PRAISE FOR CULTURE ANALYTICS

“This book is many things: a deep scientific study, an impressive

roadmap for the understanding of corporate cultures, a manual for
effective behavioural interventions and a demonstration of great
wisdom and practical experience in helping corporations to be become
better places for people.”
Professor Guido Palazzo, Professor of Business Ethics, University of
Lausanne

“This book is a detailed guide in measuring and connecting the what

with the how and the where. It is an essential advancement in applied
behavioural science and connected analytics and will change the way
we drive business success.”
William Ingham, Chief People Officer, OURO, and former SVP HR
Visa, Europe

“This book is a masterpiece and is an essential read for anyone seeking

to understand the complex interplay between cultural patterns and
data science. The author brilliantly demystifies complex analytical
techniques, offering profound insights into cultural trends and
patterns. A must-read for scholars and practitioners alike looking to
understand and leverage the power of cultural data. This book
beautifully summarizes decades of practical experience, knowledge
and invaluable contribution to behavioural and impact driven culture
studies for valuable insights including analysing cultural patterns
though the practical use of predictive analytics and its importance for
the entire organizational culture journey to drive desired outcomes.”
Ashish Sinha, Senior Partner, EMEA practice lead, People Analytics &
AI, Korn Ferry

“Even the greatest organizations simply do not survive the test of time
without a deliberate focus on their workplace culture. It is a pleasure
to read this book which articulates a powerful approach for
organizational culture measurement, using behavioural science and
predictive analytics to scientifically measure and embed desired
culture to achieve chosen business outcomes and thereby harness the
true power of culture. A must-read for leaders that want to tangibly
measure, improve and transform their workplace culture.”
Miguel Ángel Rodriguez Sola, Senior Advisor to Boston Consulting
Group

“A thought-provoking book which combines grounded research with

insightful reflection. As leaders we continue to wrestle with what
culture really means and how we measure it – this book provides new
perspectives and challenges our beliefs.”
Melanie Richards, non-executive director, UK

“It is a joy to read a book which brings hard data, deep analysis and real
insight to culture in organizations. It is a book which will make you
think and think again.”
Jo Owen, author of How to Lead and Smart Thinking

“Whilst AI will undoubtedly transform many aspects of society, culture

will continue to play a vital role in shaping how we develop, use and co-
exist with it. This book provides an amazing insight into the power of
data, how to interpret and analyse it and, more importantly, apply it in
a cultural context with practical behavioural changes. The extensive
range of models and techniques are thought provoking in themselves,
but this book isn’t just about statistics and their application, it’s about
the future of organizations. It’s a must read for anyone who is
interested in organizational and people effectiveness and wants to be
a cultural hero.”
Dave Millner, author and Founder of HR Curator

“More than just a world-class expert, the author is an amiable and

plain-speaking guide. He illustrates each theoretical framework with
relatable, how-to, applied practical examples. It’s rare and welcome to
find an analyst in any field who can bring together the interests of
scholars and practitioners so neatly. While most mere mortals get a
headache from even considering how to quantify cultural phenomena,
Hani Nabeel presents a stress-free introduction that’s full of valuable
tools and methodologies. I’ll be keeping this book close at hand as the
prime source on framing and tracking the elusive dynamics of culture.”
Roger Miles, behavioural risk researcher; Head of Faculty and Co-
Founder, Conduct Leadership Academy at UK Finance
 Culture Analytics
An evidence-based approach to company culture

Hani Nabeel
CONTENTS

List of figures and tables

Acknowledgements

PART ONE
1 Building clarity: Redefining culture and values
Leaders are the custodian of culture, but every
employee is accountable for culture
The symbiotic relationship
Notes
2 Let’s talk behaviours and demystify behavioural science!
What is behavioural science?
Behavioural science for workplace culture
Notes
3 The differences between behaviours and outcomes
Behaviours: the foundation of actions
Outcomes: the consequences of actions
Lagging and leading indicators
Path analysis
4 Understanding cultural dimensions
Behavioural measurement
Mapping behaviours to organizational
values/frameworks
Notes
5 Data analytics concepts for actionable insights
6 The link between organizational culture and its impact on
outcomes
Defining key statistical terms
Notes
7 Applying predictive analytics to culture metrics
8 You have to know the ‘why’ before you get to the ‘how’
and where’
9 Leveraging leading indicators
10 Understanding and applying path analysis as the
roadmap for driving change
Steps for developing a path analysis using SEM
modelling
11 Simplifying the actionable insights and understanding
ROI
Odds ratio
12 Cultural types
Culture type – primary behavioural clusters
Culture type – secondary behavioural clusters
Workplace culture tensions
13 Applying behavioural interventions
Stage 1: Behavioural selection and context
Stage 2: Design the intervention
Stages 3 and 4: Plan the interventions and implement
Stage 5: Validate
14 The impact of growth on organizational culture
The role of workplace culture in mergers and
acquisitions as a strategy for growth
15 Cultural navigation in the digital age
The impact of the digital age on workplace culture
The opportunity for connected data to provide
actionable workplace culture analytics
Ethical considerations for using employee data for
workplace culture analytics

PART TWO
16 Gaining predictive insight into employee sentiment
through behavioural science and analytics
Background
Context
How we crunched the numbers
Notes
17 Driving organizational change through workplace culture
and behavioural analytics
Part 1: Do your change champions have what it
really takes to effect change?
Part 2: Sizing the ‘ask’ of your organization when you
ask it to change
Part 3: Building a behavioural roadmap to drive
effective change
Notes
18 Are your HR processes supporting alignment between
culture and strategy?
Notes
19 Going beyond diversity metrics to actionable insights on
DE&I
 What’s the elephant still in the room? Maybe it’s real
commitment to DE&I programmes!
Seven behaviours of a DE&I-friendly workplace
culture
The context for the case study
Insights generated
What actions were taken?
So, did the needle move?
So what? What about the outcome?
Implications for the research approach focused on
inclusion
Notes
20 A tale of two organizations: Why innovation needs the
letter ‘C’
It’s not just about smarts and process: Why the letter
‘C’ is important
Context and methodology
The seven behaviours defining the innovation index
The traditional bank (TB)
The challenger bank (CB)
Do we need the letter ‘C’ in Innovate?
Notes
21 Safety culture and risk management
Context
The generated actionable insights
Path analysis
Notes
22 What about agile culture? Does that need the letter ‘C’?
Context
Analytics and actionable insights
 Path analysis
Note
23 The crucial role that workplace culture plays in shaping
employee wellbeing
Context
Analytics and actionable insights
Path analysis for wellbeing
Intervention design and planning
Note
24 Can we make organizational culture the hero it deserves
to be?
1. To drive success, stop focusing on results alone
and focus on the behaviours that drive these
2. Off-the-shelf models and values do not work
3. Leaders are the custodian of organizational
culture, but every employee is accountable for it
Index
LIST OF FIGURES AND TABLES
FIGURES
1. Figure 2.1 How to drive behaviour change
2. Figure 2.2 The five-step approach
3. Figure 3.1 Path analysis example
4. Figure 5.1 The behavioural factors scale interpretation
5. Figure 5.2 The dichotomous behavioural factor pairs and the scale
6. Figure 5.3 A connected data approach to provide behaviour-first or
outcome-first analytics for insights
7. Figure 5.4 The ‘where’ of behaviour presence or absence using the
quadrant dispersion method
8. Figure 6.1 A graph showing the β coefficient
9. Figure 6.2 Correlational effect size benchmarks
10. Figure 6.3 The predictive relationship between the nested five
behaviours and performance rating
11. Figure 7.1 Heatmap results for the five behaviours that drive
performance workplace culture
12. Figure 7.2 Quadrant distribution diagram showing where the five
behaviours driving performance are present
13. Figure 7.3 Line graph showing performance grade against the five
behaviours
14. Figure 8.1 Quadrant distribution graph showing behavioural
alignment vs employee engagement scores
15. Figure 9.1 Example of one set of behavioural data measurement, a
leading indicator to many thematic outcomes
16. Figure 10.1 The first two outcomes in the risk path analysis
17. Figure 10.2 The sequence of how the three outcomes should manifest
in the risk path analysis
18. Figure 10.3 The sequence of how the four outcomes should manifest
in the risk path analysis
19. Figure 10.4 The sequence of how the five outcomes should manifest
in the risk path analysis
20. Figure 10.5 The outcome sequence for the second developing path
21. Figure 10.6 The sequence of the second developing path for the
outcomes
22. Figure 10.7 The outcome sequence for the second full path
23. Figure 10.8 The outcome sequence for the third path
24. Figure 10.9 The first path with the behaviours now included
25. Figure 10.10 The second path with the behaviours now included
26. Figure 10.11 The third path with the behaviours now included
27. Figure 10.12 The behavioural leading indicators and the outcomes
they drive over the three paths
28. Figure 11.1 The odds ratio per behavioural quartile vs likelihood of
outcome occurrence
29. Figure 11.2 The odds ratio for Living our Values outcome vs
behaviour presence
30. Figure 11.3 The odds ratio for the Recognition outcome vs behaviour
presence
31. Figure 11.4 The odds ratio for the Purpose outcome vs behaviour
presence
32. Figure 11.5 The odds ratio for the Resilience outcome vs behaviour
presence
33. Figure 11.6 The odds ratio for the Psychological Safety outcome vs
behaviour presence
34. Figure 11.7 The odds ratio for the Development for the Future
outcome vs behaviour presence
35. Figure 11.8 The odds ratio for driving the Taking Ownership of Risk
outcome vs behaviour presence
36. Figure 11.9 The odds ratio for driving the Collaborate for Managing
Risk outcome vs behaviour presence
37. Figure 11.10 The odds ratio for driving the Risk Anticipation outcome
vs behaviour presence
38. Figure 11.11 The odds ratio for driving the Risk Avoidance outcome
vs behaviour presence
39. Figure 11.12 The odds ratio for driving Proactive Learning for Risk
outcome vs behaviour presence
40. Figure 11.13 The behavioural leading indicators and the outcomes
they drive as well as the likelihood of outcome manifesting over the
three paths
41. Figure 12.1 The behaviours and related culture type for a behavioural
cluster
42. Figure 12.2 The culture types with the present and targeted
behaviours marked
43. Figure 12.3 Consider First vs Act First culture types
44. Figure 12.4 Question and Listen vs Define the Answer culture types
45. Figure 12.5 Ensure Safety vs Navigate Possibility culture types
46. Figure 12.6 The Consider First vs Act First secondary culture types
47. Figure 12.7 The Question and Listen vs Define the Answer secondary
culture types
48. Figure 12.8 The Navigate Possibility vs Ensure Safety secondary
culture types
49. Figure 12.9 The percentage behavioural tension as alignment or
misalignment between Place and People
50. Figure 13.1 Intervention time plan implementation
51. Figure 13.2 Example behavioural interventions time and resource plan
52. Figure 14.1 Steps involved in the workplace cultural integration of
two independent companies
53. Figure 15.1 The big data digital approach to culture analytics
54. Figure 16.1 Path model for employee voice for this organization
55. Figure 16.2 Direct impact of the Place on employee wellbeing
56. Figure 16.3 Direct impact of the People and the Place on trust
57. Figure 16.4 Direct impact of the Place on employee voice
58. Figure 16.5 The six behaviours that drive trust, wellbeing and
employee voice
59. Figure 17.1 Behaviours metrics versus probability of change
60. Figure 17.2 The behaviours for the change metric
61. Figure 17.3 The three organizational change pillars versus
behavioural alignment
62. Figure 17.4 The profiles of the eight clusters and the alignment by
pillar
63. Figure 17.5 Combining the insight of the Place defining the size of the
ask vs the People defining the propensity for change
64. Figure 17.6 The eight-cluster mapping analysis within the size of the
ask (Place) vs propensity for change (People) metrics
65. Figure 17.7 The translation of target clusters as the starting point of
the change programme into the three pillar benchmarks
66. Figure 18.1 The behavioural defined alignment categorization zones
67. Figure 18.2 Multiple countries analysis hiring against the targeted five
behaviours
68. Figure 18.3 Employee demographic insights versus the alignment
zones and the likelihood of occurrence
69. Figure 18.4 The tenure odds ratio for five behaviours: Disenfranchise,
Delivery Focus, Collective, Consolidate and Conformity
70. Figure 19.1 Radar graph and inclusion index for the whole
organization, organizational behaviours
71. Figure 19.2 Radar graph and inclusion index for Front Office,
organizational behaviours
72. Figure 19.3 Radar graph for Front Office, personal behaviours
73. Figure 19.4 Radar graph showing the organization (Place) behavioural
differences and the inclusion index for different ethnic groups for the
whole organization
74. Figure 19.5 Regression analysis odds ratio for the first outcome,
employees being valued by the organization
75. Figure 19.6 Regression analysis odds ratio for the second outcome,
employees being valued by their team
76. Figure 19.7 Gender behavioural radar graph and inclusion index for
model one outcome, employees valued at the organization
77. Figure 19.8 Gender behavioural radar graph and inclusion index for
model two outcome, employees valued by their teams
78. Figure 19.9 Ethnicity behavioural radar graph and inclusion index for
model one outcome, employees valued at the organization
79. Figure 19.10 Ethnicity behavioural radar graph and inclusion index
for model two outcome, employees valued by their teams
80. Figure 19.11 Quadrant distribution of the three targeted behaviours:
Conformity, Team Focus and Expressive
81. Figure 19.12 Behavioural quadrant alignment matrix and descriptions
per quadrant
82. Figure 20.1 Power and hierarchy theme
83. Figure 20.2 Importance of internal risk management theme
84. Figure 20.3 External moderators to innovation theme
85. Figure 20.4 The innovation index indicator and a radar graph for the
seven behaviours that drive it for the Place
 86. Figure 20.5 The innovation index indicator and a radar graph for the
seven behaviours that drive it for the Place and the People – traditional
bank
87. Figure 20.6 Creating the best customer experience theme
88. Figure 20.7 Importance of collegial relationships theme
89. Figure 20.8 External perceptions theme
90. Figure 20.9 The innovation index indicator and a radar graph for the
seven behaviours that drive it for the Place and the People – challenger
bank
91. Figure 21.1 The odds ratio for Psychological Safety where Active
Learning, Standardized, Collective and Sequential behaviours are
present
92. Figure 21.2 The odds ratio for Inclusion at work where Active
Learning, Conformity, Relationship and People Focus behaviours are
present
93. Figure 21.3 The odds ratio for Questioning Attitude at work where
Active Learning, Conformity, Sequential and Team Focus behaviours
are present
94. Figure 21.4 The odds ratio for Safety Communication outcomes at
work where Standardized, Sequential, Relationship and Team Focus
behaviours are present
95. Figure 21.5 The odds ratio for driving a Continuous Learning culture
at work where Active Learning, Strategic, Team Focus, Sequential and
People Focus behaviours are present
96. Figure 21.6 The odds ratio for driving a culture for Accountability at
work where Strategic, Conformity, Team Focus and Sequential
behaviours are present
97. Figure 21.7 The Accountability path, also showing the behaviours’
presence needed to drive that path
98. Figure 21.8 The full outcome details for the Accountability path and
the R value illustrating the strength of the relationship between the
outcomes
99. Figure 21.9 The Proactive Safety Management path, also showing the
behaviours’ presence needed to drive that path
100. Figure 21.10 The full outcome details for the Proactive Safety
Management path and the R value illustrating the strength of the
relationship between the outcomes
101. Figure 21.11 The behaviours that need intervention design, planning
and implementation as well as the odds ratio for impacting the
outcomes’ return on investment
102. Figure 21.12 Quadrant distribution for the entire sample size for the
seven absent behaviours, Conformity, Team Focus, Collective,
Relationship, Outer Focus, People Focus and Empower
103. Figure 22.1 The odds ratio for the transparency outcome where Team
Focus, Conformity, Empower and People Focus behaviours are
present
104. Figure 22.2 The odds ratio for the growth mindset outcome where
Active Learning, Team Focus, Conformity and Strategic behaviours
are present
105. Figure 22.3 The odds ratio for the driving effectiveness outcome
where Active Learning, Conformity, Strategic and outer Focus
behaviours are present
106. Figure 22.4 The odds ratio for the psychological safety outcome
where Conformity, Expressive and People Focus behaviours are
present
107. Figure 22.5 The full agile path analysis with the leading behavioural
indicators
108. Figure 22.6 The three-stage sequential intervention plan and the odds
ratio likelihood for the return on investment
109. Figure 23.1 The odds ratio for applying the standards outcome where
Conformity and Team Focus behaviours are present
110. Figure 23.2 The odds ratio for supporting employee wellbeing
outcome where Conformity, Team Focus and Strategic behaviours are
present
111. Figure 23.3 The first established path for Psychological Safety along
with the behavioural leading indicators
112. Figure 23.4 The second established path for Wellbeing outcomes as
well as the behavioural leading indicators
113. Figure 23.5 The absent behaviours, and the proposed behavioural
intervention stages for the Psychological Safety and Wellbeing paths
as well as the return-on-investment likelihood
114. Figure 23.6 The quadrant distribution for the Expressive behaviour
functionally across the organization
115. Figure 23.7 Quadrant distribution for the Team Focus behaviour
functionally across the organization
116. Figure 23.8 Quadrant distribution for the People Focus behaviour
functionally across the organization
117. Figure 23.9 Quadrant distribution for the Conformity behaviour
functionally across the organization
118. Figure 23.10 Workshop output for the Expressive behaviour
119. Figure 23.11 Workshop output for the Team Focus behaviour
120. Figure 23.12 Workshop output for the People Focus behaviour
121. Figure 23.13 Workshop output for the Conformity behaviour
122. Figure 23.14 The interventions implementation time plan

TABLES
1. Table 4.1 CultureScope behavioural dimensions
2. Table 4.2 Full behavioural factors descriptors
3. Table 4.3 Mapping example
4. Table 4.4 Full mapping
5. Table 4.5 Cluster values behavioural mapping
6. Table 5.1 A values-based heatmap
7. Table 5.2 The presence or absence of the behaviour factors driving the
three organizational values
8. Table 5.3 A numeric heatmap for the presence or absence of the
behaviour factors driving the three organizational values
9. Table 5.4 The targeted behaviour factors and the type of intervention
needed
10. Table 6.1 Regression summary output
11. Table 6.2 Secondary regression output: Analysis of Variance
(ANOVA)
12. Table 6.3 Regression component analysis
13. Table 6.4 Regression results key components
14. Table 7.1 Comparison between the organizational values behavioural
mapping and predictive performance culture
15. Table 7.2 Behavioural mapping for Positivity
16. Table 7.3 Changed values wording and relevant behavioural mapping
17. Table 9.1 The eight behaviours mapped to all five themes
18. Table 10.1 Correlation analysis for all provided outcomes
19. Table 10.2 Correlation analysis with the mean for each outcome
calculated and ranked
20. Table 10.3 Correlation analysis with all values less than 0.2
eliminated
21. Table 10.4 Correlation analysis with values less than the column mean
removed
22. Table 10.5 The computed average mean correlations and rankings
23. Table 10.6 The path starting point outcome that has the highest
corresponding correlation coefficient to another outcome
24. Table 10.7 Risk Avoidance is our next outcome in the path analysis as
it has the highest correlation coefficient with Risk Anticipation
25. Table 10.8 The highest corresponding correlation coefficient outcome
to Risk Avoidance is Collaboration
26. Table 10.9 The highest corresponding correlation coefficient outcome
to Collaboration is Proactive Learning
27. Table 10.10 Our first path has come to a dead end
28. Table 10.11 The computed average mean correlations and rankings,
now with the first path outcomes removed
29. Table 10.12 The highest corresponding correlation coefficient
outcome to Psychological Safety is Recognition
30. Table 10.13 The highest corresponding correlation coefficient
outcome to Recognition is Resilience
31. Table 10.14 The highest corresponding correlation coefficient
outcomes have both been already used in the path
32. Table 10.15 The computed average mean correlations and rankings,
now with the first and second path outcomes removed
33. Table 11.1 The number of best and least outcome vs behavioural
quartiles
34. Table 12.1 Culture types and behaviours for the Risk Tension
35. Table 12.2 Culture types and behaviours for the Progressive Tension
36. Table 12.3 Culture types and behaviours for the Independence Tension
37. Table 13.1 Template to be completed as a team exercise to show the
current and future shift for each targeted behaviour
38. Table 13.2 Exercise showing the current and future shift for the
Empower behaviour
39. Table 13.3 Example extract for Empower behaviour analysis post-
workshop
40. Table 13.4 The COM-B function labelled with the behavioural
inhibitors for Empower
41. Table 13.5 Intervention design for an executive committee
42. Table 13.6 Intervention design for leadership team
43. Table 19.1 The two outcomes comparison before and after
interventions by gender and ethnicity demographics
44. Table 22.1 The resulting behavioural map for agile organizational
characteristics
45. Table 22.2 The sample size per county targeted for the agile
methodology
46. Table 22.3 The organizational current state vs the target agile
characteristics
47. Table 22.4 The behavioural mapping for both the current
organizational state as well as the targeted agile state
48. Table 22.5 The behavioural mapping for the current organizational
state as well as the targeted agile state
49. Table 23.1 The starting point behavioural distribution per functional
area with the organization
50. Table 23.2 All intervention designs
ACKNOWLEDGEMENTS
Writing this book was a wonderful trip down memory lane, remembering
all the special people and moments that inspired my research journey in the
quest of making organizational culture the hero it deserves to be. To all of
you I say a heartfelt thank you.

A special thank you to Bharat Shah, who has dedicated himself completely
to what became our joint mission in escalating workplace culture
measurement and design, as the key focus for all organizational leaders and
board members.

I also fondly remember and thank Eugene Burke. His deep knowledge in
psychometrics, behavioural science and people analytics was not just
inspirational, but transformative in how I bring culture insights to life.

As I also reflect back on the Neural Computing design for predictive

analytics, I thank my son Joe Nabeel and Charlie Levi both for their
amazing scientific data modelling and design, and for the awesome work
by the talented software engineers at Future Processing in bringing all the
design work to life.

My sincere appreciation goes to my colleague Michelle Renecle, a

whirlwind combination of behavioural research excellence and actionable
insights, who has and will always be a constant source of inspiration.

To family – without their constant support (and sometimes tolerance!), I

would not have been able to dedicate myself to this quest.
PART ONE
1 Building clarity: Redefining culture and
values
Culture is like the wind: blowing in your direction, it makes for
smooth sailing; blowing against you, everything is more difficult.

We hear so much about organizational culture, organizational values and

‘what we stand for’, countless articles, books, documentaries, white papers,
research studies, etc – all for the sake of addressing this crucial topic.

It is universally recognized that corporate culture is one of the most critical

levers for driving organizational success and creating stakeholder value.
When organizations successfully align strategy with culture and talent, they
elevate performance and minimize risk. Yet, culture is the one element that
most organizations fail to empirically measure and track, and therefore they
underutilize this important lever. Further, organizational culture is often the
‘villain’ when it goes all wrong, but there are many great stories out there
where culture has been the ‘hero’ for an organization.

Many organizations of all shapes and sizes battle with answering many
significant questions as part of their journey. As an example of a non-
exhaustive list:

What is the current culture of the organization and how aligned is it

with our strategy?

What are the behavioural changes required to drive culture change to

achieve our strategy?

How do we consider culture in our talent management strategy to

attract, develop and retain talent?

What is the gap between our current and desired culture?

How well do our organizational structure and practices support our

ideal culture and minimize risk?
 Will our current culture support our growth plans?

For organizations that are on a significant organic growth path, and have a
great current culture, the oddity is that such growth could be the enemy of
culture. In fact, culture is key in driving ‘how’ growth is sustained. As you
grow and add people or even suppliers to your team, your culture is at risk.
After all, the more people involved, the greater the opportunity for dilution.

So, let’s start with some definitions. Perhaps Richard Perrin’s concept of
organizational culture as ‘the sum of values and rituals which serve as glue
to integrate the members of the organization’, which suggests that
organizational culture is present in structures, routines and actions. Or
Edgar Schein’s model which suggests that organizational culture has three
components:

1. Overt elements that are apparent to outsiders (colours, fonts, furniture,

dress code, etc).

2. A distinguishable set of declared values and norms that affect how

members interact.

3. Basic assumptions which are beliefs and behaviours so deeply

embedded that they can sometimes go unnoticed.1

Nevertheless, basic assumptions are the bedrock of organizational culture

and problems can arise when they do not align with the most evident
elements.

Organizational culture refers to the shared beliefs, values, attitudes, norms

and behaviours that exist within an organization. It represents the way
things are done in the organization and shapes the overall working
environment. Organizational culture outcomes can be observed through
various elements such as dress code, communication patterns, leadership
style, decision-making processes, risk management, performance and
employee engagement.

On the other hand, organizational values are the guiding principles or core
beliefs that drive the behaviour and actions of individuals within the
organization. These values reflect what the organization stands for and
what it considers important. Organizational values can influence decision-
making, goal-setting, performance evaluations and overall workplace
behaviour.

While culture describes how things are done, values represent why they are
done. Culture is more about the collective behaviour and mindset of the
organization, whereas values are the underlying principles that guide that
behaviour. In fact, if the organizational values do not manifest in
behaviours, it renders them empty words that take up space on a website or
wall, as one key piece of evidence of ‘living the values’ is behaviours.

Understanding organizational culture and values is crucial for several

reasons:

Employee alignment: Knowing and understanding the culture and

values of an organization helps employees align themselves with the
overall goals and mission. It gives them a sense of purpose and
direction, and a feeling that they are a part of something bigger.

Decision-making: By understanding the culture and values,

employees can make decisions that are in line with the organization’s
goals and principles.

Employee engagement: Driving motivation and commitment to

perform high-quality work and contribute to organizational success.
Engaged employees exemplify eagerness to perform well, accomplish
their tasks and positively impact their organization.

Recruitment and retention: Individuals who align with the culture

and values are more likely to be successful and satisfied in their roles.
It also helps in retaining employees as they are more likely to stay in
an organization that reflects their personal values.

Organizational performance: A strong organizational culture and

values can positively impact the overall performance of an
organization. When employees are aligned with the culture and values,
 they work collaboratively, communicate effectively and strive towards
achieving organizational goals.

Effective risk management: While rules and regulations are

important, culture is your best and most effective ally for risk and
reputational management.

It is common knowledge that nurturing a positive organizational culture is

essential for creating a healthy and successful work environment. It helps to
shape employee behaviour, hence to improve performance, manage risk
and drive the overall success of the organization. In fact, if you are in a
leadership role, one of the most critical responsibilities you have is to
create the intended organizational culture. That means that you are
responsible for creating the ‘workplace’ environment for the intended
culture to fully manifest, including role-modelling – that is, own behaviour,
peer behaviour, policies, processes, procedures, performance management,
governance, organizational design and even physical office design. All
these factors work on behaviours and can be great allies in driving desired
behaviours and organizational culture.

This is critical to remember: later I will revisit this aspect by referring to it

as the ‘Place’ and the ‘People’, and I will re-emphasize the definitions later
as well. This is probably why we often hear the famous Peter Drucker
quote ‘Culture eats strategy for breakfast’.2 I would go further by saying:
let culture be your best ally in achieving your strategy!

Another interesting quote by Peter Drucker is, ‘If you can’t measure it, you
can’t change it’. Again, I am going to go further with this by adding: if you
can’t measure it, then you can’t change it, fix it, sustain it, evolve it or even
repeat it! These few words ignited my seven-year extensive and dedicated
research into behaviours and workplace culture… but more about that later.

Leaders are the custodian of culture, but every

employee is accountable for culture
In any organization, leaders undeniably hold a pivotal role. They’re often
seen as the architects, crafting vision and shaping strategic direction. In the
sphere of organizational culture, they emerge as its custodians. They set the
tone and guiding principles, and delineate the values that become the
pillars. Yet, while they initiate and nurture this culture, its true essence
thrives on collective ownership.

Leaders: steering the ship

As custodians, leaders are responsible for setting clear expectations and

exemplifying the behaviours they wish to see. Their actions, decisions and
communications shape perceptions and establish norms. This idea of
establishing the ‘tone from the top’ or role-modelling the desired culture
and behaviours is well documented and understood by most leaders, but it
is often over-relied on as the only method through which cultural norms are
established.

Although leaders are the torchbearers, illuminating the path for the rest to
follow, a torchbearer alone doesn’t light up the entire landscape. That’s
where every employee comes into play.

Employees: the keepers of the flame

While leaders set the stage, it’s the collective everyday actions of
employees that bring culture to life. Think of culture as a living organism,
pulsating and evolving with every interaction, decision and process. Each
employee, in their daily duties and interactions, either reinforces or
challenges the established cultural norms. Thus, while leaders are the
custodians, every individual is accountable for nurturing and sustaining the
cultural fabric.

For instance, consider a company that values open communication and

expressiveness. While the leadership can champion this through town halls
and transparent discussions, it truly comes to life when an entry-level
employee feels empowered to voice their opinion in a team meeting or
when colleagues actively listen and respond to each other’s perspectives.
The symbiotic relationship
There exists a symbiotic relationship between leadership and the broader
employee base when it comes to culture. While leaders lay the foundations
and uphold the standards, continuous reinforcement from every corner of
the organization ensures that the desired, targeting culture is embedded.
The greater the size and complexity of the organization, the more difficult it
is for leader role-modelling and the ‘tone from the top’ to consistently help
illuminate culture, and the more critical employees’ roles are in nurturing
and driving that culture.

Therefore, this duality means that:

Leaders must ensure that their actions mirror their words, creating an
environment of trust and setting the standard.

Every employee should recognize their influence. Whether in team

meetings, client interactions or casual coffee breaks, their behaviours
contribute to the organizational culture’s mosaic.

To truly foster a thriving organizational culture, it’s essential to embrace

this dual responsibility. Leaders must cultivate, employees must nurture,
and together they sculpt a culture that resonates, evolves and drives
success.

Let me demonstrate by sharing with you a quick story. A CEO of a

promising new start-up excitedly explained his company’s core values
which he had come to through a combination of following what he’d read
in the latest bestselling business book and copying values he knew worked
for his most successful competitors. Six months later, he had rampant
turnover and couldn’t keep his best people. Why? Because the values,
while impressive on paper, didn’t work for his unique strategy, context and
employees.

In another company, a medium-sized financial services organization, the

head of HR shared the results of their recent engagement survey – vibrant
charts, impressive infographics and enthusiastic comments. And yet she
admitted with a sigh, ‘It tells us that people are disengaged and have issues
with psychological safety, but it doesn’t answer the most pertinent question:
why? It’s as if we’ve taken a selfie. Lovely to look at, momentarily
insightful, but we don’t really know what’s going on in the background.’
The survey had captured sentiments, not the actions driving them.

Then there was the case of a fast-growing FMCG organization. Its

leadership team was often caught in a dilemma – despite booming sales,
internal conflicts and low morale were rampant. They were in the dark
about how the company’s cultural nuances were influencing these
outcomes, for better or worse.

Now think further about the so many other stories where culture has
brought large organizations to their knees: Boeing, where the ‘culture of
concealment’ and cost-cutting led to two fatal crashes of 737 Max aircraft
claiming 346 lives; the Volkswagen emissions scandal – also known as
‘emissionsgate’ and ‘dieselgate’ – which started in September 2015, when
the US Environmental Protection Agency (EPA) announced that it believed
VW had cheated the emissions tests. Also Wells Fargo’s toxic sales culture
which created minimal accountability for the questionable practice of
taking unfair advantage of customers to meet challenging sales targets,
Enron, BP, Facebook, Lehman Brothers… I’d better stop now.

So, what’s the common thread here? The cardinal misconceptions in

understanding and managing organizational culture: adopting values that
sound right or mimicking successful peers, assuming engagement surveys
or qualitative deep dives are accurate measures of your organizational
culture, and acting to enhance organizational outcomes you care about
(sales, wellbeing, engagement, risk management) in the absence of
understanding how your culture drives these outcomes.

No leader wakes up in the morning thinking, ‘I can’t wait to get to work

and do everything I can to drive the most toxic, ineffective culture possible
for my company’. Yet, due to these three cardinal errors, many leaders get
caught out by the wrong behaviours becoming dominant in their
organization, and some of them end up unintentionally harming employees,
customers or communities, which results in shutdowns, lawsuits and media
scandals.
While well-intentioned, leaders often wade through these murky waters of
culture ill-equipped as they have been poorly served regarding how to
understand and effectively manage organizational culture.

In this book, I will share my learnings with you and offer practical insights
and recommendations on how to successfully identify, measure and embed
the right culture to enable your organization to achieve great things and
stay out of trouble.

Before we move on, another thought-provoking idea is that growth can be

the enemy of organizational culture, as demonstrated in the above
examples. Why? Well, it’s about not leaving workplace culture to chance! I
will cover this idea in detail in Chapter 14.

Notes
1. 1 E Schein (2016) Organizational Culture and Leadership (5th edn),
Jossey-Bass

2. 2 S Hyken. Drucker said ‘Culture eats strategy for breakfast’ and

Enterprise Rent-A-Car proves it, Forbes, 5 December 2015,
www.forbes.com/sites/shephyken/2015/12/05/drucker-said-culture-
eats-strategy-for-breakfast-and-enterprise-rent-a-car-proves-it/
(archived at https://perma.cc/VDM5-NDMW)
2 Let’s talk behaviours and demystify
behavioural science!
Understanding human behaviour: The art and science of behavioural
science

Human behaviour is a complex amalgamation of genetics, culture (country,

organizational or both), environment and individual experiences. It shapes
our interaction with the world, influences our decision-making and
ultimately defines who we are as individuals and as a society. The field of
behavioural science offers a systematic approach to understanding human
behaviour, shedding light on the underlying mechanisms and providing
valuable insights into the way we think, feel and act.

The above is a myth-bust – that behaviour is the pure result of personality

types.

What is behavioural science?

Behavioural science is an interdisciplinary field that combines knowledge
and insights from various disciplines such as psychology, sociology,
anthropology, economics and neuroscience to study human behaviour. It
explores the factors that influence our actions, motivations, beliefs,
attitudes and emotions, often employing scientific methodologies to gain a
deeper understanding of the complexities within the human mind.

Quantitative behavioural science builds on traditional behavioural science

by adding a research and statistical lens to the study of human behaviour.
This involves the collection, description, analysis and inference of
conclusions about human behaviour from quantitative data. The
mathematical theories behind statistics rely heavily on differential and
integral calculus, linear algebra and probability theory. This theory will
later come to life to explain powerful actionable insights for driving
workplace culture.
Key concepts and theories

The behavioural science research of the renowned psychologist Daniel

Kahneman helped us articulate that human behaviour is driven by two
independent systems: System 1 and System 2:1

System 1: Our brains’ fast, automatic, unconscious and emotional

responses to situations and stimuli. It handles our more familiar tasks.

System 2: The slow, effortful and logical mode in which our brains
operate when solving more complicated problems.

On average we all have about 35,000 decisions to make each day.2 These
differ in difficulty and importance. It could be you taking a step to your left
or right when walking, or deciding whether to take the stairs or the elevator.
If you had to consciously process all these decisions, your brain would
crash. Your automatic System 1’s primary task is to protect your deliberate
System 2 thinking. It helps you prevent cognitive overload – but System 1
is susceptible to bias.

There are a few ways in which our automatic system lightens the load for
our deliberate system. First of all, it takes care of our familiar tasks by
turning them into autopilot routines, known as habits. But what your
System 1 is primarily doing is rapidly sifting through information and
ideas, without you even noticing it – prioritizing whatever seems relevant,
and filtering out the rest by taking shortcuts.

So, can we change daily behaviours to become embedded in System 1? If

so, how? The answer is yes, we can. The theories behind that process are as
follows:

Classical conditioning: One of the fundamental theories in

behavioural science is classical conditioning, proposed by Ivan
Pavlov, a Russian and Soviet experimental neurologist and
physiologist. It explores how individuals learn to associate certain
stimuli with specific responses, shaping their behaviour.
 Operant conditioning: Developed by B F Skinner, an American
psychologist, behaviourist, inventor and social philosopher. Operant
conditioning focuses on how reinforcement and punishment can affect
behaviour. Positive reinforcement encourages desirable behaviours by
offering rewards, while punishment discourages negative behaviours
through unpleasant consequences. This theory highlights the influence
of consequences on human actions.

Social learning theory: Proposed by Albert Bandura, a Canadian-

American psychologist who was responsible for contributions to the
fields of social cognitive theory, therapy and personality psychology,
and the transition between behaviourism and cognitive psychology.
Social learning theory suggests that we learn by observing and
imitating the behaviour of others. This theory emphasizes the impact
of role models, peer influence and observational learning in shaping
our behaviour.

Cognitive behavioural therapy (CBT): A therapeutic approach

derived from behavioural science, CBT focuses on understanding the
relationship between thoughts, feelings and behaviours. It helps
individuals identify negative thought patterns and replace them with
healthier and more constructive ones, ultimately leading to positive
behaviour changes.

Biases vs nudges

Before we move on, let’s address the difference between biases and nudges.

A bias is a mental shortcut used to solve a particular problem; it is a quick,

automatic and intuitive algorithm your brain uses to generate an
approximate answer to make a judgement. Biases describe at a high level
how the majority of people will behave under specific circumstances.

If you want to drive a behaviour from System 2 to System 1, then you need
to explain and train it. For example, to drive a bias towards recycling
(System 1), you need to first address System 2 by considering why we need
to recycle. The benefits of this become the ‘normal’ way of thinking.
A nudge is a corresponding set of stimuli for each bias that becomes the
tactics to actively harness the bias and change behaviour. For example,
providing recycling bins in many locations, posters reminding us to recycle,
and so on.

Practical applications

Behavioural science has numerous practical applications across various

domains, making it a valuable tool for understanding and influencing
human behaviour. Here are a few examples:

Behavioural economics: Behavioural science plays a vital role in

understanding economic decision-making. By examining the biases,
heuristics and cognitive processes that affect how individuals make
choices, economists can design more effective policies and
interventions to promote healthy financial behaviours.

Public health: Behavioural science has proven instrumental in

designing interventions to promote healthy behaviours, such as
smoking cessation programmes, dietary interventions and encouraging
physical activity. By understanding the factors that drive behaviour,
public health practitioners can develop targeted interventions to
improve overall health outcomes.

Organizational behaviour: By applying behavioural science

principles, organizations can create work environments that foster
employee engagement, productivity and satisfaction. Understanding
the motivations, biases and dynamics of individuals and groups within
an organization can inform leadership strategies, team dynamics and
organizational policies.

Needless to say, this last practical use case will be the main focus of this
book, which will include many cases across many organizational culture
themes.

Behavioural science provides us with a powerful lens to understand, predict

and influence human behaviour. By uncovering the complex interplay of
biological, psychological, social and environmental factors, behavioural
science helps us gain new perspectives on the intricacies of being human.
From creating effective public policies to improving individual wellbeing,
the insights gleaned from behavioural science enable us to design strategies
that align with our innate human tendencies and create positive change.

Behavioural science for workplace culture

Now for the critical bit for workplace culture: how do we drive effective
change? I would like to introduce you to the COM-B model of behaviour
change.

The COM-B model is widely used to identify what needs to change for a
behaviour change intervention to be effective. It identifies three factors that
need to be present for any behaviour to occur: capability, opportunity and
motivation. These factors interact over time so that behaviour can be seen
as part of a dynamic system with positive and negative feedback loops.

Capability is an attribute of a person; it is all about the ‘People’. Are they

capable of behaving in a certain way? Do they have the knowledge, skills
and abilities required to engage in a particular behaviour? Its two
components are:

psychological capability: our knowledge/psychological strength,

skills, decision process, memory, stamina or behavioural regulation

physical capability: our physical strength, skills, stamina or

proficiencies acquired through practice

Opportunity is an attribute of an environmental system; it is all about the

‘Place’. Does our workplace provide the opportunity for a behaviour to
happen? Its two components are:

physical opportunity: opportunities provided by the environment, such

as time, location and resources

social opportunity: opportunities as a result of social factors, such as

cultural norms and social cues and pressures, social comparisons and
 leadership

Motivation is also an attribute of an environmental system; it is a mental

process that energizes and directs a behaviour. Its two components are:

automatic motivation (System 1): automatic processes, such as our

desires, impulses and inhibitions, incentives, rewards and punishment

reflective motivation (System 2): reflective processes, such as making

plans and evaluating things that have already happened, beliefs about
consequences, goals, roles, intentions and identity

So, capability is about the People, whereas opportunity and motivation are
about the Place. For a behaviour to be possible or fully manifest, it is very
important to have alignment between People and Place. Of course, this is
true whether you want to drive or inhibit a behaviour.

You may remember in Chapter 1 we discussed how organizational values,

beliefs and even mindset will also affect culture – so how can we connect
that to the behaviour systems and COM-B? This is where good design is
essential and connecting the dots becomes important. Organizational values
are the first layer that should drive ‘what’ (capability), ‘how’ (opportunity)
and ‘why’ (motivation). During organizational values design, you can start
with the type of culture you are targeting and the behaviours you are
targeting to enable the intended culture to fully manifest. You can now
design the values statement.

This design method is reliant on the ability to typify the intended culture,
something that I will visit in the next chapter. Figure 2.1 brings the entire
system together.

FIGURE 2.1 How to drive behaviour change

A flow chart shows the process of behaviour change.

Figure 2.1 details

The imperative for alignment between People and Place to drive culture
change is also a critical aspect for measurement.

How do we go about transforming organizational culture using these

behavioural science principles? What does an entire project for culture
change look like? I will now introduce a simple five-step approach that is
an end-to-end process, as follows:

1. Define: what do you want to achieve, and what is your purpose for
doing so? Connecting culture to strategy is an obvious first step for
most, but it is rarely done with the rigour and precision that is
required. Start by reviewing your organizational strategy and mission,
highlighting key objectives and goals. Ask questions like ‘What
tangible results are we aiming for in the next year?’, ‘What employee
or customer experiences do we want to drive?’ and ‘What do we want
to control and minimize around risk, disruption and change?’ From
this, outline and populate the specific outcomes you wish to achieve,
such as increased sales, improved psychological safety, enhanced
customer satisfaction or high compliance with safety processes.

2. Measure: this measurement is split into three:

behaviour measurement of the People

behaviour measurement of the Place

the intended or targeted outcomes measurement

3. Actionable insights: which behaviours you must focus on to achieve

your outcomes. To be actionable, insights must explain the ‘why’ for
the outcomes, ‘why’ the targeted outcomes happen (or not), ‘how’ to
improve and the return on your investment. Additionally, insights
must deliver ‘where’ in the organization you can learn from and
‘where’ to drive interventions.

4. Culture by Design: This ties all the previous steps together.

Ultimately, it’s now about using the COM-B model to achieve culture
change.
 5. Validate: This is your feedback loop, Re-measure the impact of the
inventions taken to drive behavioural change. The three measurements
must be repeated – behaviours of the People, the Place and the
outcomes. This should happen within six to nine months of the first
intervention deployment.

Figure 2.2 gives a quick summary of the five-step approach.

FIGURE 2.2 The five-step approach

An illustration shows a summary of the five-step approach. The

shaded circle with five points are as follows: 1. Define; 2.
Measure; 3. Actionable insight; 4. Culture by design; and 5:
validate.

You may be wondering why I called the behaviour intervention, planning

and implementation ‘Culture by Design’. Well, you tend to design all other
parts of your organization, like your financial systems, your operation
models and IT infrastructure, so why would you not work on designing the
most critical aspect of your organization, your culture? Hence the name.

This approach will be used for all the thematic case studies presented later
in this book.

Notes
1. 1 D Kahneman (2011) Thinking, Fast and Slow, Allen Lane

2. 2 A Reill. A simple way to make better decisions, Harvard Business

Review, 5 December 2023, https://hbr.org/2023/12/a-simple-way-to-
make-better-decisions (archived at https://perma.cc/U4DA-N2MF)
3
The differences between
behaviours and outcomes
Understanding the distinction between behaviours and outcomes is a key
component for managing and developing workplace culture, and one that is
so often misused or even misunderstood.
In our daily lives, we often become preoccupied with the outcome of our
actions, striving for favourable results. However, it’s important to recognize
that behaviours and outcomes are distinct concepts. While outcomes are the
end results we desire, behaviours are the actions we take to achieve those
outcomes. Understanding the differences between these two pillars can help
you develop a more comprehensive approach to managing and developing
workplace culture.

Behaviours: the foundation of actions

Behaviours refer to the actions and choices we make in pursuing a
particular outcome. These encompass a wide range of tangible actions, such
as expressiveness, following the rules, being fair, etc. Our behaviours can
be a reflection of our values, beliefs and mental models, shaping our
interactions and determining how we navigate the challenges and
opportunities we encounter. However, we also know that environmental
factors (which I have referred to as ‘Place’ in prior chapters, defined as
leadership behaviours and role modelling, peer behaviour, policies,
processes, procedures, governance and physical office design) and
situations will also impact how we behave.
Key characteristics of behaviours:
 Within our sphere of control: Unlike outcomes, which may vary
due to external factors beyond our control, behaviours are fully
within our realm of influence. We have the power to shape and
improve our behaviours through intentional effort and self-reflection.
Observable: Behaviours are observable by others or even by
ourselves, e.g. whether we act ethically, communicate openly, adapt
to change or take the initiative.
Impact on outcomes: Behaviours have a direct influence on the
outcomes we strive for. Consistency in behaviours will increase the
likelihood of achieving our desired results. This will be regularly
demonstrated in the many thematic case studies later.

Outcomes: the consequences of actions

Outcomes are the results that follow our behaviours. They can be either
positive or negative, depending on various factors, including external
circumstances, luck, timing and the actions of others. Outcomes are often
measured in terms of accomplishments, incidents, financial gains,
recognition or formed opinions. While outcomes are important indicators of
success, relying solely on them to evaluate our efforts can lead to an
incomplete understanding of actions and hinder potential.
Key characteristics of outcomes:
Influenced by many factors: Outcomes are often influenced by
many factors, including behaviours, events and other outcomes.
Subjective interpretation: Outcomes can be perceived differently by
different individuals. What one person may consider a success,
another might view as a failure. It’s essential to avoid solely defining
oneself by outcomes and consider the broader context of behaviours
when evaluating success or progress.
 Learning opportunities: Regardless of the outcome, every action
provides an opportunity for learning. Outcomes, whether favourable
or not, must lead us to evaluate behaviours, refine our approach and
pursue corrected behaviours to achieve the desired results.
Let’s now illustrate the difference between behaviours and outcomes using
simple everyday examples. You get in the car with your colleague, and you
observe that they are putting on their seatbelt and asking you to do the
same. They check their seat to make sure that they are in the correct
position and have full view of the rearview and side mirrors, and they
indicate and check all mirrors before they carefully drive out.
You have just observed outcomes. The behaviours that facilitate these
outcomes are: 1) being consistent in following processes, and 2) obeying
the rules.
Now think further; such behaviours will also mean that your colleague
will take care crossing the road. This illustrates the power of behaviours as
they affect multiple outcomes. Now to stretch your thinking even further;
the outcomes above will also impact other outcomes like the reduced
likelihood of an accident. Now you see the power of a few behaviours
impacting a multilateral chain of outcomes. This insight will be
significantly demonstrated in the case studies in Part 2 of this book.

Embracing the balance

While the driving force behind our actions is the desired outcomes,
understanding the distinction between behaviours and outcomes allows us
to adopt a more holistic approach. Focusing solely on outcomes may lead to
shortcuts, unethical behaviours or neglecting the importance of the journey
itself. In contrast, placing excessive emphasis on behaviours without
considering outcomes may result in lack of accountability or reduced
motivation.
 Striking a balance by aligning our behaviours with desired outcomes
enables you to navigate challenges, adapt to evolving circumstances and
condition your workplace culture.

Lagging and leading indicators

Leading and lagging indicators are two types of measurements used when
assessing performance in a business or organization. A leading indicator is a
predictive measurement; for example, the percentage of people wearing
hard hats on a building site is a leading safety indicator. A lagging indicator
is an output measurement; for example, the number of accidents on a
building site is a lagging safety indicator. The difference between the two is
that a leading indicator can influence change, and a lagging indicator can
only record what has happened.
All too often we concentrate on measuring results, outputs and outcomes.
Why? Because they are easy to measure and they are accurate. If we want
to know how many sales have been made this month, we simply count
them. If we want to know how many accidents have occurred on the factory
floor, we consult the accident log. These are lagging indicators. They are an
after-the-event measurement, essential for charting progress but useless
when attempting to influence the future.
A different type of measurement is required to influence the future, one
that is predictive rather than a result. For example, if we want to increase
sales, a predictive measure could be to make more sales calls or run more
marketing campaigns. If we wanted to decrease accidents on the factory
floor, we could make safety training mandatory for all employees.
Measuring these activities provides us with a set of leading indicators. They
are in-process measures and are predictive.
Leading indicators are always more challenging to determine than
lagging indicators. They are predictive and, therefore, do not guarantee
success. This not only makes it difficult to decide which leading indicators
to use, but also tends to cause heated debate as to the validity of the
measure at all. To fuel the debate further, leading indicators frequently
require investment to implement an initiative before a lagging indicator sees
a result.
When developing a business performance management strategy, using a
combination of leading and lagging indicators is always good practice. The
reason for this is obvious; a lagging indicator without a leading indicator
will not reveal how a result will be achieved and provide no early warnings
about tracking towards a strategic goal.
Equally important, however, a leading indicator without a lagging
indicator may make you feel good about keeping busy with many activities.
Still, it will not confirm that a business result has been achieved.
There is a cause-and-effect chain between leading and lagging indicators;
both are important when selecting measures to track towards your business
goals. Traditionally, we tend to settle for lagging indicators; however, we do
not underestimate the importance of leading indicators.
A health warning here: do not assume that the behaviours that predict
outcomes in one organization will be the same for another organization.
Remember that earlier we discussed that the organizational environment
would be different for different organizations, like the processes,
procedures, operating model, etc… hence the danger of borrowing off-the-
shelf frameworks and assuming that such frameworks will simply work for
you. If you delve into any leading management book, white paper or
keynote speech, you’ll likely stumble upon models or values presented as
the universal remedy for organizational challenges. They draw you in with
promises such as ‘Follow these steps and you’ll see transformation’.
The realm of organizational behaviour, leadership and culture often
suggests the existence of universal truths. These are solutions that, when
applied across the board, promise to lead to outcomes every leader dreams
of – soaring sales, unwavering resilience, a loyal consumer base and so
forth. As our research commenced, we held a glimmer of hope. A hope that
within the myriad data points and diverse organizations spread across 61
countries, we’d find these universal keys to success or uncover a template
for organizational thriving.
Yet the reality that emerged was sobering: every organization’s
behavioural blueprint to drive success was different. Well, that does make
sense, as we know that organizational culture is unique and it’s a key asset
for any organization. You can replicate products and services, but it’s very
hard to replicate workplace culture.
By understanding and consciously aligning behaviours with desired
outcomes, you can enhance your chances of success. Ultimately, it is the
combination of the behaviours and the lessons you derive from outcomes
that paves the way for continual improvement and a fulfilling journey
towards your culture goals.

Path analysis
Path analysis uses is a powerful statistical technique for investigating
complex causal relationships among variables. It provides a comprehensive
framework to test hypotheses, explore direct and indirect effects and refine
models. So, how will this help us in understanding relationships between
behaviours and multiple outcomes?
Let me demonstrate using one of our above examples: wearing hard hats
on a building site. The behaviours that lead to this outcome ‘always wearing
hard hats’ are being consistent and closely following rules and procedures.
That outcome of ‘hard-hat wearing’ itself becomes a leading safety
indicator, and the outcome of ‘safety indicator’ becomes the leading
indicator of low number of injuries, etc. With this simple example we have
established the starting point behaviours, their impact on multiple outcomes
and the sequence in which those outcomes would happen.
 You can go further and add events to the developed path to test its
robustness. For example, what are the events that can happen that can be
better managed with the hard hats on or off? Is there any likely event that
has a better outcome if the hard hats are off?
Figure 3.1 shows the path.
 FIGURE 3.1 Path analysis example
A flow chart shows a sample process of path analysis.

Figure 3.1 details

If you are interested in the maths for quantitative behavioural science, I can
further explain that path analysis is established using a technique called
structural equation modelling (SEM). It is a statistical analytical
approach that allows researchers to analyse complex causal relationships
among multiple variables. Path analysis using SEM provides a rigorous
framework for testing hypotheses and exploring the interplay between
variables in a larger model.
Structural equation modelling is a multivariate statistical technique that
combines factor analysis and multiple regression analysis to examine the
relationships between measured variables and behaviours. It enables
researchers to investigate both direct and indirect relationships in a
comprehensive way. However, path analysis goes further, as it helps you
interpret the relationships between variables, providing valuable insights
into causal pathways and mediating effects.
The advantages of path analysis using SEM are numerous. First, it allows
researchers to examine complex relationships with multiple variables
simultaneously, providing a more comprehensive understanding of the
underlying mechanisms. Second, it helps to identify direct and indirect
effects, thereby allowing for a better understanding of the contribution of
each variable to the overall outcome. Third, it enables researchers to test
competing hypotheses and compare alternative models, facilitating
development and refinement.
Furthermore, path analysis using SEM can also account for measurement
error, which enhances the accuracy of the estimates and reduces bias. It is
particularly useful when dealing with latent constructs that are not directly
observed and may be measured incorrectly. By accounting for this, path
analysis using SEM provides more reliable and valid estimates of the
relationships between variables.
Many of the examples offered in Part 2 of this book will use this
approach to help you establish its practical yet powerful applications.
4 Understanding cultural dimensions
We have already covered the importance of measurement for organizational
culture insights and how measuring leading indicators can be more
complex, in particular, when the leading indicators are behaviours that are a
critical and actionable component of managing and evolving your
workplace culture. So, the two important questions are: which behaviours,
and how do we measure these behaviours?

Those were exactly the questions that ignited my seven-year research. It

was to become the most extensive quantitative behavioural research study
on organizational behaviour and culture to date, in which I studied over
51,000 employees across 60 diverse organizations in 61 countries. This was
not a venture into abstract theories or transient sentiments, but an in-depth
exploration of over 350 million data points to uncover tangible behaviours
that form the bedrock of organizational cultures.

Before we discuss the outcomes of that research and by way of

background, my earlier research started in airliner cockpits, having
observed many pilots and crew members to better understand human
factors and how they manifest in behaviours. The flight deck is a very
‘clinical’ place to conduct such research, as skill and knowledge are
somewhat a given, so you are left with the behavioural aspects.

Whenever you board a flight, it’s safe to assume that whichever airline you
have chosen, all pilots know their ‘stick and rudder stuff’; the skill of flying
and what all the buttons and gauges do. What you don’t know is how are
they going to behaviourally conduct the flight. Suddenly behaviours
become a significant leading indicator of outcomes, and of how the crew
respond to critical or somewhat benign situations.

Back to the seven-year organizational study. The below is a quick brief on

the methodology.

Research dimensions: n = 51,236 – 60 diverse organizations – 61 countries

– 229 behaviours – 350 million data points.
There were six key stages of research, as follows:

1. Principal Component Analysis (PCA) for valid item selection, scale

construction and factorial definitions. The purpose of this study was to
identify a robust set of dimensions to define a working structure for
the CultureScope diagnostic tool. After eliminating items/dimensions
with insufficient variances, initial scales were refined through an
iterative sequence of internal consistency and principal component
analysis.

2. Factorial analysis with orthogonal rotation (Varimax rotation) to

examine variability for final dimensions selection, to investigate the
interrelations between a set of behaviour variables and explain them in
terms of their latent common dimensions, called factors. It is a
technique of information reduction that does not consider the variables
as dependent or independent, since all are considered simultaneously.

3. Test-retest for diagnostic reliability to measure reliability (or

internal consistency).

4. Multi-Traits Multi-Methods (MTMM) to examine construct

validity.

5. Multivariate Analysis of Variance (MANOVA) to examine inter-

organizational variability and consensual validity.

6. Predictive analytics validity, to examine how well the leading

indicators impact outcomes as the lagging indicators using multi-level
modelling, regression analysis, multiplicity and odds ratio. By way of
a definition: multiple regression is a statistical technique that can be
used to analyse the relationship between a single dependent variable
and several independent variables. The objective of multiple
regression analysis is to use the independent variables whose values
are known to predict the value of the dependent value.

From this pioneering study, I discovered that hundreds of organizational

behaviours could be distilled into 30 distinct behaviours (out of the 229
examined behaviour factors) that make up the blueprint of how people and
organizations behave. These 30 behaviours drive a myriad of business
outcomes – from outcomes that directly impact the bottom line (sales,
performance, productivity) to outcomes that keep companies out of trouble
(risk management, financial crime, safety compliance) and even outcomes
that drive engagement and morale (wellbeing, trust, belonging,
psychological safety).

Further, the research found that these 30 behavioural factors are arranged in
15 continuums, hence two dichotomous factors for each dimension,
resulting in 15 dimensions. These 30 factors/15 dimensions are shown in
Table 4.1.

TABLE 4.1 CultureScope behavioural dimensions

Skip table

Collective: Emphasis is on Individual: Emphasis is on

VS
working in partnership individual success

Disenfranchise: Decisions Empower: Decisions

VS
escalated upwards delegated downwards

People Focus: Focus on people Delivery Focus: Outputs are

VS
rather than outputs the key focus

Neutral: Less likely to voice a Expressive: More likely to

VS
thought or opinion voice an opinion

Standardized: Focus on Flexible: Ability to work

VS
process and systems flexibly valued

Moderate: Consistency and VS Radical: New ideas and

stability valued change valued

Team Focus: Reward Self-Focus: Reward

VS
recognizes team working emphasizes individualism

Tactical: Focus is on short- Strategic: Focus is on the

VS
term results longer-term

Active Learning: Structured Passive Learning: Learning

VS
development valued is laissez-faire

Innovate: Managed risk-taking Consolidate: Control and

VS
and innovation valued risk avoidance valued

Conformity: Acting in line Nonconformity: Act in

VS
with rules and values response to the situation

Transact: Getting it done more Relationship: Getting along

VS
important more important

Achievement: People valued Status: People valued for

VS
for what they do their position

Sequential: Order and Synchronous: Changing

VS
scheduling emphasized task priorities

Internal Focus: Focus on what External Focus: Focus on

VS
happens internally what happens externally
This research was instrumental in the development of the cloud-based SaaS
product, CultureScope.

Behavioural measurement
Behavioural measurement, otherwise known as behavioural diagnostics, is
all about ‘how’ we achieve the quantitative accurate measurement of the 30
behavioural factors. Behaviour measurement is achieved using two distinct
psychometric type questionnaires, as follows:

Individual behaviour: This pertains to how single members of the

organization act and respond. It is driven by individual capability and
the willingness to act in certain ways. For instance, an employee’s
dedication to continuous learning or their responsiveness to feedback.

Organizational behaviour: This reflects the organizational

environment. It’s the behaviour that employees observe around them
and is a powerful driver or inhibitor of individual behaviour. This
pertains to organizational processes, systems and established ways of
working. For example, the organization’s informal reward systems that
give accolades to employees who support their colleagues.

The above diagnostic method is already interesting in driving actionable

insights. Remember how in previous chapters we discussed that culture
change should be directed either at capability (the ‘People’) or the ‘Place’
for opportunity and motivation. Well, this measurement exactly
corresponds to this methodology. The individual behaviour measurement
corresponds to the ‘People’, and organizational behaviour measurement
corresponds to the ‘Place’.

To drive best diagnostic validity, the following question methods were

examined:

the normative method (Likert scale type questions)

the ipsative method (non-leading, scenario based and forced choice)

 Item Response Theory (most and least type questions)

The above are psychometric diagnostic methods that I should define.

Normative diagnostic methods

A normative questionnaire asks a respondent to indicate how much they

agree with a statement to which a Likert scale is attached. Normative items
are used mainly in personality questionnaires, and are important in
psychometric testing as they can delve into the personality traits of a
respondent and compare them to others who have taken the test. This
makes personality a measurable scale, which is easier to look at when
having to go through the recruitment of many people.

An advantage of this method is that there is no way for individuals taking

the test to deviate from what is being asked of them. It is therefore a
straightforward way of assessing and comparing people’s answers,
ultimately making an index of ‘normal’ characteristics from which an
employer could pick the most suited or desirable individuals.

The main fault with this type of approach is that it lends itself to problems
such as the fatigue effect, meaning the respondent may become bored of the
questionnaire and so just randomly tick boxes to get through the test
quickly.1

Respondents can make a conscious effort to distort the results in, for
example, a high-stakes selection situation (normally favourably). They also
may semi-consciously be trying to create a favourable impression. This can
be seen as individuals making the best of what they have rather than being
highly critical (analogous in an application form to making the career
highlights sound best and minimizing or omitting the negative outcomes,
without committing the act of giving inaccurate facts).

Such impression management, conscious faking and social desirability are

often difficult to differentiate in practice.2, 3

Ipsative diagnostic methods

Ipsative measurement involves the use of items for which the respondent is
required to choose one of two options. These forced-choice formats (either
rankings or ratings) and self-report questionnaires were introduced to
reduce response biases.

A major benefit of the ipsative format is to reduce ‘faking good’ by forcing

individuals to choose between options that are similar with respect to social
desirability. Acquiescence responding, halo effects, impression
management and social desirability bias are better controlled by means of
ipsative methods than by normative methods.

The most frequently mentioned advantages of ipsative measures include the

following:

ipsative measures have higher discriminability value than normative

tests4, 5

they are said to be resistant to social desirability, and respondents can

alter their results less as compared to normative tests6

ipsative measures seem to be resistant to ‘moderate responding’7

ipsative measures might better reflect choices people make in real life,
since they cannot choose all the possibilities, but are forced to choose
only some8

research proved that ipsativity can increase validity – if it reduces

response bias9

ITEM RESPONSE THEORY

The Item Response Theory (IRT) model represents an important innovation

in the field of psychometrics. The foundation of IRT is based on a
mathematical model defined by item parameters mainly for dichotomous
items.10
IRT was developed as an underlying response to forced-choice
questionnaires. In fact, despite their advantages in reducing response
biases, forced-choice-only questionnaires have been judged as limited
because their traditional scoring methodology results in ipsative data, very
special properties of which can pose threats to construct validity and score
interpretation as well as other substantial psychometric challenges. These
can be overcome with IRT.11

The basic assumption underlying the IRT model is that it tests behaviour
not at the level of test scores, but at the item level.12 These models have
been developed to deal with responses to items that are scored
dichotomously (i.e. only two possible scored responses exist, such as true-
false, correct-incorrect, endorsed-not endorsed, etc.).

Advantages of IRT are:

detailed descriptions of the performance of individual items

indices of item and scale level precision that are free to vary across the
full range of possible scores

the score estimation process is more precise, allowing simultaneous

consideration of both the number of right/endorsed items as well as
the properties (e.g. discrimination, difficulty) of each item when
estimating each person’s score on the construct being assessed

measures of response-profile quality

computer-adaptive testing, which helps to reduce testing time

The key purpose of testing different diagnostic methods was to examine the
best psychometric properties which are specifically required for
CultureScope behavioural measurement: inter-organizational discrimination
and consensual validity (homogeneity, stability, content and construct
validity). These psychometric indices are generally not available with other
culture measurements.
Finally, a combination of the ipsative method with IRT provides the best
diagnostic validity. CultureScope goes further by making use of Adaptive
Computer Testing (ACT) to drive enhanced validity and greatly reduce
assessment time. ACT means that each respondent is connected in real time
to the CultureScope diagnostic engine and accuracy as well as consistency
are being measured in real time; hence the system can adapt by asking
more and different types of questions to drive validity. Average completion
time for each questionnaire is 12 minutes.

Mapping behaviours to organizational

values/frameworks
Moving on from the science of behaviour measurement to focus on the
adaptation for showing whether you are ‘walking the talk’ for your
organizational values. As I already mentioned in prior chapters, there is no
point in having organizational values if you are not going to live these
values through behaviours. So, the fun part now is a mapping exercise
using the above dimensions to map into the values.

For this demonstration, let’s use the following three values:

Respect: We value the rules and environment in which we operate

safely. We fairly challenge poor performance through giving and
receiving feedback respectfully. We appreciate value brought by
different people and experiences.

Humility: We empower people; the leader isn’t in the best place to

make each and every decision. We invite and value other people’s
input and points of view. We enable opportunities for ourselves and
others and are dedicated to learning and development for career
growth.

Positivity: We maintain the highest level of quality, performance and

timely delivery, but not at the expense of our people’s wellbeing.
Everyone’s achievements are celebrated regardless of position. We
 engender an innovative and curious mindset in all employees and
partners.

In order to map the behavioural factors to the values statements, you now
need to refer to the full CultureScope factorial descriptors found in Table
4.2.

TABLE 4.2 Full behavioural factors descriptors

 A table listing the descriptors of full behavioural factors. The
row-wise data listed is as follows: 1. Goals are achieved through
partnerships and alliances: culture scope's 30 behaviours:
collective and individual; Maintaining high individual
performance and winning prevail over working relationships. 2.
The structure is hierarchical ensuring that leaders retain decision-
making: culture scope's 30 behaviours: disenfranchise and
empower; Decisions are delegated to lower levels with supportive
guidance. 3. A working environment that focuses on people
rather than purely on tasks and outputs: culture scope's 30
behaviours: people focus and delivery focus; There is a clear
focus on results and task outputs; employees are accountable for
their own actions. 4. Employees show restraint when expressing
thoughts and opinions: culture scope's 30 behaviours: neutral and
expressive; People do not hesitate to express their feelings and
openness is valued. 5. People value working within clear
processes and systems: culturescope's 30 behaviours:
standardized and flexible; People value variability and deal with
each situation afresh. 6. Individuals value working steadily within
assigned and streamlined processes favouring stability: culture
scope's 30 behaviours: moderate and radical; Continuous
improvement and evolving ideas are valued; there is an emphasis
on responding differently to different situations. 7. Work is
delivered through collaborating group efforts and team results are
prioritized over individual results: culturescope's 30 behaviours:
team focus and self-focus; Individualism and independence are
encouraged and rewarded. 8. The emphasis is on short-term
delivery and results: culturescope's 30 behaviours: tactical and
strategic; The emphasis is on long-term delivery, results; and
focus is on the wider impact. 9. Active steps are taken to improve
employees' skill sets and careers; employee growth and
development is considered an integral part of the job: active
learning and passive learning; on-the-job experience is seen as
the best approach for learning and skill development. 10. New
and creative ideas are pioneered. Intelligent risk-taking is
encourage and praised: culturescope's 30 behaviours: innovate
and consolidate; predictability and control are preferred. Risk-
taking is avoided until it is absolutely necessary Tried and tested
 taking is avoided until it is absolutely necessary. Tried and tested
methods are encouraged. 11. People place a high importance on
laws, rules, values and obligations. They try to deal fairly with
people based on these rules, but rules come before relationships:
culturescope's 30 behaviours: conformity and nonconformity;
people believe that each circumstance, and each relationship,
dictates the rules that they live by. Their response to a situation
may change, based on what is happening in the moment, and who
is involved. 12. People believe that relationships don't have much
of an impact on work objectives and, although they are important,
people believe that they can work without having good
relationships: culture scope's 30 behaviours: transact and
relationship; People believe that good relationships are vital to
meeting business objectives, and that their relationships with
others will be the same. People spend time building relationships
with colleagues and clients. 13. People believe that you are what
you do, and they define your worth accordingly. These cultures
value performance independent of status: culturescope's 30
behaviours: achievement and status; People believe that you
should be valued for your status. Power, title and position matter
in these cultures, and these roles define behaviour. 14. People like
events to happen in order. They place a high value on punctuality,
quality, planning and staying on schedule. In this culture, 'time is
money,' and people don't appreciate it when their schedule is
thrown off: culturescope's 30 behaviours: sequential and
synchronous; People see the past, present and future as
interwoven periods. They often work on several projects at once,
in a fast-paced manner, and view plans and commitments as
flexible. 15. People believe that they can control nature or their
environment to achieve goals. This includes how they work with
teams and within organizations: culture scope's 30 behaviours:
internal focus and outer focus; People believe that their
environment controls them; they want to work within their
environment to achieve goals. At work or in relationships, they
focus their actions on others.

Your exercise is to read each organizational value and the call to action
associated with each value, then choose which of the above behaviour
factors map to each value. You could do this exercise with a team if you
wish. The point of this exercise is to explain ‘how’ the organizational
values should manifest in behaviours.

I suggest you break down each call-to-action statement as in Table 4.3.

TABLE 4.3 Mapping example

Skip table

Mapped
Value Text ‘call to action’
behaviour

We value the rules and environment in

Respect Conformity
which we operate safely.

We fairly challenge poor performance Expressive,

through giving and receiving feedback Achievement,
respectfully. People Focus

We appreciate value brought from Collective,

different people and experiences. People Focus

Now do the same for the other two organizational values.

If you have involved a team, it would be interesting to see what different

people have mapped and why. It’s an illustration of how different people
interpret the values and hence why a behavioural approach becomes
important. Remember, you do not have to map all the behaviours; this is
normal in organizational culture as you emphasize only the behaviours that
are perhaps critical, otherwise it becomes rather robotic.

I hope you have enjoyed this exercise. See a full mapping in Table 4.4.
TABLE 4.4 Full mapping

Skip table

Mapped
Value Text ‘call to action’
behaviour

We value the rules and environment in

Respect Conformity
which we operate safely.

We fairly challenge poor performance Expressive,

through giving and receiving feedback Achievement,
respectfully. People Focus

We appreciate value brought from Collective,

different people and experiences. People Focus

We empower people; the leader isn't in

Humility the best place for each and every Empower
decision.

We invite and value other people's input

Expressive
and points of view.

We enable opportunities for ourselves

Active
and others and are dedicated to learning
Learning
and development for career growth.
 Mapped
Value Text ‘call to action’
behaviour

We maintain the highest level of quality,

performance and timely delivery, but Sequential,
Positivity
not at the expense of our people’s People Focus
wellbeing.

Everyone’s achievements are celebrated

Achievement
regardless of position.

We engender an innovative and curious

Innovate
mindset in all employees and partners.

I hope you have ended up with something like this yourself! Let’s create a
cluster map to check that we are not over-mapping, where the behaviours
mapped are exactly the same for all the values (however, some overlapping
is great as that will show that the values are connected and hence some
behaviours are critical).

Another check would be the absence of competing behaviours, where you

have the behaviour opposites mapped. This can happen, but is rather
dangerous as the values are sending conflicting messages to employees.
See Table 4.5.

TABLE 4.5 Cluster values behavioural mapping

Skip table

Respect Humility Positivity

 Respect Humility Positivity

Conformity

Expressive Expressive

Achievement Achievement

People Focus People Focus

Collective

Empower

Active Learning

Sequential

Innovate

The mapping works, the values are distinct by behaviours and there are no
competing behaviours mapped. The behaviours of Expressive,
Achievement and People Focus are mapped twice, which is a clear
indication of the importance of these behaviours for this organization.

We now have behavioural clarity, hence launching behavioural

measurement is possible as you know what you are measuring.
Measurement as described before will articulate the following:

Are our people living our values? If not, why not (which behaviours
are absent)? Remember, this relates to capability.

Does our workplace support and drive our values? If not, why not
(which behaviours are absent)? Remember, this relates to
opportunities and motivation.

In Chapter 5 you will learn about analytics and insights using the above
example.

Notes
1. 1 C-C Bowen, B A Martin and S T Hunt (2002) A comparison of
ipsative and normative approaches for ability to control faking in
personality questionnaires, International Journal of Organizational
Analysis, 10 (3), pp. 240–59

2. 2 Bowen, Martin and Hunt, ibid

3. 3 Bowen, Martin and Hunt, ibid

4. 4 H Baron (1996) Strengths and limitations of ipsative measurement,

Journal of Occupational and Organizational Psychology, 69, pp. 49–
56

5. 5 L E Hicks (1970) Some properties of ipsative, normative and forced-

choice normative measures, Psychological Bulletin, 74, pp. 167–84

6. 6 Bowen, Martin and Hunt, ibid

7. 7 H Baron, ibid

8. 8 L V Gordon (1951) Validities of the forced-choice and questionnaire

methods of personality measurement, Journal of Applied Psychology,
35, pp. 407–12
 9. 9 Hicks, ibid

10. 10 S E Embretson and S P Reise (2000) Item Response Theory for

Psychologists, Lawrence Erlbaum Associates Publishers

11. 11 Embretson and Reise, ibid

12. 12 F B Baker and S-H Kim (2004) Item Response Theory: Parameter
estimation techniques (2nd ed), CRC Press
5 Data analytics concepts for actionable insights
Are you walking the talk?

As this is a practical and evidence-based book for culture analytics, I might as

well turn the three organizational values behavioural mapping completed in the
previous chapter into insights to answer the often-lingering question, are you
living your values? Or, as I often hear, are you walking the talk? And if you are
not walking the talk, why?

Once you know the ‘why’, you will be able to design the behaviourally
focused solution, thus turning behavioural insight into action. After all, and as
we discussed before, actionable means working on the ‘why’, which provides
the ‘how’ of culture change.

TABLE 5.1 A values-based heatmap

A table depicting a value-based heatmap. The row-wise data listed

is as follows: 1. The place: respect: present; humility: absent;
positivity: absent. 2. The people: respect: present; humility: very
absent; positivity: somewhat present.

Table 5.1 illustrates the following insights:

The value of Respect is behaviourally present at the Place and for the
People, meaning that people tend to behave in that way and the
organization tends to drive this. The presence for both people and place
results in the full manifestation of the value of Respect.

The value of Humility is absent and very absent at the Place and for the
People respectively.

The value of Positivity is absent at the Place, and somewhat present

(sometimes behave in that way) for the People. This will result in the
value not being manifested fully, as you may remember we need it to be
present for both the Place and the People.
The immediate question to follow the above insight is, why are you living your
values (or not)? We can now take a detailed look at the behaviours, as
illustrated in Table 5.2.

TABLE 5.2 The presence or absence of the behaviour factors driving

the three organizational values

Skip table

Respect Humility Positivity

Mapped
Behavioural
Dimensions
People Place People Place People Place

Conformity Present Present

Expressive Absent Present Absent Present

Achievement Present Present Present Present

People
Present Present Present Present
Focus

Very
Collective Present
Present

Vey
Empower Absent
Absent

Active Somewhat Somewhat

Learning present present
 Respect Humility Positivity
Mapped
Behavioural
Dimensions
People Place People Place People Place

Somewhat
Sequential Absent
present

Somewhat Very
Innovate
present Absent

You can see that the absence of four behaviours is the answer to ‘why’, and can
be summarized as follows:

1. Expressive (being open in expressing thoughts and opinions): The

organization generally gives employees opportunities to be open in
expressing thoughts and opinions; however, employees show restraint and
tend to not be open with their ideas and concerns. How can we increase
the capability and willingness of people to be more expressive?

2. Empower (delegating decisions to lower levels with supportive

guidance): The organization is encouraging hierarchical ways of working
where leaders tend to retain decision-making power. How can we create
more opportunities and motivation for better delegation and
empowerment? How can we build the capability for empowerment?

3. Sequential (ordering and scheduling is emphasized): People like events to

happen in order. The organization is not focused on punctuality, quality,
planning and staying on schedule. Hence the Place is not providing that
opportunity.

4. Innovate (encouraging new approaches and ideas rather than always

leaning into predictability and control): The Place does not encourage
acting on new ideas and approaches. How can we provide the opportunity
and the motivation to be more innovative?
If you like your analytics and numbers, let me provide Table 5.3.

TABLE 5.3 A numeric heatmap for the presence or absence of the

behaviour factors driving the three organizational values

Skip table

Respect Humility Positivity

Mapped Behavioural
Dimensions
People Place People Place People Place

Conformity 3.4 3.2

Expressive 2.1 3.3 2.1 3.1

Achievement 3.5 3.1 3.5 3.1

People Focus 3.4 3.2 3.4 3.2

Collective 4.1 3.6

Empower 2.4 1.6

Active Learning 2.5 2.7

Sequential 2.6 2.1

Innovate 2.5 1.5

Table 5.3 represents a behavioural factor scale from 0 to 5. The scale
interpretation is as per Figure 5.1.

FIGURE 5.1 The behavioural factors scale interpretation

A scale lists the behavioural factors with interpretation ratings.

Figure 5.1 details

Since each behavioural factor has a dichotomous paired factor, a reminder that
the absence in one direction will indicate a behaviour’s presence in the other
direction. This can be demonstrated using the radar graph in Figure 5.2. Notice
the Expressive vs Neutral result.

FIGURE 5.2 The dichotomous behavioural factor pairs and the scale

A circle with multiple axes representing different dichotomous

behavioural factor pairs.

Figure 5.2 details

Expanding on our previous chapters, we can now decide on what sort of

interventions are needed – is it for the Place (opportunity and motivation) or
the People (capability)? The same behavioural heatmap can be used to
demonstrate this, as per Table 5.4.

TABLE 5.4 The targeted behaviour factors and the type of

intervention needed

Skip table

Respect Humility Positivity

Mapped
Behavioural
Dimensions
People Place People Place People Place

Conformity 3.4 3.2

 Respect Humility Positivity
Mapped
Behavioural
Dimensions
People Place People Place People Place

Expressive 2.1 3.3 Capability 3.1

Achievement 3.5 3.1 3.5 3.1

People
3.4 3.2 3.4 3.2
Focus

Collective 4.1 3.6

Opportunity
Empower Capability &
Motivation

Active
2.5 2.7
Learning

Opportunity
Sequential 2.6 &
Motivation

Opportunity
Innovate 2.5 &
Motivation
From Table 5.4, it can be seen the two behaviours need a capability-type
intervention (Expressive and Empower), and three behaviours need an
opportunity and motivation-type intervention (Empower, Sequential and
Innovate).

The insights shown in Table 5.3 are a behaviour-first approach to delivering

insights. It means that we are opportunistically targeting the desired behaviours
based on the mapping to values exercise I introduced. Hence you can use
demographic data to analyse the presence or absence of these behaviours,
providing the ‘why’ and the ‘where’ of behaviours. This is also known as
opportunistic modelling.

This is a good time to further explain what behaviour-first or outcome-first

analytics for insights really mean. Figure 5.3 shows this concept well: having
measured the behaviours of the People and the Place, and defined the
behaviour mapping to your organizational values (in the define and measure
steps covered in Figure 2.2), you can now add and connect to demographics,
such as function, division, tenure, seniority, gender, etc.

This will result in ‘where’ the targeted behaviours are present or absent, which
can be presented in simple demographic behavioural heatmaps like Table 5.2.

FIGURE 5.3 A connected data approach to provide behaviour-first

or outcome-first analytics for insights

A composite of three illustrations shows a connected data

approach.

Figure 5.3 details

However, when you have many demographics, is there an easier way of

presenting the ‘where’ for specific and targeted behaviours on one page? The
answer is yes! This is the quadrant distribution method demonstrated in Figure
5.4.

FIGURE 5.4 The ‘where’ of behaviour presence or absence using the

quadrant dispersion method

A graph shows the presence and absence of where of behaviour in

quadrants.
 The bubbles represent size of the population as a proportion of the
total population within each quadrant.

Figure 5.4 details

Figure 5.4 plots the Empower behaviour, with the X axis showing presence for
the People, and the Y axis showing presence at the Place. This can then be
populated with the applied demographics for the four resulting quadrants.
Exploring each quadrant can prove the ‘where’ for Empower: for example,
which of your organizational functions exists in each quadrant? Is there a
difference by gender? By tenure or by seniority? A quick and easy way to
examine one or more behaviours with length heatmaps.

As for the analysis, the demographics existing in the ‘Strongly Aligned’ zone
have the Empower behaviour present for People and Place, and that is where
you should go to learn from with a view to replicating it. The zone below
marked ‘Look at the Place’ is where the capability for the Empower behaviour
is present for the People, but absent at the Place – that is, the Place doesn’t
provide the opportunity nor the motivation, resulting in this behaviour not
being fully manifested. (Remember what I explained before – both the People
and Place are needed for a behaviour to fully manifest.) The People making up
the demographics in this zone can be a quick win if you can fix the issues with
opportunity and motivation. Hence, in future you would expect them to exist in
the ‘Strongly Aligned’ zone.

The zone marked ‘Look at the People’ shows you that the Empower behaviour
is present at the Place as both opportunity and motivation are present, but
absent by capability, which is what interventions should address. As an
example, an organization is communicating to all leaders at all levels that
empowerment is something they need to do in their daily work, and the
organization is explaining the benefits of empowerment, but leaders in this
zone indicate the absence of capability. This means that leaders are not sure
how to empower their teams or the responsibilities, decisions and actions for
empowerment.

That leaves us with that last zone. Essentially, the People making up these
demographics indicate the absence of capability, opportunity and motivation
for the Empower behaviour to manifest. Normally the People in this zone are
the biggest drag when driving change and transformation.
The analytics for actionable insights process described in this chapter is
framework-led; the process starts with frameworks that are available, followed
by the behavioural mapping exercise to generate the behavioural clusters, to
answer the key question of ‘are we living our framework or our values?’. For
clarity, a framework could include organizational values, risk and compliance,
regulatory frameworks, etc.

If the specific question is about which behaviours drive or even predict a given
outcome, the approach will be outcome-led analytics to drive actionable
insights. This method will be covered in the next chapter.
6The link between organizational culture and its impact on outcomes
Why did an incident happen? Which behaviours do we need to focus on to drive productivity and performance
while maintaining psychological safety? How can we drive innovation?

These are all questions that need to be answered, and it may not surprise you to know that workplace culture plays
a significant part in answering them. That is, behaviours become the actionable insight that you need to work on to
affect the outcomes above.

Following on from the last chapter, in this chapter I will demonstrate the outcome-first approach. It is inevitable
that I will have to visit some aspects of statistical maths to achieve the crown of all insights that is predictive
analytics for actionable insights. So strap in while we visit some definitions first, to demystify predictive analytics.

Defining key statistical terms

The previous chapter metrics were all driven by what is known as ‘descriptive statistics’. As the name suggests,
this method helps articulate the current picture, giving you the ability to describe the trend using the following:1

Mean: this is the average of the measured numbers, and is calculated by dividing the sum of measured
numbers (say the measured behaviour factors) by the total number of numbers. Mean = (Sum of all the
measured observations/Total number of observations, also known as sample size). This method was used in
the behaviour heatmaps in the last chapter.

Median: this is the middle number in a sorted ascending or descending list of numbers, and can be more
descriptive of that data set than the average. It is the point above and below which half (50 per cent) of the
observed data falls, and so represents the midpoint of the data.

Standard deviation: this is a measure of the amount of variation of a random variable expected about its
mean. A low standard deviation indicates that the values tend to be close to the mean, while a high standard
deviation indicates that the values are spread out over a wider range. This is very useful to describe the data
in the distribution quadrant methodology I demonstrated in the last chapter.

Now let’s move on from descriptive to predictive statistics. The term ‘predictive statistics’ (or analytics) refers to
the use of statistics and modelling techniques to make predictions about future outcomes and performance.
Predictive analytics looks at current and historical data patterns to determine if those patterns are likely to emerge
again. Predictive analysis can also be used to improve operational efficiencies and reduce risk.

In behavioural science and specifically for culture metrics, you will get to accurately know ‘why’ it happened,
‘how’ to improve, change or sustain your outcomes, and, in our approach, your return on investment to drive
behaviour change. The primary methodology to achieve predictive analytics is multiple regression, which needs
some definition.

Multiple regression is a statistical technique that can be used to analyse the relationship between a single
dependent variable and several independent variables.2 The objective of multiple regression analysis is to use the
independent variables whose values are known to predict the value of the dependent value. Multiple regression
analysis allows researchers to assess the strength of the relationship between an outcome (the dependent variable)
and several predictor variables, as well as the importance of each of the predictors to the relationship.

Multiple regression is used when we want to predict the value of a variable based on the value of two or more
other variables. The variable we want to predict is called the dependent variable (or sometimes, the outcome,
target or criterion variable). The variables we are using to predict the value of the dependent variable are called the
independent variables (or sometimes, the predictor, explanatory or regressor variables).

For example, you could use multiple regression to understand whether exam performance can be predicted based
on revision time, test anxiety, lecture attendance or gender. The output from this analysis will explain which of the
four variables predict exam performance and how good this prediction is.

Multiple regression also allows you to determine the overall fit (variance explained) of the model and the relative
contribution of each of the predictors to the total variance explained. For example, you might want to know how
much of the variation in exam performance can be explained by revision time, test anxiety, lecture attendance or
gender ‘as a whole’, but also the relative contribution of each independent variable in explaining the variance.

The parsimonious model is a model that accomplishes the desired level of explanation or prediction with as few
predictor variables as possible. The goodness of fit of a statistical model describes how well it fits a set of
observations. The spurious model occurs when two factors appear causally related to one another, but are not. The
appearance of a causal relationship is often due to similar movement on a chart that turns out to be coincidental or
caused by a third ‘confounding’ factor.3

The above will be sufficient to understand the methodology I am about to explain; however, if you’d like to know
more about regression, I will provide you with a quick summary of how to read regression results.

Typically, a regression primary output summary will look like Table 6.1, and the parameters are defined as
follows:

Multiple R is the correlation coefficient that measures the strength of a linear relationship between two
variables. The correlation coefficient can be any value between -1 and 1, and its absolute value indicates the
relationship strength. The larger the absolute value, the stronger the relationship:

1 means a strong positive relationship

-1 means a strong negative relationship

0 means no relationship at all

R square is the coefficient of determination, which is used as an indicator of the goodness of fit. It shows
how many points fall on the regression line. The R2 value is calculated from the total sum of squares; more
precisely, it is the sum of the squared deviations of the original data from the mean.

Adjusted R square is the R square adjusted for the number of independent variables in the model. You will
want to use this value instead of R square for multiple regression analysis.

Standard error is another goodness-of-fit measure that shows the precision of your regression analysis – the
smaller the number, the more certain you can be about your regression equation. While R2 represents the
percentage of the dependent variables variance that is explained by the model, standard error is an absolute
measure that shows the average distance that the data points fall from the regression line.

Observations is simply the number of observations in your model, also known as sample sizeTable 6.1.

TABLE 6.1 Regression summary output

Skip table

SUMMARY OUTPUT

Regression Statistics

Multiple R 0.491332608

R Square 0.241407732
 Adjusted R Square 0.238887492

Standard Error 0.392270373

Observations 303

A secondary regression output will look like Table 6.2, and the parameters are defined as follows:

df is the number of the degrees of freedom associated with the sources of variance.

SS is the sum of squares. The smaller the residual SS compared with the total SS, the better your model fits
the data.

MS is the mean square.

F is the F statistic, or F-test for the null hypothesis. It is used to test the overall significance of the model.

Significance F is the P-value of F. The lower this value, the higher the significance of this model.

TABLE 6.2 Secondary regression output: Analysis of Variance (ANOVA)

Skip table

ANOVA

df SS MS F Significance F

Regression 1 14.73941599 14.73941599 95.78759297 8.0871E-20

Residual 301 46.31668962 0.153876045

Total 302 61.05610561

And the final part of a regression output is understanding the components of your analysis. As Table 6.3 shows,
the most useful component in this section is Coefficients as it enables you to build a prediction; basically, how
quickly the outcome will improve by changing the input, also known as the beta (β) coefficient. The beta describes
the slope of the line for the outcome prediction.

As an example, see Figure 6.1. Notice how quickly the predicted outcome changes when the input changes. This
slope is described by the β coefficient.

TABLE 6.3 Regression component analysis

Skip table

Coefficients Standard t Stat P-value Lower 95% Upper 95% Lower Upper
 Error 95.0% 95.0%

4.54599E-
Intercept 1.951043878 0.088186475 22.12407158 1.777503784 2.124583972 1.777503784 2.124583
65

X
8.0871E-
Variable 0.274297483 0.028026392 9.787113618 0.219145004 0.329449962 0.219145004 0.329449
20
1

FIGURE 6.1 A graph showing the β coefficient

Figure 6.1 details

Well, now that we’ve got the definitions out of the way, I hope I have demonstrated that interpreting predictive
analytics results and turning them into actionable insights can be a complex business. This is exactly why I will be
providing you with a vastly simpler way of representing the power of predictive analytics without compromising
the methodology. This will be demonstrated in Chapter 11.

Let’s turn our attention to the outcome-first method and how to use it.

As the name suggests, the method is about starting with the outcome, hence no behaviour mapping is assumed, as
follows.

The first step is to obtain the data set containing behavioural data for the people and the place as per the diagnostic
method outlined in Chapter 4. Obtain the outcome data at individual level as a preference, e.g. performance
metrics or participants responding to outcome/sentiment questions.

Next, the multiple R coefficient value between each behaviour and the outcome should be obtained, using the
multiple regression – this is the first-level modelling. There are 16 variations of regression analysis, e.g. linear,
non-linear, polynomial, logistic, quantile, etc. The most relevant ones are linear regression for linear outcomes and
logistic regression for binary outcomes. The behaviours with the best multiple R should be selected for each
outcome.

Differences in regression models should be tested to identify the most parsimonious nested model – the best
combination of CultureScope behaviours for predicting the outcome – this is second-level modelling. After years
of research, I have always found that a nested behavioural model is a better predictor of outcomes than singular
behaviours.

Never rely on just one thematic outcome to create the most parsimonious model. You should try to get as many
outcome metrics as you can for a given theme, which could include sentiment, situational and measured data; this
approach will act as a good validation method for your resulting model. To demonstrate, I will elaborate further in
Chapter 9 on workplace culture outcome themes and how to leverage the behavioural data collected for multiple
themes for connected analytics and actionable insights.

A key question I need to address is: for a given population size, what is the data collection completion rate needed
to provide model statistical validity, and for analytics for insights validity? Let’s define and answer each one as
follows.

Statistical validity can be defined as the extent to which drawn conclusions of a research study can be considered
accurate and reliable from a statistical test. To achieve statistical validity, it is essential for researchers to have
sufficient data and to choose the right statistical approach to analyse that data. The statistical validity approach I
took during my seven-year research, as outlined in Chapter 4, was derived from the effect size method, which tells
you how meaningful the relationship between variables or the difference between groups is.4 It indicates the
practical significance of a research outcome; in most cases a minimum 30 per cent completion rate of a given
population achieves statistical validity.

Analytics validity is defined by the number of criteria-based benchmarks one needs to generate to show
demographic differences once a module has been developed. Essentially, the more the demographics you apply, a
higher completion rate is needed. In my research outlined in Chapter 4, I found that ideally n ≥ 5 (five participants
or more are needed) for any generated criteria-based benchmark.

You may wonder: why five? This protects anonymity and, critically, will stop an outlier from totally deforming the
model for a generated benchmark so as to render it unreliable.

Putting the above method into practice, let’s go back to our example from Chapter 5, where the collected data was
from 3,122 participants representing a 62 per cent completion rate. We now need to forget about the mapping
exercise and assume nothing; that is, apply all 30 behavioural factors for the People and 30 for the Place as the
independent variables, and the outcome as the dependent variable.

For this organization, the key question they wanted answering was: does our workplace culture support
performance? If so, which behaviours do we need to focus on? I will now apply the performance scorecard data
that this organization has been collecting as the dependent variable. The performance scale is as follows:

1. Does not meet performance expectations

2. Partially meets performance expectations

3. Consistently meets performance expectations

4. Often exceeds performance expectations

5. Consistently exceeds performance expectations

You can see from the above that the outcome is on a five-point scale, indicating that linear regression analysis is
likely to be the best analysis method. Having applied this regression, the key component results are shown in
Table 6.4 (focusing on a few important metrics only to simplify).

TABLE 6.4 Regression results key components

Skip table
 Behavioural Factors Achievement, Empower, Radical, Consolidate, Team Focus

Multiple R 0.33

Standard Error 0.812

Number of observations 3,122

Significance F 3.30753663061259E-08

Beta Coefficient 0.55

The regression showed that five People behaviours are the predictors for performance. The behavioural factors
are:

Achievement: People believe that you are what you do, and they define your worth accordingly. These
cultures value performance independent of status.

Empower: Decisions are delegated to lower levels with supportive guidance.

Radical: Continuous improvement and evolving ideas are valued; there is an emphasis on responding
differently to different situations.

Consolidate: Predictability and control are preferred. Risk taking is avoided until it is absolutely necessary.
Tried-and-tested methods are encouraged.

Team Focus: Work is delivered through collaborative group efforts and team results are prioritized over
individual results.

These behaviours yield a very good multiple R of 0.33, which ranks at the 80th predictive percentile of effect sizes
observed as compared to the research I mentioned earlier.5 For ease of reference, see Figure 6.2.

Analysing the other regression components, standard error is low and significance F is almost zero, which implies
a highly significant model. A beta coefficient of 0.55 provides a good prediction model, as per Figure 6.3.

FIGURE 6.2 Correlational effect size benchmarks

A table lists the benchmarks based on correlational effect size.

Figure 6.2 details

FIGURE 6.3 The predictive relationship between the nested five behaviours and performance rating
 Figure 6.3 details

Now that we have developed the behavioural model for performance as a workplace culture theme, in the next
chapter I will illustrate how we can apply the same analytics lens as in Chapter 5 to this model, to show to what
degree these five behaviours are present or absent, and crucially where in the organization. Remember, if you want
to drive performance, that’s what you need to focus your interventions on.

Notes
1. 1 P Saville and T Hopton (2014) Psychometrics@work, Saville Consulting

2. 2 A W Moore, B Anderson, K Das and W-K Wong (2006) Combining Multiple Signals for Biosurveillance,
Handbook of Biosurveillance, pp. 235–42

3. 3 A Gupta. Model Selection Techniques – Parsimony & Goodness of Fit, Medium, 13 December 2020,
https://medium.com/geekculture/model-selection-techniques-parsimony-goodness-of-fit-fc2f1863ccfd
(archived at https://perma.cc/7TSG-9ZWL)

4. 4 F A Bosco, H Aguinis, H, K Singh, J G Field and C A Pierce (2015) Correlational effect size benchmarks,
Journal of Applied Psychology, 100,pp. 431–49

5. 5 Bosco et al, ibid

7 Applying predictive analytics to culture
metrics
Having established from the previous chapter, through predictive analytics,
the five behaviours that drive your workplace culture for performance, we
can now establish a behavioural heatmap – firstly for the entire population,
then looking at all benchmark demographics through that performance
behavioural lens.

Figure 7.1 shows that for the cluster of all five behaviours for the entire
measured population, the behaviours that drive performance are not all that
present: the People yield a mean of 2.78 and the Place 2.67 (the scale is
from 0 to 5). We really need to be aiming at a mean of 3.1 or more to truly
drive performance, and as you can see, both the People and the Place need
interventions to drive capability, opportunity and motivation.

FIGURE 7.1 Heatmap results for the five behaviours that drive
performance workplace culture

A composite of three illustrations shows the heatmap results for

the five behaviours that increase performance in the workplace.

Figure 7.1 details

To make the actionable insight more meaningful, we can use the quadrant
method to examine where in the organization all five behaviours are present
for the Place, or the People, or both. Further, I will set the quadrant
threshold to 3 for both People and Place to drive meaningful insights (if
you are wondering why 3, remember this is the presence threshold on the
behaviour scale).

The result is shown in Figure 7.2, revealing that one team in the
organization does have all five behaviours present for both People and
Place, the Operations team. Furthermore, 63 per cent of the Marketing team
have all five behaviours present, although they are absent at their Place.
This is also encouraging, as fixing the Place for this team means that we
move them to the greatness zone! Still, the team to learn from is
Operations, and they can help by being the change champions.

(I just revealed another advantage to the quadrant distribution method; that

is, to show you where your culture champions are.)

What about the two remaining zones in the quadrant? The top left zone
indicates that Place is enabling the five behaviours; however, you will need
to build capability for the People. One example here is the Technology
team. For the final quadrant, in the bottom left, the five behaviours are
absent for both Place and People. In this zone, you have the most ‘drag’ as
you will need to design interventions that will build capability and provide
opportunity and motivation to enact these behaviours. As you can see from
Figure 7.2, the Support department requires these focused interventions.

FIGURE 7.2 Quadrant distribution diagram showing where the

five behaviours driving performance are present

A graph shows the presence of where five behaviours driving

performance results in quadrants.

Figure 7.2 details

You can also use this quadrant distribution method to examine singular
behaviours, ending up with five diagrams giving further insight into each
behaviour. This more detailed approach can be very useful if you want to
find singular behaviour champions, especially in relatively large
organizations, where a change programme may need many culture
champions to maximize success.

I have mentioned workplace culture champions twice now, and you may be
wondering: what does that really mean? Why call individuals in the top
right corner of the quadrant ‘champions’? The champions have all five
targeted behaviours as present for them and their Place, hence they are
walking the talk! That alignment results in highly engaged individuals, and
this is meaningful engagement, as it drives the targeted performance culture
outcome. Hence you will want them to actively inspire and energize the rest
of the organization.
By contrast, the zone at the bottom left corner of the distribution quadrant
will have individuals that are aligned to Place as all five behaviours are
absent for them and their Place, hence behavioural alignment and possible
engagement; however, for the wrong reasons, as this alignment works
against the performance culture outcome. I will cover this critical topic
further in the next chapter.

We can further validate this workplace culture for performance model by

producing five benchmarks for the five-point performance scale; the scale
banding is from ‘consistently exceeds’ to ‘does not meet’ performance
criteria, as already shown in the previous chapter. This will illustrate how
the presence or absence of the five predictive behaviours fares against the
performance scale.

Figure 7.3 is a line graph showing and validating that the individuals
awarded the best performance do indeed have the five behaviours present.
You also see that the individuals with the lower performance grades exhibit
lower presence of the five behaviours. In fact, ‘does not meet performance
criteria’, which is the lowest performance grade, has all five behaviours
absent.

Notice the difference between the People and the Place in Figure 7.3; in all
the performance criteria, the People behaviours are more likely to be
present. So the Place will impede performance, which is not what you
want.

These results also beg a few questions: is the Place the key reason for the
People’s behaviours and hence performance? Should the five behaviours at
the Place be somewhat consistent across all performance groups? So many
questions to further investigate about performance culture, which should be
a combination of quantitative and qualitative.

FIGURE 7.3 Line graph showing performance grade against the

five behaviours

A line graph shows the performance-based results of the five

behaviours.
 Figure 7.3 details

Having established a behavioural predictive model for performance culture,

you can now compare the behaviours that drive your outcome with the
behaviours mapped to your organizational values. The reason for
comparing the mapping against the developed model for one or more
outcomes is to answer a secondary question regarding your organizational
values. Let me elaborate: you may remember we did a mapping exercise
back in Chapter 4, in order to look at the behavioural results with that
applied lens in Chapter 5, answering the primary question of: are we living
our values? Ok, so what? If we are living our values, do we also achieve
our desired outcomes? After all, that is what you want out of your values
and workplace culture.

To answer this secondary, but very important question for organizational

values, we need to make sure that some if not all the mapped behaviours
are the same as per the outcome or even several outcomes in the developed
model, and we also need to make sure that we do not have any competing
behaviours, referring to the dichotomy construct of the behavioural pairs.

As an example, if one of your organization’s values is Standardized

behaviour, but the regression model for outcomes indicates the need for
Flexible behaviour, being the dichotomous partner as illustrated in Figure
5.1 in Chapter 5, this will result in a conflict between living your
organizational values and driving your desired outcomes.

The comparison between the organizational values behavioural mapping

and predictive performance culture is illustrated in Table 7.1. It can be seen
that we have a match with two of the behaviours, Achievement and
Empower. A further two behaviours are not matched, but are simply
additive as part of the performance culture model. That is fine – they could
be unique to this culture theme, hence need focusing on just for that
outcome.

If these two behaviours keep recurring while we develop more outcome

culture themes, we can look to integrate them into your organizational
values and perhaps adapt or evolve your values – an interesting idea and
method in making your organizational values true to intended behaviours,
and another example of linking outcomes to values.

TABLE 7.1 Comparison between the organizational values

behavioural mapping and predictive performance culture

Skip table

Performance Culture
Respect Humility Positivity
Model

Conformity

Expressive Expressive

Achievement
An icon showing a
Achievement Achievement
solid circle with a tick
mark in it.

People People
Focus Focus

Collective

Empower
An icon showing a
Empower
solid circle with a tick
mark in it.
 Performance Culture
Respect Humility Positivity
Model

Active
Learning

Sequential

Consolidate
An icon showing a
Innovate
solid circle with an
exclamation mark in it.

Radical

Team Focus

What about the elephant in the room? I am referring to that fifth behaviour
– yes, you may have noticed that Consolidate is the dichotomous partner of
Innovate, which is mapped to your organizational values. Shock and
horror! We don’t need to be innovative to drive performance culture; this is
a good example of ‘never assume’. It’s that old syndrome: ‘if it works
somewhere, it should work for us’. And this example shows how each
organizational culture is unique and there is no one-size-fits-all; again, I
will further elaborate in the next chapter.

For now, let me offer an interesting observation rather than a hypothesis

based on the performance outcome; could this example simply be a case of
wrongly mapped behaviour? Notice that Radical behaviour in Table 7.1 –
perhaps that should be the mapped behaviour and we just need to fine-tune
the organizational values behavioural mapping? It could be an easy
mistake, and the difference in culture can be simply summarized as
‘evolution rather than revolution’.

How do we adapt your organizational value ‘Positivity’ to reflect this

change? Let us start by looking at the definitions of Innovate vs Radical:

Innovate: New and creative ideas are pioneered. Intelligent risk taking
is encouraged and praised.

Radical: Continuous improvement and evolving ideas are valued;

there is an emphasis on responding differently to different situations.

Now let’s recall the mapping for ‘Positivity’ from Table 4.3 – this is shown
in Table 7.2.

TABLE 7.2 Behavioural mapping for Positivity

Skip table

We maintain the highest level of quality, Sequential,

Positivity performance and timely delivery, but not People
at the expense of our people’s wellbeing. Focus

Everyone’s achievements are celebrated

Achievement
regardless of position.

We engender an innovative and curious

Innovate
mindset in all employees and partners.

The words used in your organizational values, ‘innovative and curious

mindset’, should now be changed, based on what we have learnt from the
performance outcome. Perhaps that should read, ‘As we evolve, we
engender continuous improvement in all employees and partners’. The
resulting mapping is demonstrated in Table 7.3.
 TABLE 7.3 Changed values wording and relevant behavioural
mapping

Skip table

We maintain the highest level of quality, Sequential,

Positivity performance and timely delivery, but not People
at the expense of our people’s wellbeing. Focus

Everyone’s achievements are celebrated

Achievement
regardless of position.

As we evolve, we engender continuous

improvement in all employees and Radical
partners

Although this example of organizational values adaptation is based on just

one outcome, and in practice we will need to validate against many
outcomes to be absolutely sure, I hope that the process is now clear for how
you can make your organizational values steeped in reality. Walking the
talk drives your desired and intended outcomes.

The process in this chapter illustrates the linking that should happen
between organizational values, workplace culture, behaviours and intended
outcomes. Unfortunately, this process rarely happens in organizations, and
if it does it’s a disconnected, qualitative process at best, rendering
organizational values ‘just nice words’!
8 You have to know the ‘why’ before you get to
the ‘how’ and ‘where’
I hope you can see from the example in the previous chapter how analytics
for behaviour-based actionable insights are a continuous loop of an
outcome-first analytics approach leading to a behaviour-analytics-first
approach.

Once a model has been developed with behavioural leading indicators, you
need to perform further steps to deliver the full insights picture. The
journey from Chapter 4 through Chapter 7, and assuming the behavioural
measurement stage has been completed, I can summarize as follows:

1. Define the outcomes: you may have done so already as part of the
‘Define’ stage. Remember the five-step approach for measuring and
driving workplace culture change that I shared with you in Chapter 2?
Defining the outcomes was part of that; however, post behaviour
diagnostics, you can still define further outcomes to help you validate
or add to your insights.

2. Gather outcome data: that is, if you have not done so already. The
process entails making sure that you have clean metric data, as per the
example in Chapter 6. Check that you have up-to-date performance
data – are there any gaps or even possible erroneous data, e.g.
performance data reported that is outside the performance scale? If so,
you can eliminate or correct it.

3. Regression analysis: for each outcome data, perform the regression

analysis so you end up with the nested behaviours that predict your
desired outcomes.

4. Generate benchmarks using quadrant graphs and/or heatmaps:

based on the behaviour cluster, you can plot behavioural heatmaps and
quadrant distribution graphs for People and Place to show exactly
where the behaviours are present or absent based on the
demographics. You can also deliver remarkable insights by generating
 behavioural benchmarks for the outcome data, as I demonstrated using
a line graph in the previous chapter (Figure 7.3). Remember you can
also generate quadrant dispersion graphs for singular behaviours,
giving a further deep layer of actionable insights. Depending on your
audience, you may want to generate other infographics to illustrate the
picture; again, the key is to tell the story of ‘why’ the outcomes
happen, ‘how’, with examples of ‘what’ can we do to sustain or
improve, and ‘where’ in the organization we need to focus on. In
Chapter 11 you will try an easy method to show your audience ‘so
what?’ – what do we get for it; what is the return on our investment?

The process just outlined follows the title of this chapter exactly: steps one,
two and three together provide you with the ‘why’, answering the question,
‘Why does an outcome happen (or not)?’.

Step three provides you with the ‘how’, answering the question, ‘How can
you improve, sustain or suppress an outcome?’ – which behaviours, as the
leading indicators, will maximize your chances for success. Step four will
provide you with the ‘where’, answering the question, ‘Where by
demographics are the behaviour present or absent?’, helping you know
where in the organization to learn from, where to drive change and what
type of intervention is needed, i.e. capability, opportunity, motivation or all
three.

I am indeed hoping that the title of this chapter will stick, as it’s far easier
to remember than a detailed process.

This insight reporting mechanism is your best ally for helping you with the
known unknowns, so they become known knowns. Let me explain. We
know that workplace culture and behaviours impact outcomes – that’s the
known bit – but we don’t know which behaviours and where – that’s the
unknown. The result of this process is turning the unknown to a known.

You may wonder why I am mentioning known unknowns. I am hoping that

this will act as a reminder for you whenever you are examining outcomes,
regardless of how good the outcome metrics are, to stop and think – are
there any known unknowns? This is a critical question if you are to stop
what I call ‘false positives’.
I promised in the previous chapter to elaborate further on behavioural
alignment vs engagement, and I am sure you would have often looked at
employee engagement results and seen great results. However, you also
know that other outcomes are not great. So what’s happening?

The first thing to remember is that employee engagement is another

outcome of workplace culture. Secondly, engagement survey data is
naturally very transient, and can be sensitive to the current situation. In
recent research that I have conducted, I have found that behavioural
alignment between the People and the Place drives a positive engagement
score.

As you may recall from Figure 7.2, the quadrant distribution graph
example, there are two zones where there is behavioural alignment – the
top right corner showing a behavioural presence for both People and Place,
driving meaningful engagement, and the bottom left corner showing a
behavioural absence for both People and Place, driving engagement, but for
the wrong reasons.

To demonstrate the ‘false positives’ issue, I will share with you an example
from some recent research I did. Figure 8.1 shows a quadrant distribution
for large professional services organizations.

FIGURE 8.1 Quadrant distribution graph showing behavioural

alignment vs employee engagement scores

A graph shows the difference between behavioural alignment

and employee engagement scores.

NOTE Sample size: n = 6,560, representing 54 per cent

completion rate of the entire organization

Figure 8.1 details

As before, the plot is for the targeted behaviours for both the People and the
Place, however this time I have added their engagement index score to all
four quadrants of the quadrant distribution graph. Notice the engagement
score for the top right corner where the behaviours are present and aligned.
Now look at the bottom left corner where the behaviours are absent but
aligned. They have very similar engagement scores of 4.8 and 4.1
respectively. If we did not have the behavioural measurement, we would
simply see both zones as showing positive employee engagement, masking
issues and giving us a far rosier image as we would be reporting an
excellent engagement score for 55 per cent of the organization.

In fact, knowing the leading indicators can now explain the ‘why’ for each
zone, ‘how’ to improve and if we add demographics we will get to the
‘where’.

Another important aspect of the known unknowns is that organizational

culture is unique, and you cannot simply borrow models from other
organizations and hope that all will be well. This uniqueness is driven by
your organizational operating model, market conditions, processes and
polices and organizational structure, just to name a few. Hence why I am
normally very sceptical about off-the-shelf workplace culture models that
can be adopted with no applied research of your own to explore how they
will help, and critically what will be the outcomes. What I am really saying
is that your known unknowns may not be the same as other organizations. I
will demonstrate with a quick example, borrowing from Charles Dickens
with a variation – ‘a tale of two organizations’.

If you recall from the example in Chapter 6, the behaviours that will help
you drive performance culture were Achievement, Empower, Consolidate,
Radical and Team Focus.

For another organization in the same sector, and having applied the same
performance culture behavioural research, I have found that the following
behaviours drive their performance as follows: Collective, Relationship,
People Focus and Achievement.

Comparing the two developed behavioural models yields only one common
behaviour. Essentially, these two organizations are competitors, but their
culture for performance is significantly different.

This demonstration is also the reason why people who have performed well
in one organization may struggle in another. I have often come across this
exact question relating to talent: why is this person struggling? They came
to us as a top performer in their previous employment! To answer what is
happening, I will have to take you back to the known unknowns, and there
is no wrong or right answer. You will need to understand your
organizational behavioural model as the leading indicator and how aligned
or misaligned your talent is to the targeted behaviours. Remember that
alignment does not have to be for all behaviours, just the critical few that
drive your outcomes.

To further illustrate with another simple example, imagine you are a New
Yorker, living and working in New York, and you are asked for the first
time to travel and conduct detailed business transactions in Tokyo. As soon
you arrive, it will all feel so different and most likely be very puzzling. You
will need to work so hard to demystify and process what is happening, and
although you are a top performer, suddenly in this very different
environment you are not able to perform! Knowing ‘why’, then the ‘how’,
is the only way forward to impact the future outcome of success.

In Part 2 of this book, I will provide further example case studies for
workplace culture differences and driving similar outcomes between
organizations, so the ‘tale of two organizations’ will continue.

The actionable insights process outlined so far is critical for any workplace
culture initiative, management or change. These initiatives often target an
outcome without knowing ‘why’ the outcomes really happen; so, at best,
any success will be by chance.

Unfortunately, chance does not always work in your favour, and eventually
your luck will run out! For so many organizations I have worked with, I
often come across this syndrome known as ‘get-me-there-itis’ – just get me
there now! We need to drive performance while we drive efficiency and
drive retention, etc. My response is, no! What you need is to drive your
workplace culture in a way that enables and predicts performance, efficacy
and retention. You cannot take shortcuts or dangerously assume the ‘why’.
With digital quantitative tools available to us today like CultureScope,
where behavioural science and people analytics work in harmony, it is far
easier to deliver impactful and actionable insights in a matter of weeks for
any size of organization.
The further good news with this onlined quantitative measurement
approach is the ability to rate and validate. It’s easy to redeploy to make
sure that your behavioural action plan is working, and that the intended
outcome is changing.

So, my last words of wisdom for this chapter are: do not leave workplace
culture change to chance.
9
Leveraging leading indicators
How do you leverage one set of leading indicators to drive
multiple outcomes? Is that really possible?
In today’s fast-paced, competitive and demanding workplace
environment, organizations are constantly seeking ways to
optimize their operations and improve their overall
performance. One effective strategy that has emerged in recent
years is the use of leading indicators to drive multiple
outcomes. As a reminder, leading indicators are defined as
measurable factors that can be used to predict and influence
future performance.
Traditionally, and as I have already demonstrated in previous
chapters, organizations have relied on lagging indicators, which
measure past performance to assess their success. While
lagging indicators provide valuable insights into historical data,
they do not provide proactive information that can drive future
outcomes. Leading indicators, on the other hand, are forward-
looking and can provide organizations with opportunities to
improve performance before problems may occur.
The power of leading indicators lies in their ability to affect
multiple outcomes simultaneously. By identifying and
tracking a set of key leading indicators, organizations can
proactively manage a range of objectives and drive superior
results across various areas of operation. The different
outcomes can be collected together thematically – for example,
risk management, performance, efficiency, diversity, equity and
inclusion, wellbeing, etc. Each theme can, and should, have
multiple outcomes and metrics to manage the risk of overly
depending on singular data points and enhancing validity or
eliminating the inference without sufficient evidence.
Firstly, let us explore some example themes, but without the
enrichment of behavioural data. My main reasoning behind
doing so is to illustrate what adding behavioural data will
actually provide.
This particular organization is a mid-sized nuclear energy generator, with a total sample
size of 681.
Safety and accident prevention
One crucial aspect of many organizations is the safety of their workforce. By identifying
indicators such as near-miss incidents, safety training completion rates or adherence to
safety protocols, organizations can proactively address potential hazards and prevent
accidents. This not only protects the wellbeing of employees, but also reduces costs
associated with workers’ compensation claims and lost productivity.
Quality and customer satisfaction
Another important outcome for organizations is the quality of their products or
services and the level of customer satisfaction. Indicators such as customer feedback
scores, product defect rates or employee training on quality standards can help
organizations identify areas of improvement to enhance product quality and customer
satisfaction. Addressing these indicators can help to increase customer loyalty, improve
brand reputation and ultimately lead to higher revenue.
Employee retention
Organizations that prioritize employee retention enjoy numerous benefits, including
lower turnover costs and less talent/knowledge drain. By tracking indicators like
employee satisfaction surveys, training hours or employee turnover rates,
organizations may identify potential issues or areas for improvement that directly
impact employee retention.
Operational efficiency
Efficient operations are critical for organizations aiming to maximize profitability and
minimize waste. Indicators such as production cycle times, equipment downtime or
waste-reduction initiatives provide insights into the effectiveness and efficiency of
operations. Analysing and actively managing these indicators can help organizations
identify bottlenecks, streamline processes and optimize resource allocation, leading to
increased productivity and reduced costs.
Financial performance
Financial success is a paramount objective for any organization. Indicators such as
revenue growth rates, cost management metrics or investment return ratios can
provide opportunities for improvement in financial performance. By proactively
addressing these indicators, organizations can make data-driven decisions to optimize
revenue generation, reduce costs and improve overall profitability.
I am sure the above cases all sound interesting and valid – but
what is missing? Where is the role of workplace culture in these
outlined metrics in each of the above cases? Armed with the
knowledge that workplace culture translates into behaviours
for Place and People, we simply cannot ignore such metrics for
grassroots leading indicators.
As you have already learnt from the previous chapters, for
each of the above illustrated themes you can run the
behavioural diagnostic and regression analysis for the theme-
relevant outcomes to add leading indicators and then, with the
aid of multiple demographic data and benchmark generation,
you can generate the quadrant distributions to articulate the
‘where’ – i.e., where we can learn from and where we need to
improve.
But do you have to redeploy the behavioural diagnostic
measurement for each theme? The (very) good news is that you
don’t! As long as the behavioural diagnostic includes all 15
dimensions (30 behavioural factors) outlined in Chapter 4, you
can simply measure once and generate multiple and endless
actionable insights. See Figure 9.1 as an example of one
behavioural dataset driving multiple workplace culture themes.
Such a behavioural diagnostic approach to predicting
multiple outcomes is a key differentiator between measuring
behaviours and measuring sentiment, employee management
or even theme-specific employee opinion surveys. Such
methods will result in the need to deploy many different online
questionnaires, driving employee questionnaire fatigue.
 FIGURE 9.1 Example of one set of behavioural data
measurement, a leading indicator to many thematic
outcomes

Figure 9.1 details

Using the above five thematic examples, let’s explore how
leveraging one set of leading indicators can positively impact
multiple outcomes.

Theme 1: Safety and accident prevention

The outcomes that we have metrics for are:
reported near-miss incidents
safety training completion rates
reported safety compliance incidents
Differences in regression models for each of the above
outcomes were tested to identify the most parsimonious model
– i.e. the best number of behaviours required to predict the
outcome. The regression model yielded 0.33 for reported near-
miss incidents, and the predictive behaviours are
Nonconformity, Flexible and Self-Focus. We know the opposite
behaviours will suppress the number of near-miss incidents –
they are Conformity, Standardized and Team Focus.
The regression model yielded 0.41 for the second outcome
and predictive behaviours Active Learning, Standardized and
Conformity.
Finally, the third outcome regression model yielded 0.38 and
the predictive behaviours Neutral, Nonconformity,
Synchronous and Self-Focus. To reduce the number of
compliance incidents, you will need the oppositional
behaviours, which are Expressive, Conformity, Sequential and
Team Focus.
Already from the above you can see the advantages of the one
set of leading indicators, i.e. the behaviours that drive more
than one outcome. If you focus on Conformity, Standardized,
Team Focus, Expressive and Sequential, you will gain all three
outcomes.

Theme 2: Quality and customer satisfaction

For this theme, the outcomes that we had metrics for were
quality related:
improvement projects completion metrics
team design quality metrics
Regression model analyses yielded 0.36 and 0.42, and shared
similar behaviours – Sequential, Team Focus and Active
Learning. Interestingly, we have seen all these behaviours in
the first theme.

Theme 3: Employee retention

Apart from the regression analysis for the employee tenure
outcome, we also looked at the alignment index between Place
and People for the resulting behaviours. Regression model
analysis yielded 0.32 for Collective, Conformity, Team Focus,
Sequential and Active Learning. Apart from Collective, we have
seen all these behaviours in the previous themes.
Looking at the above three examples, you can already see
how a few behaviours are predicting multiple outcomes and
multiple themes. In fact, for the last two themes, which are
Operational efficiency and Financial performance, all the same
behaviours already discovered are predictors with the addition
of Achievement.
Table 9.1 demonstrates the eight predictive behaviours
mapping against the five themes – notice how two critical
behaviours (Team Focus and Sequential) are mapped to all
themes. Sequential is a present behaviour; however, Team
Focus is an absent behaviour, hence intervention design and
implementation will be critical as this one behaviour will move
the dial for all five themes.
 TABLE 9.1 The eight behaviours mapped to all five
themes
Skip table
Safety and Quality and
Accident Customer Employee Ope
Prevention Satisfaction Retention Effi
Sequential
Conformity
Standardized
Active Learning
Collective
Team Focus
Achievement
Expressive

The tick denotes the presence of a behaviour, while the exclamation mark denotes
the absence of a behaviour

So, how have we transformed the leading indicators story? We

went from seemingly unconnected themes looking at outcomes
only to finding out that the key leading indicators are eight
behaviours. And for this organization the story gets better! Five
of the behaviours are present (Sequential, Conformity,
Collective, Active Learning, Standardized). So, they can now
focus on improving the three remaining behaviours
(Expressive, Achievement, Team Focus) for driving 16 outcomes
across five themes.
This demonstrates how managers and leaders don’t have to
be overwhelmed thinking about solutions for multiple
outcomes across multiple themes and will only need to focus on
driving interventions for a few behaviours. This is another
demonstration of the power of organizational behavioural
science as a leading indicator. In fact, if time and resources are
an issue, then the absolute must-focus-on priority in this
example is Team Focus, as this one behaviour will impact all
five themes and will yield significant return on investment.
Apart from having robust and scientifically valid behaviour
measurement, to effectively leverage a single set of leading
indicators, organizations must also establish robust and
relevant outcome measurement metrics. Regular monitoring,
analysis and action planning based on these leading indicators
are also essential to drive the desired outcomes.
In conclusion, leveraging one set of leading indicators can be
a game-changer for organizations looking to drive multiple
outcomes. By identifying key behavioural leading indicators
and actively managing them, organizations can be very
efficient at driving multiple outcomes with reduced effort.
Having demonstrated in this chapter how one set of
behavioural data can drive multiple thematic outcomes, the
question we have not answered is whether there is a sequence
for the outcomes, as some outcomes will need to happen first to
impact other desired outcomes. This is known as path analysis,
and is the subject of my next chapter.
10
Understanding and applying
path analysis as the roadmap for
driving change
Unveiling the connections in complex
systems
As I illustrated the power of behavioural leading indicators in
Chapter 9, it’s time to turn our attention to the relationships
between the outcomes. As you will have seen in Chapter 3,
some outcomes will sequentially predict other outcomes.
In this chapter, I will be getting into some deep maths. This
chapter is not required for all readers as the topic of path
analysis at a high level has been covered in Chapter 3, hence it
is only for those readers who want to gain a deeper
understanding of the maths behind path analysis.
As I explained in Chapter 3, path analysis uses a powerful
statistical technique for investigating complex causal
relationships among variables. It provides a comprehensive
framework to test hypotheses, explore direct and indirect
effects and refine models.
Path analysis is established using a technique called
structural equation modelling (SEM). It is a statistical analytical
approach that allows researchers to analyse complex causal
relationships among multiple variables. Path analysis using
SEM provides a rigorous framework for testing hypotheses and
exploring the interplay between variables in a larger model.
SEM is a multivariate statistical technique that combines
factor analysis and multiple regression analysis to examine the
relationships between measured variables and behaviours. It
enables researchers to investigate both direct and indirect
relationships in a comprehensive way. However, path analysis
goes further, as it helps you interpret the relationships between
variables, providing valuable insights into causal pathways and
mediating effects.
The advantages of path analysis using SEM are numerous,
and can be summarized as follows:
It allows researchers to examine complex relationships
with multiple variables simultaneously, providing a more
comprehensive understanding of the underlying
mechanisms.
It helps to identify direct and indirect effects, thereby
allowing for a better understanding of the contribution of
each variable to the overall outcome.
It enables researchers to test competing hypotheses and
compare alternative models, facilitating development and
refinement.
Furthermore, path analysis using SEM can also control for
measurement error, which enhances the accuracy of the
estimates and reduces bias. It is particularly useful when
dealing with latent constructs that are not directly observed
and the measurement may not be error-free. By accounting for
measurement error, path analysis using SEM provides more
reliable and valid estimates of the relationships between
variables.
 SEM has also emerged as a powerful tool in uncovering the
intricate relationships and connections even within complex
systems. SEM goes beyond traditional linear regression
techniques by considering both observed and unobserved
variables simultaneously. SEM methodology combines elements
of multivariate analysis, regression analysis and factor analysis.
It enables researchers to examine the relationships between
latent variables and observed variables within a framework.
For a definition of regression analysis, please refer back to
Chapter 6. Here I will define multivariate and factor analysis.
Multivariate analysis is a statistical technique that allows
researchers to simultaneously analyse multiple variables to
gain a deeper understanding of the data and extract meaningful
insights. By considering the interconnections among variables,
multivariate analysis enables us to uncover hidden patterns,
identify correlations, make predictions and ultimately make
more informed decisions.
One of the main advantages of multivariate analysis is its
ability to handle complex data sets that involve several
variables. Traditionally, univariate analysis focuses on one
variable at a time, disregarding the potential interactions
among other variables. By contrast, multivariate analysis
considers the relationships among multiple variables
simultaneously, providing a more comprehensive picture.
Multivariate analysis techniques can be broadly classified
into two categories: exploratory analysis and confirmatory
analysis. Exploratory multivariate analysis aims to uncover
patterns and relationships in the dataset without any
preconceived hypotheses. Techniques like principal component
analysis (PCA), factor analysis and cluster analysis fall under
this category. These methods help researchers explore the
underlying structure of the data, reduce its dimensionality and
identify groups or clusters.
Factor analysis is a statistical method used to identify
underlying factors or dimensions in a set of observed variables.
It is commonly used in psychology, sociology, marketing
research and other fields where researchers aim to understand
the latent structure or underlying concepts in a large set of
data. Factor analysis offers several advantages over other data
reduction techniques. It allows researchers to reduce the
dimensionality of a large dataset, making it easier to analyse
and interpret. It also provides insights into the
interrelationships between variables, revealing underlying
constructs that may not be apparent from the raw data.
In SEM, the underlying assumption is that variables can be
decomposed into two main types: endogenous and exogenous
variables. Endogenous variables are predicted by the other
variables in the model, while exogenous variables are
predictors in the model that are not predicted by other
variables. SEM allows the researcher to estimate the strength
and direction of relationships between endogenous and
exogenous variables, ultimately providing a comprehensive
understanding of the system under investigation.
SEM consists of three main components: the measurement
model, the structural model and the identification strategy:
The measurement model focuses on establishing the
relationships between observed variables and their
corresponding latent constructs. Latent constructs are
often variables that affect the observed variables. Through
 the measurement model, researchers can assess how well
the observed variables measure the latent constructs.
The structural model explores the relationships between
the latent constructs. It examines how the latent
constructs influence each other directly or indirectly
through observed variables. These relationships are
represented by path coefficients, which indicate the
magnitude and direction of the influence. By evaluating
the strength of these paths, SEM allows researchers to
understand the causal relationships and interactions
between various constructs.
The identification strategy ensures that the parameters
estimated in the model possess a unique solution. Without
identification, different model specifications could yield
equivalent results, making it impossible to determine the
true relationships between variables. Various
identification strategies, such as setting anchor variables
or using measurement model constraints, can be applied
to achieve a unique solution.
SEM is widely used in behaviour and social sciences to study
complex constructs such as behavioural traits. Researchers can
evaluate the relationships between latent factors and observed
variables to gain insights into the underlying dynamics of these
constructs.
The advantages of using SEM are:
Simultaneous analysis: One of the key advantages of SEM
is its ability to analyse multiple relationships
simultaneously. By considering multiple constructs and
their interconnections, SEM offers a holistic
understanding of complex systems.
 Model fit assessment: SEM provides explicit measures for
evaluating the fit between the hypothesized model and the
observed data. This allows researchers to determine the
effectiveness of their models and make necessary
adjustments, ensuring the validity and reliability of their
findings.
Mediation and moderation analysis: SEM allows
researchers to evaluate mediation and moderation effects.
Mediation analysis assesses the mechanisms through
which one variable influences another, while moderation
analysis examines how the relationship between two
variables varies depending on a third variable. These
analyses help unravel the underlying processes and
provide a deeper understanding of complex systems.
One of the most significant issues with SEM is model
misspecification. Hence the plethora of case studies in Part 2 of
this book aiming at gaining experience in how to interpret it,
particularly in the workplace culture context.

Steps for developing a path analysis using SEM

modelling
To gain further practical understanding, let us explore a
workplace culture example. This is a large financial institution
interested in knowing how to drive proactive risk management
and six key organizational culture outcomes, as follows:
Living our Values
Recognition
Purpose
Resilience
 Psychological Safety
Development for the Future
The outcomes for proactive risk management were:
Taking Ownership
Collaboration
Risk Anticipation
Risk Avoidance
Proactive Learning
The organization had a recent outcome metric measure for all
the outcomes, and had launched behavioural measurement for
circa 10,500 with a response rate of 4,622, representing a 44 per
cent completion rate.
As described in Chapter 6, the methodology for predictive
analytics was applied to each of the targeted themes using
multiple regression analysis to find the most parsimonious
model. Each theme ended up with a list of predictive
behaviours. I will exhibit those behaviours later in this chapter;
for now, let us focus on the new learning – that is, path analysis
using SEM.

Step 1
Build a correlation coefficient table for all the thematic
outcomes as per Table 10.1. This will show the relationship (if
any) between all the outcomes. You will see the outcome titles
and correlation value against other outcomes. The outcomes
are not arranged in any order, and at this stage we are not
building any hypothesis, we are just exploring the relationships
between the outcomes (if any).
 TABLE 10.1 Correlation analysis for all provided
outcomes
Skip table
Management
Collaboration Role? Living our Values Recog
Collaboration 1 –0.036 0.177 0.232
Management –0.036 1 0.005 –0.011
Role?
Living our Values 0.177 0.005 1 0.317
Recognition 0.232 –0.011 0.317 1
Purpose 0.189 0.137 0.145 0.167
Resilience 0.186 0.104 0.256 0.319
Psychological 0.225 0.041 0.325 0.419
Safety
Development 0.14 0.095 0.126 0.193
Risk Anticipation 0.343 0.003 0.204 0.211
Risk Avoidance 0.339 0.029 0.181 0.236
Proactive 0.325 –0.06 0.152 0.232
Learning
Taking Ownership 0.409 –0.021 0.175 0.244

As a reminder, the correlation coefficient scale is from –1 to +1.

A correlation coefficient greater than zero indicates a positive
relationship, while a value less than zero signifies a negative
relationship. A value close to zero indicates a weak relationship
between the two variables being compared. A correlation
coefficient of 0.2 and above will be interesting for our purposes.
From the generated table, we are interested in a correlation
of 0.2 and above (refer back to Chapter 6 re correlational effect
size). You will also notice that the second metric, Management
Role?, is irrelevant as it has no correlation to any other outcome
– the largest correlation is 0.137 to the Purpose outcome.
Step 2
Calculate the mean correlation for each outcome. Based on the
mean correlation value, we can assign an impact rank value as
shown in Table 10.2.
 TABLE 10.2 Correlation analysis with the mean for
each outcome calculated and ranked
Skip table

Management
Collaboration Role? Living our Values Recog
Collaboration –0.036 0.177 0.232
Management –0.036 0.005 –0.011
Role?
Living our Values 0.177 0.005 0.317
Recognition 0.232 –0.011 0.317
Purpose 0.189 0.137 0.145 0.167
Resilience 0.186 0.104 0.256 0.319
Psychological 0.225 0.041 0.325 0.419
Safety
Development 0.14 0.095 0.126 0.193
Risk Anticipation 0.343 0.003 0.204 0.211
Risk Avoidance 0.339 0.029 0.181 0.236
Proactive 0.325 –0.06 0.152 0.232
Learning
Taking Ownership 0.409 –0.021 0.175 0.244
Mean 0.230 0.026 0.188 0.233
Rank 6 12 9 4

Step 3
Using the Table 10.2 correlation analysis, remove any
correlation values below 0.2 and recalculate the mean as per
Table 10.3. Essentially, this step will reduce the data noise and
you can now start developing the real connections. The cell
shading is applied to help you see the most and least
correlations.
 TABLE 10.3 Correlation analysis with all values less
than 0.2 eliminated
Skip table
Management
Collaboration Role? Living our Values Recog
Collaboration 0.232
Management
Role?
Living our Values 0.317
Recognition 0.232 0.317
Purpose
Resilience 0.256 0.319
Psychological 0.225 0.325 0.419
Safety
Development
Risk Anticipation 0.343 0.204 0.211
Risk Avoidance 0.339 0.236
Proactive 0.325 0.232
Learning
Taking Ownership 0.409 0.244
Mean 0.312 0.276 0.276
Rank 3 8 7

Step 4
In each column, we can remove all values that are less than the
column’s mean. For example, for the Collaboration column, we
can see that the correlation of 0.232 with Recognition and 0.225
with Psychological Safety are both less than 0.312 (the average
correlation for Collaboration). Then by removing these values
we can see clearly which criteria are most correlated to
Collaboration. We repeat this process for each column.
 Consequently, we obtain Table 10.4. Calculate the resulting
mean and rank for each column.
 TABLE 10.4 Correlation analysis with values less than
the column mean removed
Skip table
Management
Collaboration Role? Living our Values Recog
Collaboration
Management
Role?
Living our Values 0.317
Recognition 0.317
Purpose
Resilience 0.319
Psychological 0.325 0.419
Safety
Development
Risk Anticipation 0.343
Risk Avoidance 0.339
Proactive 0.325
Learning
Taking Ownership 0.409
Mean 0.354 0.321 0.352
Rank 6 – 8 7

Step 5
To find our path starting point, we need to compute the average
mean correlations and rankings from the three matrices of
steps 2, 3 and 4 as per Table 10.5.
 TABLE 10.5 The computed average mean correlations
and rankings
Skip table
Manag
Collaboration Role?
Step 2 Mean 0.230 0.026
Rank 6 12
Step 3 Mean 0.312
Rank 3
Step 4 Mean 0.354
Rank 6 –
Avg of mean 0.299 0.026
correlations
Avg rank 5.00 12.00

The column with the resulting highest mean correlation

coefficient which will have also have the closest value to 1 in
our resulting mean ranking will be the starting point. In this
example it’s the Taking Ownership outcome. Our first key
practical outcome insight: we must get this outcome to manifest
before we progress.

Step 6
We can now continue to build our path. To find the next
outcome, we can use Table 10.4 from step 4 and look to see the
highest corresponding correlation coefficient to another
outcome. See the resulting Table 10.6 – the next outcome to
manifest in our path is Risk Anticipation.
We can start drawing our resulting diagrammatic path – see
Figure 10.1.
TABLE 10.6 The path starting point outcome that has
the highest corresponding correlation coefficient to
another outcome
 FIGURE 10.1 The first two outcomes in the risk path
analysis

To continue with our path, we can remove what we have

already used and look at the highest corresponding correlation
coefficient to Risk Anticipation. See Table 10.7 showing that Risk
Avoidance is our next outcome in the path analysis.
 TABLE 10.7 Risk Avoidance is our next outcome in the
path analysis as it has the highest correlation
coefficient with Risk Anticipation

We can now also continue drawing the path – see Figure 10.2.
 FIGURE 10.2 The sequence of how the three outcomes
should manifest in the risk path analysis

For our next outcome, we will need to look at the highest

corresponding correlation coefficient to the Risk Avoidance
outcome, and as before we can remove what we have already
used. Table 10.8 shows that the next outcome is Collaboration.
 TABLE 10.8 The highest corresponding correlation
coefficient outcome to Risk Avoidance is
Collaboration

The resulting diagrammatic path can be seen in Figure 10.3.

 FIGURE 10.3 The sequence of how the four outcomes
should manifest in the risk path analysis

Let us look at the next outcome in our risk path. Having

removed what we have already used in our path, let us
examine the Collaboration column. The next outcome with
highest correlation coefficient is Proactive Learning, as shown
in Table 10.9.
 TABLE 10.9 The highest corresponding correlation
coefficient outcome to Collaboration is Proactive
Learning

The resulting path analysis diagram can be seen in Figure 10.4.

 FIGURE 10.4 The sequence of how the five outcomes
should manifest in the risk path analysis

All good so far – we can now once again remove what we have
used so far and move to the Proactive Learning outcome
column to see which is the corresponding outcome with the
highest correlation coefficient. As you will see from Table 10.10,
we have come to a dead end! But we do have more outcomes
with a usable correlation coefficient. What is happening?
 TABLE 10.10 Our first path has come to a dead end

This means that our outcomes correlation matrix has more

than one path. It is interesting to note that any outcomes
regarding risk have all attributed to the risk management path,
and we have come to this conclusion without any assumptions,
illustrating the power of SEM for path analysis.
So, what have we learnt about our risk path for this
institution? Examining Figure 10.4 tells us that the ability to
take ownership is key as it’s the starting point. The organization
aces this outcome, which will positively impact their ability to
anticipate risk issues before unforeseen events happen. They
can now continue to improve as they drive down the path. This
path is critical as each outcome metric will inform an
organization where they are down the path, and of course
which behaviours they need to focus on to improve.
Furthermore, and as an illustrative example of the power of
path analysis insights, there is no point in focusing on the
Proactive Learning for Risk outcome if they have not secured
the Taking Ownership of Risk outcome.
 As I mentioned earlier, I will come back to Figure 10.4 and
add the behaviours. For now, let’s develop our second path.

Step 7
To discover our second path, we will need to examine Table
10.5 from step 5 and remove the outcomes that formed our first
path related to risk management. See resulting Table 10.11.
We will now need to find the highest remaining mean
correlation coefficient (the result established from steps 2, 3
and 4); that should correspond to the average ranking nearest
to 1 from the remaining outcomes. Your choice should fall on
Psychological Safety as the starting point for the second path.
 TABLE 10.11 The computed average mean
correlations and rankings, now with the first path
outcomes removed

Back to the same process again. Look at the Psychological Safety

outcome column and find out the corresponding outcome with
the highest correlation coefficient, which is Recognition, as can
be seen in Table 10.12.
 TABLE 10.12 The highest corresponding correlation
coefficient outcome to Psychological Safety is
Recognition

We can also start drawing our second path, as seen in Figure

10.5.
 FIGURE 10.5 The outcome sequence for the second
developing path

Moving our focus to the Recognition outcome column, look at

Table 10.13 and you will see that after removing what we have
used already, the highest corresponding correlation coefficient
outcome is Resilience.
 TABLE 10.13 The highest corresponding correlation
coefficient outcome to Recognition is Resilience

So we can continue drawing our path – see Figure 10.6.

 FIGURE 10.6 The sequence of the second developing
path for the outcomes

Next in line is the Resilience outcome column, and as you can

see from Table 10.14, we have used both possible corresponding
outcomes. However, step back to the Recognition outcome
column and we will see that we still have one additional
outcome with a very good corresponding correlation
coefficient; that is the Living our Values outcome as circled in
Table 10.14.
This means that Recognition drives two outcomes, as
demonstrated in Figure 10.7.
TABLE 10.14 The highest corresponding correlation
coefficient outcomes have both been already used in
the path

Notice the split paths from Recognition to another outcome, Living our Values
 FIGURE 10.7 The outcome sequence for the second
full path

If we remove Recognition and Living our Values, we have now

reached a dead end, but we still have two outcomes with a good
correlation coefficient. It seems we have a third path to
develop.

Step 8
I am sure by now you have got this process all sorted! So back
to Table 10.5 from step 5 and remove the outcomes that formed
our first and second paths. See resulting Table 10.15.
As before, we will now need to find the highest remaining
mean correlation coefficient (the result established from steps
2, 3 and 4), which should correspond to the average ranking
nearest to 1 from the remaining outcomes. Your choice should
fall on Purpose as the starting point for the third path as the
mean correlation coefficient is higher than for the Development
outcome, although they both share a similar ranking.
 TABLE 10.15 The computed average mean
correlations and rankings, now with the first and
second path outcomes removed

Since we only have two outcomes, the third path is easy to draw
without any further steps – see Figure 10.8.
 FIGURE 10.8 The outcome sequence for the third path

I am sure by now you may feel that we have been on a

marathon with SEM and path analysis, but I hope you can see
its importance. It can significantly inform which interventions
to invest in as outcomes will drive other outcomes, and it
should stop erratic organization interventions that will not
yield any benefit and may even fail.
You may have also noticed from the developed paths that for
this institution, their risk management and workplace culture
are separate paths. This did myth-bust for them, as leaders
assumed ‘if we live our culture, we will manage our risk’. Hmm
– no! These are separate paths you need to monitor and drive.
What about the behaviours that predict this path? As I did
promise, I will now add behaviours to our three paths.
Remember the behaviours were established using the same
process outlined in Chapter 6.
Path one behaviours are Team Focus, Strategic, Innovate,
Outer Focus and Conformity. These behaviours are critical for
the starting point, and if absent for People or Place (or both),
they must be the focus for intervention design and
implementation. These behaviours will take you down the path
to Collaborate for Risk, which will need another behaviour,
Collective, and to achieve the final outcome, Proactive Learning
for Risk, the behaviour People Focus needs to be present.
Figure 10.9 illustrates the first path with the behaviours now
included.
 FIGURE 10.9 The first path with the behaviours now
included

Figure 10.9 details

Consider the first path, which is the risk management one. Here
is a shock for you: Innovate is one of the predictor behaviours.
You may have never thought so – again, the dangers of
assumptions and hypotheses with no evidence. For this
organization, Conformity, Strategic and Team Focus are present
behaviours. Innovate is absent for People and Place, needing
interventions to address capability, opportunity and motivation.
Outer Focus is absent at the Place, needing opportunity and
motivation-type interventions. Looking further into the path,
Collective is present; however, People Focus is absent for the
People, hence a capability-type intervention is required.
So, the message to this organization has great clarity and
simplicity now. Don’t chase or even invest in driving all your
outcomes in Figure 10.9. You need to achieve your starting point
first, as by improving the two behaviours, you gain and
improve all your sequential outcomes till you arrive at the final
outcome and another behaviour needs improvement. Before
this actionable insight, the organization was simply trying to
address each outcome on its own without the knowledge of the
behavioural leading indicators.
Let’s look at the behaviours for the second path, Figure 10.10
reveals that four behaviours are driving the starting point. It
shares one behaviour with the first path, that is Team Focus,
which as you know already is a present behaviour; however,
the other three behaviours all need attention. Empower is
absent at the Place, and Expressive and People Focus are both
absent for the People. Ace the starting point behaviours and
you’ve got the entire path in sequence, as Active Learning and
Conformity are both present.
It’s fascinating to look closely at this path, to see how
important Psychological Safety is as an outcome. Without
achieving that goal, Recognition and Resilience will simply not
happen and walking the talk by Living our Values will be
impossible.
 FIGURE 10.10 The second path with the behaviours
now included

Figure 10.10 details

Interestingly, this organization thought that Living our Values
would be the starting point – another reason not to assume, as
workplace culture is so unique! Basically, if you don’t have
Psychological Safety and Recognition, you will not get to Living
our Values.
The third path behaviours are illustrated in Figure 10.11. We
have seen all these behaviours already as they are shared with
the two previous paths. Innovate is absent for Place and People,
and Outer Focus is absent at the Place. It’s interesting to note
that Innovate is part of living their purpose as an organization,
and there is no point in future employee development if you do
not live your purpose first!
 FIGURE 10.11 The third path with the behaviours
now included

Figure 10.11 details

For the final part of this chapter, we can now analyse how the
behavioural leading indicators will help in shifting the dial for
multiple outcomes. Ten behaviours in total drive the three
paths; five are present and five need intervention design,
planning and implementation. Figure 10.12 illustrates the
return on investment for this organization as they focus on the
five behaviours for improvement.
 FIGURE 10.12 The behavioural leading indicators and
the outcomes they drive over the three paths

Figure 10.12 details

Although Figure 10.12 shows you the number of outcome gains
by category, it does not show you by how much. Interesting
questions for each outcome would be ‘What is the likelihood for
success?’ and ‘What is the likely return on investment?’ –
questions that will be answered in the next chapter for this
exact example.
11
Simplifying the actionable
insights and understanding ROI
In today’s data-driven organizational environment, showcasing
the return on investment (ROI) of a project or initiative has
become essential. Executives and stakeholders need concrete
evidence that their investment will yield positive outcomes.
Statistics offer a reliable and tangible means of demonstrating
ROI. In previous chapters I have already built the case for
leading indicators and predictive analytics, but as you may
remember from Chapter 6, reading and interpreting regression
analysis can be complex and may take away from the insight’s
gravitas.
You will already know that to build predictive analytics, we
need leading and lagging indicators. In the context of
workplace culture, leading indicators are the valid and
scientific behaviour measurement of the People and the Place.
Lagging indicators include any outcome, sentiment or opinion
data, and we also discussed that per workplace culture theme,
it’s important to examine multiple lagging indicators to validate
the predictive analytics process.
Multiple regression analysis provides us with interesting and
actionable insights as it produces a restricted set of nested
behaviours that drive multiple outcomes across multiple
themes. Understanding the regression analysis entails looking
at multiple variables to articulate how good the resulting model
is. As a reminder, that included the following:
Multiple R: the correlation coefficient that measures the
strength of a linear relationship between two variables.
The correlation coefficient can be any value between –1
and 1, and its absolute value indicates the relationship
strength. The larger the absolute value, the stronger the
relationship:
1 means a strong positive relationship.
–1 means a strong negative relationship.
0 means no relationship at all.
Adjusted R square: the R square adjusted for the number
of independent variables in the model. You will want to
use this value instead of R square for multiple regression
analysis.
Standard error: another goodness-of-fit measure that
shows the precision of your regression analysis – the
smaller the number, the more certain you can be about
your regression equation. While R square represents the
percentage of the dependent variables variance that is
explained by the model, standard error is an absolute
measure that shows the average distance that the data
points fall from the regression line.
Significance F: the P-value of F. The lower this value, the
higher the significance of this model.
Assessing model fit (residual): residual plots help
determine whether the linear regression model is an
appropriate fit for the data. If the residuals are randomly
 scattered around the horizontal axis, it indicates that the
model is a good fit.
Model coefficients: these enable you to build a
prediction; basically, how quickly the outcome will
improve by changing the input, also known as the beta (β)
coefficient. The beta describes the slope of the line for the
outcome prediction.
As you can see, it can be a complex process to explain how good
the prediction is and what is the likely ROI.
Since I have just mentioned the word ‘likely’ in the context of
ROI, I had better explain! In regression analysis, we are trying
to articulate the nature of the causative relationship between
the leading and lagging indicators, if indeed the causative
relationship exists; hence, what is the likely magnitude of gain
from changing the leading indicator? In actual statistical terms
we cannot surmise the exact gain, only the likelihood of
increase or decrease in the lagging indicators.
So, is there a way to show the predictive power without
having to explain to the audience all the regression analysis
variables? Is there a way to bring the story to life in a simple
diagram or plot? The good news is that you can, with something
called the ‘odds ratio’. This process follows regression analysis
to produce the restricted nested behaviours.

Odds ratio
The odds ratio is a statistical measure used to determine the
association between two events or variables. It is commonly
used in medical and social sciences research to assess the
strength and direction of an association between exposure to a
risk factor and the likelihood of developing a disease or
experiencing an outcome.
The odds ratio is calculated by comparing the odds of an
event occurring in one group to the odds of the same event
occurring in another group. It is commonly denoted as OR and
is expressed as a ratio of two odds.
To illustrate how the odds ratio works, let’s consider an
example. Suppose a study aims to investigate the association
between smoking and lung cancer. The researchers divide
participants into two groups: smokers and non-smokers. They
then calculate the odds of developing lung cancer in each
group.
Let’s say the odds of developing lung cancer in the smoker
group are 0.3 (meaning that for every one case of lung cancer,
there are three non-cases), and the odds in the non-smoker
group are 0.1 (meaning for every one case of lung cancer, there
are nine non-cases). The odds ratio would be (0.3/0.1) = 3.
The interpretation of the odds ratio depends on its value. If
the odds ratio is 1, it suggests no association between the
exposure and the outcome. If the odds ratio is greater than 1, it
indicates a positive association, meaning the exposure
increases the likelihood of the outcome. In our example, an
odds ratio of 3 implies that smokers are three times more likely
to develop lung cancer than non-smokers.
On the other hand, if the odds ratio is less than 1, it suggests a
negative or protective association. For instance, an odds ratio of
0.5 would mean that the exposure decreases the likelihood of
the outcome by half.
It is important to note that the odds ratio only provides
information on the association between the exposure and
outcome and does not establish causality. It’s the regression
analysis that provides the information that we need about
which behaviours are the leading indicators and what are the
outcomes that they influence. This is why the odds ratio in our
context will be calculated after the regression analysis.
Specifically, for our purpose, odds ratio is an easy way to
represent the likelihood of an outcome based on the absence
and/or presence of a nested behaviour model to show how good
the predictive model is.
To apply the odds ratio method, we will arrange the
behaviours into four quartiles, and count the number of
outcome occurrences per nested model quartile (remember, the
nested behavioural model is the result from the regression
analysis). On the behavioural scale of zero to five, the quartiles
are defined as:
0 to 1.25 (less than or equal to)
1.25 to 2.5
2.5 to 3.75
3.75 to 5
Let us look at the step-by-step approach to developing an odds
ratio for a specific outcome. In this illustrative example, the
context is a small engineering organization with outcome data
for classroom-type training and learning that includes
simulation for incident management. Having deployed the
CultureScope online diagnostic for the full saturated 30 factors
(15 dimensions) behavioural measurement, and conducted the
hierarchical regression analysis, a restricted behavioural nested
model was developed that includes two behaviours, Active
Learning and People Focus. The model yielded a regression
value of R=0.51 and total sample size of n=94: nothing new here,
just explaining the context.
To generate the odds ratio you will need to follow these steps.

Step 1
Develop a table with the number of occurrences per the nested
behavioural quartile (i.e. how people achieving the desired
outcome are in the 4th quartile, that is 3.75 to 5; how many in
the 3rd quartile, 2.5 to 3.75, etc.). See resulting Table 11.1.
 TABLE 11.1 The number of best and least outcome vs
behavioural quartiles
Skip table
Occurrences

Quartiles Least Outcome Best Outcome Totals

1st Quartile 0 – 1.25 4 2 6
2nd Quartile 1.25 – 2.5 5 14 19
3rd Quartile 2.5 – 3.75 2 29 31
4th Quartile 3.75 – 5 2 36 38
Totals 13 81 94

The table shows the behavioural quartiles, as well as the

number of occurrences for each quartile vs the best and least
desired outcome (this outcome can be a metric or even
qualitative). You can then divide the number of occurrences of
the highest number by the lowest number, providing you with
the odds ratio per quartile. For the first behavioural quartile,
the least outcome column occurrence is higher than the best
outcome occurrence, hence the odds ratio will be 4 divided by 2
= 2. However, you should designate a minus to this number as
the best outcome value is less than the least outcome. Insert the
value in the ‘Odds against’ column.
You can now repeat for each behavioural quartile, applying
the same logic. As another working example for the 2nd
behavioural quartile, the best outcome has more occurrence
than the least outcome, so divide 14 by 5 = 2.8. Since more
occurrences are for the best outcome, insert the value in the
‘Odds for’ column.

Step 2
You are now ready to draw the diagram to reflect Table 11.1, as
per Figure 11.1 (odds ratio figures have been rounded for
simplicity). We can now shift the story away from describing
the predictive analytics insights using six variables to one
compelling simple story. Where the behaviours’ presence is in
the upper fourth quartile, you are 18 times more likely to get
the desired outcome, and where the behaviours’ presence is in
the first quartile you are two times less likely to achieve the
outcome. Now, that is far simpler language to answer the
question: ‘Why should anyone invest in the two behaviours?’.
 FIGURE 11.1 The odds ratio per behavioural quartile
vs likelihood of outcome occurrence

NOTE n=94, R=0.51. Behavioural factors are Active Learning and People Focus

Figure 11.1 details

To put this learning into practice, we can now apply an odds
ratio to all the outcomes for the financial institution case study
from the previous chapter. As a reminder, the behavioural
measurement using CultureScope was deployed for around
10,500 employees, with a response rate of 4,622 representing a
44 per cent completion rate.
Let’s now explore each outcomes odds ratio, and then we can
look at the entire system.
LIVING OUR VALUES
Regression analysis for the saturated model led to the nested behaviours of Team
Focus, People Focus, Conformity and Expressive with R=0.314 for this model. Figure
11.2 shows the odds ratio, and the interpretation is easy: this organization is almost
120 times more likely to live its values where these four behaviours are in the top
quartile of their presence, and only twice as likely to live its values if these behaviours
are in the first quartile of their presence.
 FIGURE 11.2 The odds ratio for Living our Values
outcome vs behaviour presence

NOTE Probability of behaviours’ presence or absence. Behaviours are Team

Focus, People Focus, Conformity and Expressive

Figure 11.2 details

RECOGNITION
Regression analysis for the saturated model led to the nested behaviours of Team
Focus, People Focus, Conformity and Active Learning with R=0.436 for this model.
Figure 11.3 shows the odds ratio.
This organization is almost 190 times more likely to drive recognition where these
four behaviours are in the top quartile of their presence, and three times less likely to
drive recognition if these behaviours are in the first quartile of their presence.
 FIGURE 11.3 The odds ratio for the Recognition
outcome vs behaviour presence

NOTE Probability of behaviours’ presence or absence. Behaviours are Team

Focus, People Focus, Conformity and Active Learning

Figure 11.3 details

PURPOSE
Regression analysis for the saturated model led to the nested behaviours of Team
Focus, Outer Focus, Conformity, Innovate and Active Learning with R=0.302 for this
model. Figure 11.4 displays the odds ratio.
This organization is almost 52 times more likely to live its purpose where these five
behaviours are in the top quartile of their presence, and two times less likely to live its
purpose if these behaviours are in the first quartile of their presence.
 FIGURE 11.4 The odds ratio for the Purpose outcome
vs behaviour presence

NOTE Probability of behaviours’ presence or absence. Behaviours are Team

Focus, Outer Focus, Conformity, Innovate and Active Learning

Figure 11.4 details

RESILIENCE
Regression analysis for the saturated model led to the nested behaviours of Team
Focus, Empower, Conformity, People Focus and Active Learning with R=0.338 for this
model. Figure 11.5 shows the odds ratio.
This organization is almost 53 times more likely to be resilient where these five
behaviours are in the top quartile of their presence, and two times less likely to be
resilient if these behaviours are in the first quartile of their presence.
FIGURE 11.5 The odds ratio for the Resilience
outcome vs behaviour presence

NOTE Probability of behaviours’ presence or absence. Behaviours are Team

Focus, Outer Focus, Conformity, Innovate and Active Learning

Figure 11.5 details

PSYCHOLOGICAL SAFETY
Regression analysis for the saturated model led to the nested behaviours of Team
Focus, Empower, Expressive and People Focus with R=0.433 for this model. Figure 11.6
demonstrates the odds ratio.
This organization is almost 195 times more likely to drive psychological safety where
these four behaviours are in the top quartile of their presence, and only one and a half
times as likely to drive psychological safety if these behaviours are in the first quartile
of their presence.
FIGURE 11.6 The odds ratio for the Psychological
Safety outcome vs behaviour presence

NOTE Probability of behaviours’ presence or absence. Behaviours are Team

Focus, Empower, Expressive and People Focus

Figure 11.6 details

DEVELOPMENT FOR THE FUTURE
Regression analysis for the saturated model led to the nested behaviours of Active
Learning and Team Focus with R=0.300 for this model. Figure 11.7 conveys the odds
ratio.
This organization is 16 times more likely to drive talent development where these
two behaviours are in the top quartile of their presence, and is only one time as likely to
drive talent development if these behaviours are in the first quartile of their presence.
FIGURE 11.7 The odds ratio for the Development for
the Future outcome vs behaviour presence

NOTE Probability of behaviours’ presence or absence. Behaviours are Active

Learning and Team Focus

Figure 11.7 details

TAKING OWNERSHIP OF RISK
Regression analysis for the saturated model led to the nested behaviours of Team
Focus, Strategic, Outer Focus, Innovate and Conformity with R=0.371 for this model.
Figure 11.8 shows the odds ratio.
This organization is almost 14 times more likely to drive taking ownership of risk
where these five behaviours are in the top quartile of their presence, and two times less
likely to drive taking ownership of risk if these behaviours are in the first quartile of
their presence.
 FIGURE 11.8 The odds ratio for driving the Taking
Ownership of Risk outcome vs behaviour presence
A bar graph shows the difference in the odds ratio between ownership of risk
outcomes and behaviour presence.

NOTE Probability of behaviours’ presence or absence. Behaviours are Team

Focus, Strategic, Outer Focus, Innovate and Conformity

Figure 11.8 details

COLLABORATE FOR MANAGING RISK

Regression analysis for the saturated model led to the nested behaviours of Team
Focus, Strategic, Outer Focus, Innovate and Collective with R=0.402 for this model.
Figure 11.9 indicates the odds ratio.
This organization is almost nine times more likely to drive collaborative risk
management where these five behaviours are in the top quartile of their presence, and
three times less likely to drive collaborative risk management if these behaviours are in
the first quartile of their presence.
 FIGURE 11.9 The odds ratio for driving the Collaborate
for Managing Risk outcome vs behaviour presence
A bar graph shows the difference in the odds ratio between managing risk
outcomes and behaviour presence.
NOTE Probability of behaviours’ presence or absence. Behaviours are Team
Focus, Strategic, Outer Focus, Innovate and Collective

Figure 11.9 details

RISK ANTICIPATION
Regression analysis for the saturated model led to the nested behaviours of Team
Focus, Strategic, Outer Focus, Innovate and Conformity with R=0.347 for this model.
Figure 11.10 shows the odds ratio.
This organization is almost 18 times more likely to drive the anticipation of risk
where these five behaviours are in the top quartile of their presence, and two times less
likely to drive the anticipation of risk if these behaviours are in the first quartile of their
presence.
 FIGURE 11.10 The odds ratio for driving the Risk
Anticipation outcome vs behaviour presence
A bar graph shows the difference in the odds ratio between risk anticipation
outcomes and behaviour presence.

NOTE Probability of behaviours’ presence or absence. Behaviours are Team

Focus, Strategic, Outer Focus, Innovate and Conformity

Figure 11.10 details

RISK AVOIDANCE
Regression analysis for the saturated model led to the nested behaviours of Team
Focus, Strategic, Outer Focus, Innovate and Conformity with R=0.331 for this model.
Figure 11.11 displays the odds ratio.
This organization is 131 times more likely to drive risk avoidance where these five
behaviours are in the top quartile of their presence, and two times less likely to drive
risk avoidance if these behaviours are in the first quartile of their presence.
 FIGURE 11.11 The odds ratio for driving the Risk
Avoidance outcome vs behaviour presence
A bar graph shows the difference in the odds ratio between risk avoidance
outcomes and behaviour presence.

NOTE Probability of behaviours’ presence or absence. Behaviours are Team

Focus, Strategic, Outer Focus, Innovate and Conformity

Figure 11.11 details

PROACTIVE LEARNING FOR RISK

Regression analysis for the saturated model led to the nested behaviours of Team
Focus, Strategic, People Focus, Innovate and Collective with R=0.414 for this model.
Figure 11.12 shows the odds ratio.
This organization is almost four times more likely to drive proactive learning for risk
where these five behaviours are in the top quartile of their presence, and almost five
times less likely to drive proactive learning for risk if these behaviours are in the first
quartile of their presence.
 FIGURE 11.12 The odds ratio for driving Proactive
Learning for Risk outcome vs behaviour presence
A bar graph shows the difference in the odds ratio between risk outcomes for
increasing proactive learning and behaviour presence.

NOTE Probability of behaviours’ presence or absence. Behaviours are Team

Focus, Strategic, People Focus, Innovate and Collective

Figure 11.12 details

We have just been on a small marathon translating all the

outcome predictive models into easier language using the odds
ratio method. That is all well and good for each individual
outcome, but can we show that insight as a whole system? Yes,
we can; and as I have already illustrated in Figure 10.12, we can
simply add the odds ratio to create Figure 11.13.
 FIGURE 11.13 The behavioural leading indicators and
the outcomes they drive as well as the likelihood of
outcome manifesting over the three paths
An illustration shows the process of behavioural
indicators and the outcomes in the three paths that increase
the likelihood of outcome manifestation.

Figure 11.13 details

*Likelihood of this outcome manifesting from a behavioural intervention (i.e. 23 × more
likely to manifest if you have the presence of behaviours and other outcomes)

The detailed workings in this and Chapter 10 are so well

summarized by Figure 11.13, and the key message is that you
will get all your desired outcomes if you focus interventions on
the five absent behaviours and look at that likelihood of
success!
Demonstrating possible ROI using odds ratio statistics is
crucial for stakeholders to assess the success and value of their
investment. With the outlined method, you can effectively
communicate the impact.
Additionally, using comparative analysis and addressing
limitations and risks enhances credibility when you present the
likelihood of ROI, by showing: What will happen if we do
nothing? How absent are these behaviours currently? And so
on.
12
Cultural types
Difficult as it may be, we often find people trying to put labels
on organizations. The label acts as a description of that culture;
essentially it is typifying or characterizing the organization.
Why would anyone find it useful to do that? Does it connect to
the organization’s aspired-to values? Does it help to
communicate change from one workplace culture type to
another?
Indeed, these are all valid questions and useful to answer,
particularly after having gone through the process of predictive
behaviour analytics. Workplace culture types will simply be an
additional lens.
As I have already discussed in Chapter 4, the seven-year
extensive research study I conducted provided the evidence
and clarity that the 15 behavioural dimensions (30 behavioural
factors in dichotomy – as outlined in Table 4.2) are mutually
exclusive and collectively exhaustive, in a behavioural
approach for measuring organizational culture. However, I
have illustrated in previous chapters that we can create
behavioural nested clusters as a behavioural map for
organizational values, or how these clusters are developed as a
result of predictive analytics.
Therefore, can we now create clusters of behaviours to
represent workplace culture types? Yes, we can! And once we
have done so, you can map the actual diagnostic results to the
culture type clusters to typify your current versus target
culture.
There are two layers to the culture types: primary and
secondary.

Culture type – primary behavioural clusters

Figure 12.1 shows the behavioural clusters, five behaviours to a
cluster, and the culture type is tagged for each generated
cluster. The behaviours are organized in the outermost layer;
the first inner layer has the culture type label.
 FIGURE 12.1 The behaviours and related culture type
for a behavioural cluster

Figure 12.1 details

In the primary culture types, we have generated six clusters as
follows:
1. Partner type: in this culture it’s all about the people, the
teams, relationships and external alliances.
 2. Envision type: in this culture, it’s all about driving
innovation, being pioneering and working at a fast pace
with competing properties and some errors along the way.
3. Explore type: in this culture, it’s about providing
empowerment and flexibility so the environment can
drive exploration without boundaries.
4. Own type: in this culture, it’s about owning it and making
it happen. The focus is individual accountability and
performance.
5. Apply type: in this culture, it’s about applying what we
have and know with no space or time for errors.
6. Verify type: in this culture, it’s about command and
control, following the rules, standards and processes and
getting on with minimum noise.
Organizations are very rarely one culture type, so the exercise
here is to circle the present behaviours for Place and People.
You will get two or more relevant clusters showing that
workplace culture is type ‘X’ as the dominant type, with some
aspects of type ‘Y’. To demonstrate this concept, I will apply the
present and the target behaviours for the financial institution
results from the previous chapter to the culture cluster types:
The present behaviours are Team Focus, Conformity,
Strategic, Collective and Active Learning.
The targeted behaviours are Expressive, People Focus,
Empower, Outer Focus and Innovate.
The results are shown in Figure 12.2 and the present
behaviours are circled indicating that the current culture types
are Partner and Envision. The target culture type is a lot more
focused on Envision, along with Partner and some Explore.
 FIGURE 12.2 The culture types with the present and
targeted behaviours marked

SOURCE Present behaviours are circled and targeted behaviours are squared

Figure 12.2 details

Culture type – secondary behavioural clusters

The secondary clusters are generated at the innermost circle, as
demonstrated in the combined Figures 12.3, 12.4 and 12.5. This
is another useful lens and can in fact show you how drastic the
workplace culture type change is that is targeted.
FIGURE 12.3 Consider First vs Act First culture types

Figure 12.3 details

FIGURE 12.4 Question and Listen vs Define the
Answer culture types

Figure 12.4 details

 FIGURE 12.5 Ensure Safety vs Navigate Possibility
culture types

Figure 12.5 details

The definitions for these dichotomies are as follows:
Consider First: Taking time to think and learn, partnering
to make it happen and ensure compliance; vs Act First:
Explore and test on the go, own and deliver as we don’t
 have time to consider, it’s all about the here and now, we
don’t need to worry about compliance.
Questions and Listen: People and partner opinions are
important; openness and collaboration are the best route
for exploring, learning and pioneering for long-term
success; vs Define the Answer: We know what needs to
happen and no one needs to question it, it’s down to us to
make it happen now as time is of the essence, and as long
as we are compliant, we will follow the tried-and-tested.
Navigate Possibility: We are pioneers at what we do, and
regardless of the market condition and competition, we
will revolutionize regardless of market needs; we should
not worry about compliance, it’s all about being highly
competitive and being first in the market; vs Ensure
Safety: Tried-and-tested is the best way forward, being
compliant, on time and to budget are critical factors, we
don’t need to invent it, let’s seek alliances to deliver.
As you may have noticed, the secondary inner clusters are
constructed from three primary clusters and 15 behavioural
dimensions.
Using the same working example from the previous chapter,
let’s mark and then analyse the present behaviours and
targeted behaviours for the inner secondary clusters. See
Figures 12.6, 12.7 and 12.8.
 FIGURE 12.6 The Consider First vs Act First secondary
culture types

SOURCE Present behaviours are circled and targeted behaviours are squared

Figure 12.6 details

In Figure 12.6 present behaviours indicates that the culture
type is ‘Consider First’. When taking the targeted behaviours
into consideration, that still seems to be the dominant type.
 Let us now examine the next secondary cluster in Figure 12.7.
The target seems to be more of ‘Question and Listen’, so no
change in culture type there.
 FIGURE 12.7 The Question and Listen vs Define the
Answer secondary culture types

SOURCE Present behaviours are circled and targeted behaviours are squared

Figure 12.7 details

Moving on to the final secondary cluster in Figure 12.8, this
does indeed show a culture type change, targeting ‘Navigate
Possibility’ as the dominant culture type while maintaining
some ‘Ensure Safety’.
 FIGURE 12.8 The Navigate Possibility vs Ensure
Safety secondary culture types

SOURCE Present behaviours are circled and targeted behaviours are squared

Figure 12.8 details

So, with the culture types approach, we know now that this
organization is targeting a culture for ‘Navigate Possibilities’,
which means a drive towards Envision and Explore. The
behaviours core to this change are: Outer Focus, Innovate,
Empower and Expressive.
I hope this illustrates how culture types can be useful to
communicate and explain a targeted culture change.
Another good use for the culture types is in generating or
changing organizational values. With the diagnostic approach
measuring the ground-up behaviours of an organization,
followed by the predictive analytics process and then applying
the culture types lens, you will be able to create the values
based on the targeted culture rather than a ‘gut feel’ approach
that is not steeped in reality and evidence.
Using the example above, perhaps one of the organizational
values for this financial institution should be ‘pioneering for a
better future’.

Workplace culture tensions

The culture types methodology also shines a light on the idea
that your organizational culture can’t be all things to all people,
and you will need to make choices. Essentially there are three
workplace culture tensions you will need to make a choice on
as to where you lean towards:
The risk tension: As the name may suggest, this is about
how you manage risk. It’s the culture types dichotomy
between Explore and Verify. For ease of reference, Table
12.1 shows the behaviours and definition reminders.
The progressive tension: This is about how progressive
the organization wants to be. It’s the culture types
dichotomy between Apply and Envision. For ease of
reference, Table 12.2 shows the behaviours and definition
reminders.
The independence tension: This is about how
partnership-, alliance- and team-oriented the organization
wants to be. It’s the culture types dichotomy between Own
and Partner. For ease of reference, Table 12.3 shows the
behaviours and definition reminders.
TABLE 12.1 Culture types and behaviours for the Risk
Tension
Skip table

Verify Explore
Neutral Expressive
Feelings and opinions are expressed Feelings and opinions are expressed
delicately and with concern for how they openly whether they are comfortable
will be received; there is a focus on being or not; there is a focus on people
audience aware and frankness is seen as calling it like they see it and frankness
blunt and inconsiderate. is seen as honest and helpful.
Disenfranchise Empower
Hierarchical, command-and-control work Enabling, non-hierarchical work styles
styles dominate, and senior leaders are dominate where senior leaders trust
held directly responsible; there is a clear others. Decisions are delegated to
focus on keeping decision-making at the lower levels with supportive guidance.
top and directing.
Moderate Radical
Stability and working in steady, known Continuous improvement and
ways are admired and actively originality are admired and actively
encouraged; there is a clear focus on encouraged; there is a clear focus on
following process and trusting what has needing to try new things and push
worked in the past. boundaries.
Conformity Nonconformity
High emphasis and regular attention are Rules are seen as a starting point which
given to formal laws, rules, standards and should guide, but not overly constrain,
obligations; there is a clear focus on judgement; there is a clear focus on
trusting in the rules and sticking tightly to interpretation and doing what feels
them. right in the moment.
Standardized Flexible
Those who eliminate ambiguity and Those who appreciate ambiguity and
establish efficient operating templates build flexible, easily varied ways of
are admired; there is a clear focus on working are admired; there is a clear
getting consistency and uniformity in focus on giving license to work in
how things are done. different ways and be adaptable.
TABLE 12.2 Culture types and behaviours for the
Progressive Tension
Skip table

Apply Envision
Consolidate Innovate
‘Risk’ is seen primarily as a negative ‘Risk’ is seen primarily in terms of
threat to be minimized; there is a clear opportunity, and a degree of failure is
focus on reinforcing existing, known accepted as inevitable; there is a clear
value and steering clear of what could focus on seeking new value and being
go wrong is seen as praiseworthy. creative is seen as praiseworthy.
Tactical Strategic
‘Getting things done’ is a driving ‘What we are trying to achieve’ is a
mantra and this is seen in terms of driving mantra and is seen in terms of
imminent delivery; there is a clear broader direction; there is a clear focus
focus on execution and meeting on looking past delivery to overarching
pressing, immediate business needs. purpose and longer-term needs.
Passive Learning Active Learning
Development is seen in terms of know- Development is seen as ongoing personal
how picked up as a byproduct of on- and professional growth; learning is seen
the-job experience; learning is as inherently valuable and ongoing and is
prioritized when there is a clear treated as a valuable outcome in of itself.
practical need for specific skills or
knowledge.
Sequential Synchronous
Perspectives are grounded in the here- Those who explore differing perspectives
and-now and those who work in an and see issues within a wider context are
ordered, applied way are admired; admired; there is a clear focus on the
there is a clear focus on not getting ‘why’ and treating issues and decision
distracted and keeping it simple. making as interconnected.
Internal Focus Outer Focus
Bringing the environment and those Looking outside and being influenced by
around us into line with the ‘right’ way other environments, people and
of doing things is valued; there is a perspectives are valued; there is a clear
clear focus on asserting what we know focus on questioning assumptions and
is important and what works for us. actively checking how what we do affects
others.
 TABLE 12.3 Culture types and behaviours for the
Independence Tension
Skip table

Own Partner
Transact Relationship
Relationships are valued for their Fostering longer-term relationships is
usefulness and are seen as a means to an seen as important and admirable;
end; there is a clear focus on prioritizing there is a clear focus on deepening and
your immediate, practical stakeholders sustaining connections over time
and then moving on. beyond key stakeholders.
Individual Collective
Goals are formulated and achieved based Goals are formulated and achieved
on individual responsibilities; across the through groups and partnerships;
business there is a clear focus on across the business there is a clear
individual competition and outperforming focus on succeeding (and failing)
others. together.
Self Focus Team Focus
Personal competence is actively prized, Team capability is actively prized, and
and independence is admired; within collaboration is admired; within teams
teams there is a clear focus on ensuring there is a clear focus on prioritizing
people meet their own obligations before what will help the group the most
helping others. before personal interests.
Delivery Centric People Centric
Getting things done is more highly prized How things got done and the impact
than the manner in which they are on people is as important as whether
achieved; there is a clear focus on tangible they got done; there is a clear focus on
outcomes and organizational creating a sense of community where
effectiveness. everyone is encouraged.
Achievement Status
Status is downplayed and the best Status is highly prized and ‘who thinks
contributions and inputs are recognized what’ can be a bigger focus than what
no matter where they come from; what makes sense; there is a clear focus on
needs to be said is more important than what important people want and
who is saying it or who is listening. being loyal to one’s seniors.

To showcase these tensions, I will apply the financial institution

results for present and targeted behaviours to the three
tensions:
The risk tension: they are targeting two behaviours from
Explore and one behaviour from Verify, hence they are
slightly leaning towards risk-taking, but not massively as
it’s only two behaviours.
The progressive tension: they are targeting four
behaviours from Envision and none from Apply, hence
heavily leaning towards being more progressive.
The independence tension: they are targeting three
behaviours from Partner and none from Own, hence
reasonably leaning towards more partnership and
alliances.
These tensions are yet another lens to help you articulate your
current culture and what you are aiming to be.
That chapter should also help you turn quantitative into
qualitative, to tell culture stories and illustrations. It helps you
simplify the message, having explained the potential ROI as per
the previous chapter.
Talking about tensions, there is another tension type to
consider: the tension between the People and the Place,
showing how aligned or misaligned the behaviours are. This is
useful as we do want both aligned against the target for the
behaviours to fully manifest. Figure 12.9 illustrates the
alignment or tension between Place and People. The diagram
shows the behavioural dimensions tension in percentage terms
for the dichotomous factors for each dimension, and it also
indicates for each targeted behaviour how aligned or
misaligned it is and to what degree is this alignment against the
targeted behaviour. Consider the Empower behaviour: it has
the most tension between Place and People, and the People are
more aligned to the targeted behaviour; hence the Place is the
inhibitor of that behaviour manifesting.
 FIGURE 12.9 The percentage behavioural tension as
alignment or misalignment between Place and People
A graph shows the alignment between place and people
with behavioural dimensions tension for the dichotomous
factors for each dimension.

Figure 12.9 details

Figure 12.9 is not showing you new data; it is, however,
showing the behavioural tensions between Place and People
and their alignment with the targeted behaviours. Again, it
could be useful to demonstrate the behaviours most in tension.
13
Applying behavioural
interventions
As I have already shown in Chapter 2, it’s important to take
action on behavioural insights. I have already outlined the
details behind the Culture by Design approach using capability,
opportunity and motivation, so to remain practical let us now
apply this approach to this chapter.
Culture by Design has five stages, summarized as follows:
Stage 1: Behavioural selection and context:
Understanding how the target behaviours manifest in
your organization and using a model of behavioural
change to understand exactly what needs to change and
where. This involves workshops with selected teams based
on ‘where’ the targeted behaviours are absent or present.
This information can be found using the quadrant
distribution graphs showing functional/geographical areas
to understand localized context.
Stage 2: Design the intervention: Using the outputs from
Stage 1, this will lead you through the process of designing
impactful behavioural interventions and the most
appropriate methods to bring about the desired change at
the targeted level. The proposed interventions should be
assessed according to the APEASE criteria – Affordability,
Practicability, Effectiveness, Acceptability, Side effects and
Equity.
 Stage 3: Plan the intervention: Establishing a clear plan
for the specific intervention content, sequence the
interventions and provide an implementation roadmap
ready for implementation.
Stage 4: Implement: Implement according to the action
plan, keeping track of who is doing what, where and by
when. Progress should be reviewed regularly – I suggest
weekly – to make sure that you stay on track and apply
corrections to the master plan.
Stage 5: Validate: This is a remeasure stage, measuring
the behaviours and targeted outcomes to check progress.
This should be done between six and nine months from
implementation. Compare before and after to validate that
what you have implemented is working – you should
expect behaviour and outcome changes. This stage will
provide with the evidence needed to keep going or make
adjustments. If any adjustments are needed, you will need
to go back to Stage 2.
Let’s get into the details for each stage.

Stage 1: Behavioural selection and context

The first action to take is to articulate the shift from current to
future for the targeted behaviours, which is another team
exercise that you can do.
To do this classroom-based exercise, you will need to select
teams from each of the quadrants of the quadrant dispersion
graph. This process was outlined in Figure 5.3. What you are
seeking to know is ‘what happens’ for each targeted behaviour
for the teams in the top right corner of the quadrant (as a
reminder; the targeted behaviour is present for Place and
People) versus ‘what happens’ for the teams in the other three
quadrants.
To achieve this, you should provide breakout rooms for the
teams as per the quadrant information, and for each targeted
behaviour they should complete Table 13.1. Before the start you
should explain to them what the behaviour means, and that you
are asking them to explain the context for this behaviour in
their own world. The teams in the top right corner of the
quadrant will most likely elaborate well on ‘where are we now’
and will struggle with ‘where we need to be’. That’s OK, as for
this team, you are really interested in the ‘where are we’ as that
will be the target for other quadrants.
 TABLE 13.1 Template to be completed as a team
exercise to show the current and future shift for each
targeted behaviour
Skip table
Where Are We? Where Do We Need to Be?
Organization Organization
Behaviour Behaviour
– What happens here – What should happen
Belief Belief
– Why it happens – Beliefs we will embrace
People People
Behaviour Behaviour
– What I do – What I will do
Belief Belief
– Why I do it – Beliefs I will own

Notice that Table 13.1 includes People and Place (Place is

labelled ‘Organi- zation’). You can get more specific if the target
behaviour is only a Place or a People issue, which is in line with
the COM-B model as outlined in Chapter 2.
The number of workshops will be in line with the number of
targeted behaviours; as an example, if you are targeting three
behaviours then you should plan for three workshops, and the
participants for each workshop should be selected to represent
all four quadrants.
To provide you with an example, see Table 13.2. The targeting
behaviour is Empower and the example is an output from one
of my many workshops.
TABLE 13.2 Exercise showing the current and future
shift for the Empower behaviour
Skip table
Where Are We? Where Do We Need to Be?
Organization Organization
Behaviour Behaviour
– Decisions are escalated up to our – Taking and escalating decisions
managers most of the time are both common
– A lot of managers talk about – A lot of managers talk about
needing to know what is trusting their people to know
happening in their area when to involve them
– Those who act without direct and – Those who act without formal
formal approval are frowned upon approval are often spoken about
– Managers delegate responsibility as having shown initiative
only once they are sure there is no – Responsibility is delegated to the
risk people who are best placed to
– The vast majority of meetings manage the risks
concentrate on what the most – Meetings concentrate on
senior people think relevance of views and who is
– Communication happens through doing the work
senior managers – Communication happens through
key project drivers
Belief Belief
– The default assumption is: ‘check – The default assumption is: ‘I am
in with the boss’ trusted to know what to own and
– The status of the person speaking what to escalate’
is more important than insight – The substance is more important
– It is a manager’s job to ensure than the speaker
everyone is doing what they ought – It is a manager’s job to create an
to be doing environment to allow their team
– It is not junior people’s place to to do their jobs well
speak for the team – Everyone in the team should have
a voice in the business
People People
 Where Are We? Where Do We Need to Be?
Behaviour Behaviour
– I look to my manager to tell me – I reach out to my manager to
what I should and should not be figure out together what I should
doing be doing
– I make sure I am only doing what I – I do what I feel supports the goals
am confident I have been told to of the team and my manager
do – I counsel others to be willing to
– I give feedback to peers who try lead discussions
and have a voice which may differ – I give feedback to people who just
from that of their boss wait to be told
– I share stories of those who got – I call out others for telling
into trouble by not working within cautionary tales about people
the system failing to fall into line
– I wait to be told where the team is – I set to play an active part in
prioritizing where the team should be going
Belief Belief
– My manager knows better than – My manager and I know different
me, their manager knows better things
than them, and so on – My manager needs my help to
– It is safer to do as you are told lead the team well
– It is not my place to lead – It is often my place to lead
– If I try to lead, my manager will be – If I help lead, my manager will
upset appreciate the support
– I will be better off if we stick to – Rigid hierarchy can sometimes
clear lines of authority hold us back

If during the workshop you find the teams are struggling to

complete Table 13.1, you can further facilitate by asking direct
questions about capability, opportunity and motivation.
Example questions for capability might be:
How aware are they of the behaviour and precisely what
it entails?
How well do they know how to do it?
 How well do they understand the benefits of doing it? Or
not doing it?
For opportunity, example questions might be:
How far is it considered ‘normal’ within their
environment?
How much support do they get to do it?
Do they have time to do it?
For motivation, example questions might be:
How worthwhile is it to do it?
Will doing it service a need?
Will people be recognized for it?
Post-workshop, you will need to assess the responses by
quadrant and create a table for each target behaviour
summarizing the findings in terms of the ‘what’ and the
‘where’. This will help you crystallize all your findings in one
simple table before the design stage. See Table 13.3. Note that
you can have more than one entry per behaviour, i.e. more
than one row.
TABLE 13.3 Example extract for Empower behaviour
analysis post-workshop
Skip table

EMPOWER (targeted behaviour) – Some people are able to empower others;

however, the place generally doesn’t provide the opportunity to delegate
decisions nor provide supportive guidance. What changes can we make at the
place to provide this opportunity?
vs
DISENFRANCHISE – The structure is hierarchical, ensuring that leaders retain
decision-making.

Where does this

Behavioural Behaviour manifests happen/not
Dimension (reason it happens) happen Who is involved
Empower vs Some people are able Throughout the Everyone starting
Disenfranchise to empower others; company with the from executive
Need to increase however, the place exception of the leadership team
empowerment generally doesn’t Operations team down is suffering
and reduce provide the (that’s the place to from this. Even
disenfranchise opportunity to learn from). executive leaders
throughout the delegate decisions nor suffer from a lack
organization provide supportive of empowerment
guidance. What from the board of
changes can directors.
we make at the place
to provide this
opportunity?
Empower vs Too many team All organization, As above
Disenfranchise leaders, at all levels, slightly less
turn up to every prominent in
meeting. Operations team.
Empower vs Everyday decisions All organization, As above
Disenfranchise have to be checked by better in
many leaders and Operations team.
committees.
Stage 2: Design the intervention
Once you have completed this exercise for each of the targeted
behaviours, and have generated the summarized analysis for
the context, you can now move to designing the intervention.
Before we start the intervention design, it is important to
remind you of the COM-B process for driving behaviour change.
I introduced this concept in Chapter 2.
Figure 2.1 should now start featuring in Table 13.3 as we
identify for each row whether it’s a physical capability,
psychological capability, physical opportunity, social
opportunity, reflective motivation or automatic motivation. You
can also tag each entry as an inhibitor or enabler. See Table
13.4.
TABLE 13.4 The COM-B function labelled with the
behavioural inhibitors for Empower
Skip table
EMPOWER (targeted behaviour) – Some people are able to empower others; howeve
place generally doesn’t provide the opportunity to delegate decisions nor provide su
guidance. What changes can we make at the place to provide this opportunity?
vs
Disenfranchise – The structure is hierarchical ensuring that leaders retain decision-m

COM-B Functions
physical
capability
psychological
capability
physical
opportunity
social
opportunity,
reflective
Behaviour motivation
Behavioural manifests (reason automatic
Dimension it happens) motivation Inhibitor Enable
Empower vs Too many team psychological x
Disenfranchise leaders, at all capabilityphysical
levels, turn up to opportunitysocial
every meeting. opportunity
reflective
motivation
Empower vs Everyday psychological x
Disenfranchise decisions have to capabilityphysical
be checked by opportunitysocial
many leaders and opportunity
committees.
Table 13.4 provides clarity for the type of intervention needed.
The intervention must turn the inhibitor into an enabler, and it
must function as a psychological capability, physical and social
opportunity enabler and provide reflective motivation. Some
interventions can target multiple behaviours.
As an example, to use a psychological capability to enable
Empower, training must be provided to show them how to
delegate effectively, and how to communicate the outcomes
they require from their teams with clear parameters and
supportive (explanatory, not directive) guidance. Empowered
teams can design and decide as to ‘how’ they can achieve these
outcomes.
As another example, using a physical opportunity to enable
Empower, the board and leadership team will need to define
the role and position within the ‘delegation chain’ and
communicate that to all relevant teams and functions. This
intervention will also cascade into ‘who’ should be attending
‘which’ meeting and ‘why’.
For each intervention the following validity criteria must be
considered:
The intervention must be context-applicable – can we do
this at our organization?
Is the intervention focused on unlocking the target
behaviour?
Is it a sustainable intervention, or temporary?
Is it measurable in terms of behaviour and outcome?
Can it be applied company-wide, or for the entire targeted
team? You certainly don’t want it to work against
inclusion!
Eventually you should summarize all interventions in one table.
As examples, please see Table 13.5 for an executive committee,
and Table 13.6 for team leaders addressing more than one
behaviour.
TABLE 13.5 Intervention design for an executive
committee
Skip table
Intervention &
Aim Training Process Technology Intern
To review role and – X – Interna
position within
the ‘delegation
chain’. Need to
pull back from the
line and empower
subordinates and
be properly
empowered by
the Board.
Outcomes must
be communicated.
Team dynamics X – – Partne
training. Provide
Executive team
with improved
team dynamics
moving forward,
the ability to
understand how
event-driven
behaviours
manifest in teams;
to improve
situational
awareness,
managing
complexity,
resolving conflict
and thereby
enhancing
teamwork for
minimizing risks
and driving
performance.
TABLE 13.6 Intervention design for leadership team
Skip table
Intervention & Aim Training Process Technology In
Leadership delegation X – – Pa
training. This training
focuses on leaders
being able to
effectively delegate –
being able to
communicate the
OUTCOMES they
require from their
team with clear
parameters and
supportive (not
directive) guidance.
Colleagues are then
empowered as to
HOW they achieve
these outcomes.
360 leader reviews. – X X In
360 leader reviews
should be conducted
focused on
Empowerment and
People Focus. Specific
questions on HOW
things are achieved as
well as WHAT, and also
on the ability of a
leader to delegate
effectively.
 Intervention & Aim Training Process Technology In
Review and allocate X X – In
mandates for
leadership group. Each
leadership level should
have clear and
empowering
parameters in terms of
their mandate, to
include things such as
budget, paper sign-
off/review/recruitment
and regulatory impact.
This will detail when
they are expected to be
the sign-off and when
it should be delegated
or promoted up the
line.

Stages 3 and 4: Plan the interventions and

implement
You can have a high-level time plan for ease of reference and
presentation, but you should also have an Excel-type time plan
for tracking progress and resources. See Figure 13.1 for an
summary time plan.
 FIGURE 13.1 Intervention time plan implementation

Figure 13.1 details

It is critical to review progress on the above events weekly. You
should have an Excel-style sheet to track the implementation.
As an example detailed time and resource implementation
plan, see Figure 13.2 covering interventions for Expressive
behaviour only.
 FIGURE 13.2 Example behavioural interventions time
and resource plan

Figure 13.2 details

Additionally, you should consider how applying behavioural
nudges might help. Nudges have emerged as a powerful tool for
influencing individual choices and fostering positive
behavioural change. In both personal and organizational
contexts, applying behavioural nudges strategically can lead to
the improved adoption of desirable behaviours.
Behavioural nudges involve subtle interventions designed to
guide individuals towards making better choices without
restricting their freedom. These aim to influence behaviour by
leveraging cognitive biases that often drive decision-making.
For completeness, I would like to share the core principles of
effective nudging as follows:
Defaults and opt-out mechanisms: Nudges often
leverage the power of defaults. By setting a particular
option as the default (which individuals can opt-out of if
desired), decision-makers are more likely to stick with the
default choice.
Social norms: People are influenced by what they
perceive as normal behaviour within their social context.
 Nudges can tap into this by providing information about
what others are doing.
Simplicity and clarity: Nudges should be simple and easy
to understand. Complex messages can lead to decision
paralysis. Clarity in communication is key to ensuring that
the nudge is effective in guiding behaviour.
Immediate feedback: Providing immediate feedback on
behaviour can reinforce positive actions.
Application timing: The timing for applying the nudges
would be important – i.e. when you start, and for how
long.
Back to our context, this will be an interesting and fun exercise
to do with your team. Divide the team by four and take each of
the above principles and design a nudge.
For this example, the organization provided digital nudges in
the form of emails and group Teams messages across the entire
year. The nudge message was clear, and consisted of the
following:
Do you need to attend all the meeting in your calendar for
this week? Will you make a difference?
By empowering your team, you can, on average, gain 32
per cent effective time back in your calendar.
Have you blocked some think time in your calendar
today?
In fact, if you have deployed an organizational network and
calendar analysis system, you will be able to customize
feedback messages weekly showing progress. And I mean
customized for each individual; an excellent method for a
customized nudge!
Stage 5: Validate
This is a remeasure stage, measuring the behaviours and
targeted outcomes to check progress. Compare before and after
to validate that what you have implemented is working – you
should expect behaviour and outcome changes.
As an example, having targeted Empower for seven months
as per the intervention design and implementation outlined in
this chapter, an organization redeployed the behavioural
measurement and remeasured the targeted outcomes.
Empowerment increased by 18 per cent from absent to
present at the Place. The outcomes related to Empowerment
improved by 48 per cent, which is a significant shift; these
outcomes included speed of decision-making, improvement in
leadership time management and process-related performance
efficiencies, to name a few.
This is an excellent result and will help the organization stay
focused on their implementation plan to yield more outcome
benefits.
If any adjustments are needed, you will need to go back to
Stage 2 – design the intervention.
14 The impact of growth on organizational
culture
Navigating the dynamics of change

Growth is a natural and often sought-after aspect of an organization’s

journey towards success. However, as organizations expand and evolve,
they must confront the profound impact that growth can have on their
workplace culture.

I will now remind you of the thought-provoking idea I shared at the end of
Chapter 1, that growth can be the enemy of organizational culture. Hence,
growth must be treated as a workplace culture change, and as I have
already covered throughout this book so far, one must not leave workplace
culture to chance! This concept is even more critical if any change is
happening or about to happen.

Many organizations of all shapes and sizes also battle with answering a
significant question as part of their journey – do we drive growth
organically, or through a merger or acquisition? Both paths can present
significant opportunities and risks; however, leaders often forget to think
about the significant role that culture will play. For organizations that are
on a significant organic growth path and have a great current culture, the
oddity is that such growth could risk partial or total destruction by culture!

We have seen this before – remember the Boeing story I shared with you in
Chapter 1? And this is the key reason why one must not gamble with
workplace culture. The key is validated measurement and management. In
fact, culture is key in driving ‘how’ growth can be sustained.

As you grow and add people or even suppliers to your team, your culture is
at risk. After all, the more people involved, the greater the opportunity for
dilution. To help curb this effect, recruit with your target culture in mind
and hire people whose values and behaviours are compatible with your
organization. Before I hear the shout ‘What about diversity?’, I am not
suggesting you only hire people from a certain background, ethnicity,
gender, etc… not at all. I am suggesting that you must know which few
behaviours are critical for your workplace culture, as they drive and predict
many of your desired outcomes, and hire in line with the those behaviours.

I must also remind you that I am not talking about personality measurement
as part of the recruitment strategy – that may be hazardous for diversity. I
am merely referring to the few behaviours that are critical for you.

Effective communication is also key to manage the growth effect on

organizational culture. As organizations grow, communication channels
may become more complex and fragmented. This can lead to
misunderstandings, reduced transparency and an overall breakdown in the
flow of information. Leaders must invest in robust communication
strategies, leveraging technology and fostering open lines of dialogue to
maintain a cohesive and inclusive culture throughout periods of growth.

Growth often necessitates changes in leadership styles and organizational

structures. The leadership that successfully guided a small start-up may not
be equipped to manage a larger, more complex organization; hence, as the
organization grows, leaders must adapt their styles, delegate effectively and
foster a workplace culture of empowerment. This adaptability is crucial for
maintaining a positive and cohesive organizational culture. Again, the only
way leaders will be well informed and prepared is by articulating the
behaviours that got us here so far and the behaviours that they need in their
workplace culture for their next destination.

The role of workplace culture in mergers and

acquisitions as a strategy for growth
Having already stated that growth can be driven by mergers and
acquisitions (M&A), why is it that the results of mergers and acquisitions
don’t always live up to expectations?

M&A growth strategies promise a plethora of strategic opportunities; from

rapid growth, to eliminating competition, to access to new markets. And
many organizations are currently embarking on, or have embarked on,
merger and acquisition growth strategies to varying effect.

When asked about the primary causes of these mixed results, most leaders
cite a misalignment between the two organizations’ cultures. This friction
can wreak havoc as the members of different groups integrate to drive the
performance gains that M&A strategies forecast.

Companies of all shapes and sizes are actively engaged in M&A growth
strategies to help position themselves to compete. And organizations of all
sizes must be mindful of how company culture plays a role in M&A
success.

Typically, M&A activities are conducted with a focus on the external

impact. How are we going to gain access to new opportunities? How will
this position us better to gain market share? How will this afford us
economies of scale to keep ahead of our competitors? Post-M&A, the focus
shifts to systems – how do we integrate our human resource, financial and
customer management systems? How will we ensure that our operations
can integrate as quickly as possible?

The risk here is two-fold. First, if the transaction was initiated for
externally focused reasons, becoming hyper-focused on internal integration
can distract people from why the deal happened in the first place. Second,
and perhaps more problematic, is that the promises of financial gain can be
seductive, leading the organization to find data and examples to support the
notion that the organizations are ideally suited to integrate, rather than
looking at data objectively.

It’s not uncommon to hear executives talk about how the cultures of the
organizations are well-aligned, and how this will facilitate an easier
integration. We’ve seen it play out time and time again as leaders convince
themselves that the integration is a ‘marriage made in heaven’ when, in
fact, they have no real understanding or data on which to base that claim. It
seems ludicrous that companies would roll the dice with something so
critical without understanding the magnitude of organizational culture and
its impact on the M&A outcomes.
A culture-first approach to mergers and acquisitions

With significant advancement in behavioural diagnostics and people

analytics, it is now possible to approach post-M&A integration with a
scientific focus to measure organizational culture and effect on key
performance indicators and outcomes.

Take the opportunity to really understand the culture of each organization.

You can achieve this using behaviour diagnostics (the leading indicator) for
each company to examine ‘how’ each company lives its culture and, very
importantly, the impact on each organization’s outcomes (the lagging
indicator).

Having looked at each organization’s own culture, you can now start
building the ‘third’ resulting target culture by knowing what behaviours
you want to sustain, change or eliminate. Where are the culture
misalignments that will cause pressure points? Most importantly, you can
build a predictive model for how the ‘third’ resulting culture will be our
hero in making the M&A a great success, rather than the villain that will
destroy it, Figure 14.1 summarizes this approach and sequence of actions.

FIGURE 14.1 Steps involved in the workplace cultural

integration of two independent companies

A flow chart shows the approach and sequence action of four

steps for workplace cultural integration of two independent
companies.

Figure 14.1 details

Using this approach will also allow you to shed a critical light on what each
organization ‘says’ and how it relates to what they actually ‘do’. To
demonstrate, two organizations may use the term ‘customer-centric’;
however, their approach to this word is very different and drives very
different behaviours. The risk in this situation is that both organizations
define customer success differently, and how they deliver success looks so
different that it can create frustration or even confusion as employees
struggle to figure out what the new ‘right way’ looks like across the new
integrated organization.

Guiding a successful ‘culture first’ integration is a tricky transition and

needs focus and intention, as most leaders want to complete the integration
process as quickly as possible in order to reap the financial benefits of the
transaction, but this can come back to bite them.

Using behaviour analytics to find out which behaviours are the key drivers
for the manifestation of organizational culture will allow you to dig deep
into ‘what right looks like’ in each organization, to ensure you don’t jump
onto a directionless culture bandwagon and drive yourself into the first
ditch.

In today’s environment, you’d be hard-pressed to find a business executive

who has not lived through (some may say survived) a merger or acquisition
in their careers. Why is it, then, that we continue to try to convince
ourselves that culture is not going to present a significant risk to our future
transactions and their ability to drive the returns that we hope for?
Whatever expansion strategy you embark on, organic or through M&A,
organizational culture is ‘valued currency’ in driving sustainable growth.

Whatever your approach for growth and even if you seek to modify your
organizational culture as you grow, behavioural diagnostics and analytics
should be your guiding ally for success.

Growth is an exciting and challenging phase in any organization’s journey.

Recognizing the impact of growth on workplace culture and proactively
addressing the associated challenges, by taking an active approach through
measuring, adapting and validating to better navigate the dynamics of
growth, will help you preserve the intended aspects of your unique
workplace cultural identity while driving your objectives and projected
outcomes.
15
Cultural navigation in the digital
age
As we find ourselves firmly entrenched in the digital age,
organizations are facing unprecedented challenges and
opportunities. The rapid evolution of technology has not only
transformed institutional operations, but has also significantly
influenced workplace culture.
The digital age brings with it a wave of technological
advancements that can reshape the way organizations operate,
from cloud computing and artificial intelligence (AI) to remote
collaboration tools. Embracing so-called ‘digital transformation’
is essential as the way we work is rapidly changing.
Organizations must navigate this shift with a keen
understanding of how these technologies impact their
workplace culture, and what opportunities this brings for
culture and people analytics.
In this chapter, I want to address three aspects: the impact of
the digital age on workplace culture, the opportunity for
connected data to provide actionable workplace culture
analytics, and finally the ethical use of data for workplace
culture analytics.

The impact of the digital age on workplace

culture
I am sure you will immediately connect with this, as the
emergence of virtual work environments and remote working
has become a defining feature of the modern workplace.
Suddenly workplace culture has no walled confines – but I
would argue that workplace culture had no walled confines in
many industries already. To explain this concept and as an
example, think about the transport industry. The airline or train
operators rarely have their core workforce in a single office
building, or even in one geographical location, yet
organizational culture travels so well and you can observe how
the majority of employees enact their respective workplace
cultures. Additionally, that applies to managing risk and
operational performance. Having myself lived on a flight deck
for part of my career, I am still fascinated by how behaviour
risk travels remarkably well from head office to every airliner
cockpit and cabin literally thousands of miles away.
This concept is also no different with big brands that have
cross-border satellite offices. It is also remarkable to study how
such workplace cultures manifest alongside the geographical
culture (known as ethnography).
So, the elephant in the room is: how do you manage this level
of complexity for organizational culture? The good news is that
the approach outlined in this book will be exactly the same for
all types of employees: remote, part-time, interns, casual, full-
time… you name it! Hence it is an inclusive approach and a
good demonstration of how a valid, evidence- and data-based
approach is imperative in managing complexity.
The key approach remains the same – valid, easy-to-deploy
behavioural diagnostics, the availability of multiple relevant
thematic outcomes to drive actionable insights, and the work
on the leading indicators. Remember it is an iterative cycle so
keep measuring, adapting and validating.
Additionally, managing workplace culture in the digital age
demands a unique set of leadership skills: the ability to lead
remotely, communicating effectively through digital channels
and leveraging data-driven insights are all crucial skills.
Progressive organizations in the digital age go beyond
traditional hierarchies and leaders should create the
environment that emphasizes collaboration, empowerment and
inclusivity.
The digital age should not mean that we must all have a knee-
jerk reaction and become ‘off the scale’ innovative. You should
specifically assess what it means for your organization, as I
have already made the case for in Chapter 1. A copy-and-paste
approach will not work, and it cannot replace the evidence-
based process outlined.

The opportunity for connected data to provide

actionable workplace culture analytics
To drive effective and actionable analytics and insights, I have
already outlined the importance of connecting the lagging and
leading indicators. Actually I went a lot further towards
building the full analytics picture and I have already covered
how you can bring in quantitative and qualitative outcomes, as
well as any organizational values into the mix. Therefore, I
would like to argue the need for a big data approach. Even if
you have not set up the technology to connect all the needed
data through application programming interfaces (APIs), no
problem. I am sure most platforms have an export function so
you can test what connected analytics would do for your
organization.
Figure 15.1 helps you visualize everything I have already
covered in previous chapters in one image, and as you look at
this image, do allow your thoughts to go wild! You should be
inquisitive as to what your workplace culture is driving without
you even knowing.
 FIGURE 15.1 The big data digital approach to culture
analytics

Figure 15.1 details

Let’s analyse Figure 15.1 as to the different layers and elements:
The diagnostics layer includes any leading indicators
measurement. In the context of workplace culture, it is the
behavioural measurement for People and Place. You will
also notice that I am suggesting the injection of HR
demographics data at that layer. It may seem odd that I am
suggesting so many pre-analytics; however, HR
demographics will play a role in two ways. Firstly, it will
help you decide if you have achieved statistical and
analytics validity – the so-called ‘effect size’. I have already
explained what that is in Chapter 6. Secondly, HR
demographics will provide you the ‘where’ in your
analytics for actionable insights.
 The analytics layer includes organization thematic
quantitative outcomes. As is shown in Figure 15.1, that can
include performance data, compliance incidents,
organizational network analysis data, talent retention
data, etc. This data can be ‘line by line’ or at team level for
each outcome, and remember you can have more than
one outcome per theme. This layer will also include
qualitative data and organizational values or any
frameworks-type mapping. Qualitative data can include
employee engagement, wellbeing, psychological safety and
any other outcomes. This data is generally collected
through quick Likert scale-type sentiment surveys, but
other means of data collection can be added – perhaps a
combination of qualitative and quantitative data feeds can
be added for some themes.
The culture analytics function in the middle is where all
the model development happens (regression analysis, path
analysis, distribution analysis and odds ratio calculations).
The functionality will also include all the graphics to show
the insights, as described in previous chapters.

Ethical considerations for using employee data

for workplace culture analytics
As we are in an era where data is fast becoming a cornerstone
of organizational decision-making, measuring and analysing
organizational culture through data is gaining prominence.
While data-driven insights offer valuable perspectives on
organizational dynamics, it is crucial to address the ethical
considerations surrounding the use of data in this context, with
the emphasis on the importance of responsible practices to
ensure integrity and fairness.
The first notion to consider here is that the type of
behavioural diagnostics and analytics data will be analysed at
team, division or organizational level, and not intended for
individual reporting. Also, this solves an immediate critical
issue around data privacy as you will not need to process
personal identifier-type data, e.g. name, email address or
employee ID number. All you really need to examine the data
and produce a model for predictive analytics is a pseudo-
randomized ID that will be used to connect the diagnostics with
the corresponding demographics and outcome metrics.
A word of warning about the HR demographics, as these
should not include any criteria that can only be attributed to
one employee. I have already explained in Chapter 6 why
generated benchmarks for analytics should use five or more
employees.
You will still need to obtain informed consent from
employees, and individuals should be made aware of the data
collection methods, the purpose of gathering data and how the
insights will be utilized. The best approach is to prepare a
culture metrics and data analytics privacy notice, which should
include:
what data are you collecting
why you need this data
where the data is stored
conditions for data processing
how the data will be used
who can access the data and any access restrictions
data retention
This transparency builds trust and ensures that employees feel
comfortable contributing to the data collection process without
concerns about privacy violations.

Bias mitigation in data collection

Ethical concerns arise when data collection methods introduce
biases that may skew the interpretation of organizational
culture. However, the CultureScope behavioural diagnostic
mitigates this risk through the questionnaire methods and
approach to drive validity, as described in Chapter 4.
It is also essential to use inclusive and diverse sampling
methods to ensure that data accurately reflects the entire
workforce. Additionally, you should regularly review and
update your data collection sampling processes to identify and
rectify any unintentional biases.

Fair and objective analysis

When analysing data to measure organizational culture, it is
crucial to maintain objectivity and fairness. Avoiding
preconceived notions and letting the data speak for itself
ensures that interpretations are unbiased. Ethical
considerations should guide the interpretation of data insights
to avoid reinforcing stereotypes or discriminatory practices.
As a final word on navigating organizational culture in the
digital age, it is a multifaceted challenge that requires strategic
thinking, adaptability and a commitment to providing the best
actionable insight based on multiple data sources.
Embracing the possibilities of connected multiple data
sources will provide you with the best approach for analytics
validity, resulting in actionable insights focused on the leading
indicators and vastly improving the organization’s odds of
yielding their targeted outcomes and return on their culture
investment, making organizational culture the hero it should
be!
PART TWO
16 Gaining predictive insight into employee
sentiment through behavioural science and
analytics
Why four parts Place and one part People matters in whether employees
share their voice at work

Welcome to Part 2! Having established in Part 1 of this book all the

principles and methodologies for driving culture analytics using leading
and lagging indicators, and having provided practical exercises and
working examples, it’s time to examine thematic case studies. Each chapter
from this one onwards will be dedicated to a single case study.

To satisfy as many readers as possible, I will follow two methods for

presenting the case studies. Chapters 16 to 19 will be ‘white paper’ style –
this approach will remind you of the entire culture analytics process.
Chapters 20 onwards will be purely focused on analytics for actionable
insights.

Although the case studies are anonymized, all the analytics and insights are
‘real world’. Also, in some case studies, where the remeasure has been
deployed post-culture intervention design, planning and implementation,
the progress in behaviour and outcome results will be included.

Whether it is the sharing of information between co-workers to get ahead of

problems, refine and improve working practices and diffuse better practices
throughout the organization, or whether it is the upward communication
that helps organizational leaders keep abreast of what is happening in their
organization, most recognize that encouraging employee voice is an
essential factor in the success of organizations.

Background
Research published by the McKinsey Institute, ‘Why agility pays’, shows
that involving employees in shaping an organization’s vision is one of four
critical management practices that drive the agility of organizations in
adapting to and successfully introducing change.1

At the level of the individual employee, research supports the link between
employee voice and William Kahn’s original conception of employee
engagement, and voice or lack of it has been shown to relate to staff
turnover, higher employee commitment and higher satisfaction with both
work and employer.2

Before we go further, I should define what I mean by employee voice,

which refers to the communication of ideas, suggestions, opinions and
concerns about work-related issues with the intention of improving the
functioning of the organization. It includes the constructive challenge of the
current status quo in the organization or work unit about the way in which
tasks are assigned and executed. It does not include criticism for the sake of
criticism. The key attributes of effective employee voice are that it is
constructive and focused on improvement.

In contrast, employee silence represents the withholding of valuable input

and often reflects a feeling of futility (voice will be ignored or dismissed),
fear about negative consequences from sharing a point of view or a desire
to protect co-workers.

The connection between employee voice and employee engagement is

easily understood when you consider that sharing information is key to
employee learning, and that the notion of talent in many organizations is
predicated on the employee being an active contributor as recognized in
many performance appraisal and reward systems used by organizations.

Yet, while evidence supports the individual and organizational benefits of

encouraging employee voice, research also shows that employees are often
reluctant to speak up. That reluctance has remained stubbornly at around 50
per cent of employees feeling uncomfortable at expressing voice at work
for nigh-on the last three decades.3
What does that mean? Essentially, it means that speaking up and voicing a
view or an opinion is not the natural default behaviour in the workplace. It
means that expressing a voice or remaining silent is a choice, and that
choosing to remain silent is not necessarily a passive act.

Whether it is out of fear of negative consequences or a sense of futility

(‘they won’t listen anyway’), employee silence may actually represent a
conscious choice reflecting the belief that speaking up is too risky in terms
of the impact that may have on their relationships with co-workers, with
their line management, on their employment prospects and on their own
wellbeing at work.4

If the assumption that employees will naturally share their voice is

misplaced, then that has serious implications and potential consequences
for organizational leaders. For one, that misplaced assumption may lead to
complacency and the belief that ‘no news is good news’ when in fact
organizational leaders may have a much more limited view of what is
happening in their organization than they credit themselves with. Silence
may not be quite as ‘golden’ as leaders might assume. More about this in
Chapter 19.

That blindness to what is happening in the organization may also have

serious and very tangible consequences for the organization. Regulators
now recognize the importance of employee voice and employee silence as
factors related to the potential for organizational misconduct and even
illegal practices. Indeed, regulators in the banking and finance sectors now
look at employee voice or its absence as signalling what they are calling
‘broken cultures’. To quote one UK regulator, ‘Employee voice is critical –
are people able to contribute and challenge, be listened to, and therefore be
treated fairly?’5

So, how do organizations promote effective employee voice? The

prevailing wisdom seems to be that if organizations put the right structures
and processes in place, then employee voice will follow. Examples include
manager training schemes, town hall forums and employee suggestion
processes as well as whistleblowing systems adopted in many banking and
other companies.
These efforts are based on at least two assumptions. The first is that
employees have the talents required to constructively raise a suggestion or
voice a concern. The natural extension of that assumption is that if we, the
organization, recruit and train the right people then they will speak up.

The second assumption is that processes, systems and manager and

executive training will naturally create the right context to encourage
employee voice. Research on psychological safety and employee voice
suggests that these may be further examples of unwarranted assumptions.6

To borrow from the field of architecture, these assumptions represent a

misconception that form will follow function.7 That is, by creating the right
organizational ‘forms’ (structure and processes), employee voice will
naturally function.

This misconception fails to recognize that form and function will only work
to support employee voice if the culture of the organization promotes the
behaviours needed for those forms to function. And the proof that those
forms are functioning lies in whether those structures and processes are
promoting behaviours that support effective voice.

Context
Here is where our story begins. A large global financial institution had
invested in a variety of initiatives to promote employee voice over several
years, with mixed success by their own judgement. Conscious of the active
interest of regulators in whether financial organizations understand the
mechanisms through which employee voice is supported, we were engaged
by a global retail and investment bank to help them identify the key factors
in their organization that acted for and against employees speaking up.

To help this organization gain insight into what drove or suppressed

employee voice in their organization, we deployed the CultureScope
behavioural analytics platform. This approach takes a very tangible
approach to understanding culture by focusing on behaviour in the
workplace and applies two lenses to those behaviours by combining data on
the People, individual employees, and the Place, the workplace at the level
of the organization through to the work unit or work team.

How People or individuals behave is likely to make intuitive sense to you,

but you might be curious about the idea of how the Place behaves.
Research in the field of social psychology has long shown that the social
context in which we find ourselves can have a substantial, even overriding
effect on how we behave. Whether it is people acting against their
conscience, conforming to opinions shown objectively to be clearly
erroneous or the condition of learned helplessness, the Place sets a
powerful frame for how the People behave at work.8

You might ask: ‘Isn’t how the Place behaves somehow the sum of the
behaviour of individual employees?’ How the Place behaves reflects
strongly embedded and often obscure beliefs and feelings that develop into
norms for behaviour. Those behavioural norms may have developed
historically, reflecting an inheritance from previously significant figures in
an organization’s development, or may be a factor of the impact of
significant figures, leaders and influential employees.

They may also be a factor of the geographical region in which the

organization is located. Our data across many organizations shows that the
Place can behave in very different ways depending on where you look, so
organizations can be made up of a number of distinct ‘Places’ reflecting
very different expectations of employees.

What we have often seen in our data is a conflict between how the Place
behaves and how People typically behave. That conflict creates
opportunities for misbehaviour in the workplace and frequently derails
efforts to achieve work goals and the strategic aims of organizations.

So, why not explore the beliefs, feelings and attitudes of employees to
understand those norms? We believe that exploring these influences on
employee behaviour does offer value, but we also believe that exploring the
Place in terms of how the People and the Place behave provides a much
more direct and tangible approach to understanding organizational culture
by focusing on observable actions and outcomes.
Like many similar establishments, this organization regularly surveys
employees to gauge employee attitudes and sentiment. Survey questions are
deployed on a quarterly or annual basis and include questions regarding
whether the employee feels able to share opinions and to speak up without
negative consequences (we will refer to these questions as Speak Up) and
believes that there is mutual trust between employees and leaders (Mutual
Trust), and whether line managers take into account the wellbeing of
employees (Wellbeing).9

We tracked the pattern of results from employee surveys for three quarters
after the deployment of CultureScope to develop a predictive model
through two waves of analysis. The first explored the relationship between
the three sets of survey questions to understand the functional relationships
between them. Through testing the fit of various models to the survey data,
we arrived at a model in which Mutual Trust and Wellbeing acted as key
drivers of whether employees reported comfort in speaking up.

Our next step was to develop and evaluate a model that linked the
behaviour of the People and the Place to trust and wellbeing at work, and to
use that model to understand the direct and indirect impacts of the
behaviour of the People and the Place on employee voice.

What we found surprised us and the organization. Finding a relationship

between culture, trust, wellbeing and speaking up was not where the
surprise lay. That surprise was in the strength of the predictive relationships
we found and how that relationship helped to unpack the behavioural
drivers behind employee attitudes in the workplace.

The various waves of this organization’s surveys showed that those

employee attitudes were fairly stable and, on the surface, potentially
reassuring for the organization. As we have already seen, various surveys
and research studies over the past three decades have shown the odds of an
employee speaking up at work have sat stubbornly at around 50:50.

This organization’s surveys showed that, overall, 60 per cent of employees

reported high or very high comfort in speaking up, ten percentage points
above the baseline set by those three decades of wider research. Yet, there
was also a persistent percentage of 16 per cent or around one in six
employees who reported low or very low comfort in speaking up and
sharing their opinions at work and a further 24 per cent, or around one in
four employees, who reported uncertainty about the value of stating their
opinions at work.

This is where using behavioural analytics helped us and the organization

get under the skin of these survey statistics. Here is a summary of what we
found:

That below the surface of those employee survey results lay a very
strong relationship with the behavioural profiles of the People and the
Place, and that relationship was only apparent once the connections
between behavioural diagnostics and employee attitudes had been
surfaced.

That the relationship remained strong up to nine months after the

behavioural analytics had been deployed and explained why employee
sentiment remained so consistent over the time period we explored
with this organization. In short, if the behaviours do not change, then
employee attitudes, including the likelihood of speaking up, are
unlikely to change.

That the relationship was driven by a specific set of common or

complementary behaviours, and that this finding in itself offered the
organization value in narrowing down what they needed to focus on.

That the Place has an overriding impact specifically on levels of

mutual trust, and this is where we found the impact of behaviour to be
around four parts Place to one part People.

Our model showed that the People and the Place operate as drivers of
employee voice both directly and indirectly through mutual trust and
employee wellbeing.

The path model we landed on has been illustrated in Figure 16.1. As you
can see, the Place acts both as a direct driver of employee voice and as an
indirect driver through employee wellbeing and mutual trust.
 FIGURE 16.1 Path model for employee voice for this
organization

A flow diagram shows the path model for employee voice at an

organization.

Figure 16.1 details

When it comes to mutual trust – do I trust my leaders, and do I believe they

trust me? – both the People and the Place have an impact roughly in the
ratio of four parts Place to one part People. Why that ratio? We think that
this is confirmation of the overriding impact that the Place has when it
comes to behaviour in context. This is not about whether People,
employees, naturally trust others, but about whether they trust and feel
trusted by their leaders where they work.

Figures 16.2, 16.3 and 16.4 show the level of the direct impact that the
People and the Place have on wellbeing, mutual trust and employee voice
respectively. What these quadrant figures show are the odds of employees
positively endorsing their wellbeing at work, their sense of mutual trust and
their comfort in speaking up at work. These figures show how the odds
vary for employees reporting high levels of trust (though we refined that by
focusing on one specific trust question as explained below), high levels of
wellbeing at work and high comfort in sharing their voice at work (again
we focused on a specific aspect of employee voice).

FIGURE 16.2 Direct impact of the Place on employee wellbeing

A graph shows the level of the direct impact of the place on

employee wellbeing.

Figure 16.2 details

FIGURE 16.3 Direct impact of the People and the Place on trust

A graph shows the level of the direct impact of mutual trust on

the people and the place.
 Figure 16.3 details

FIGURE 16.4 Direct impact of the Place on employee voice

A graph shows the level of the direct impact of the place on

employee voice.

Figure 16.4 details

For trust and speaking up, the results we shared focus on two specific
questions. Regarding trust, employees reported lower trust in leaders than
the trust they experienced from leaders. As such, it seems that the dynamic
of trust amongst these employees is asymmetric and may reflect differences
in trust with immediate leaders and the broader leadership in the
organization. To boost trust as a driver of employee voice, it seems that
trust in leaders was the key aspect of trust to focus on.

For employee voice, the surveys captured both promotive voice (sharing
ideas and opinions aimed at improving team and organizational
performance) and prohibitive voice (speaking up to stop harm to others and
the organization). The data showed lower endorsement rates among
employees for promotive voice, which was of concern to the organization
at a time when competition in their industry had increased, with the arrival
of challenger banks leveraging new customer service models and
applications of customer service technologies.

While we focus on these specific aspects of trust and employee voice, we

checked our analyses to see whether this focus gave a biased view of the
functioning of our model. It does not, so we feel comfortable sharing the
view of our findings based on these specific questions.

We developed two indices to capture the relative strength of the

behavioural profiles of the People and the Place. In Figure 16.4, employees
have been segmented by broad levels of those indices to create four
quadrants. Essentially, and for the purposes of this case study, we have split
our People and Place indices at the 50th percentile to create a lower and a
higher banding of employees on each index.
The multiplier you see in each of the four quadrants created by splitting our
indices in this way is the ratio of positive endorsements (high or very high
responses to a survey question) to the ratio of negative endorsements (low
or very low responses to a survey question). Those ratios give the odds of
an employee in any quadrant reporting higher trust, higher wellbeing or
higher comfort in speaking up. The higher the multiplier in a quadrant, the
higher the odds of an employee reporting higher trust, higher wellbeing and
higher comfort in sharing their opinions at work.

Our analysis shows that comfort among employees in sharing voice in the
workplace is 7.5 times higher when the Place is stronger and aligned. While
the data shows that there is some comfort in sharing voice even when the
Place is weaker in its support for employee voice, the strength of that
endorsement is exponentially higher when the Place is stronger in those
critical behaviours that drive employee voice.

For the quadrant we have called Virtuous Alignment (we will look at these
more fully in Chapter 18), employees were 13.6 times more likely to report
comfort in sharing opinions. In contrast, the odds for that comfort were 1.8
times for the quadrant we have called the Double Negative, and the ratio of
those two numbers is how we arrive at the 7.5 times higher comfort for
when the Place is stronger and aligned.

Note that the multipliers for employee voice by quadrant are substantially
higher when our index for the Place is stronger (upper two quadrants),
emphasizing the stronger direct impact of the Place on employee voice.
Also note that these multipliers represent the direct impact of the Place on
employee voice and that the Place also has an indirect impact on voice
through employee wellbeing and trust, which we will explore next.

When we look at wellbeing, we see a similar pattern in the impact of the

Place. Employees in our Virtuous Alignment quadrant are more likely to
report higher wellbeing than those in the Double Negative quadrant by 6.7
times.

For wellbeing, the difference between employees in the quadrant we have

called Strong Foundation and the Virtuous Alignment quadrant suggests an
opportunity to develop employees in the behaviours they can utilize to
support their own wellbeing at work.

In other words, rather than look to the Place and specifically the line
manager as the sole driver of employee wellbeing, there is an opportunity
for this organization to explore how employees in the Strong Foundation
quadrant can become more active agents in supporting their own wellbeing
at work. This is a good example of how combining behavioural analytics
and employee survey data helped the organization to understand levers in
the organization for promoting employee wellbeing and, through wellbeing,
the sharing of employee opinion.

Another lever from our model is trust in leadership and here we see a direct
impact of both the People and the Place to the extent that employees in the
Virtuous Alignment quadrant are 17.2 times more likely to report high
comfort in speaking up than employees in the Double Negative quadrant.

The synergistic impact on trust of both the People and Place can been seen
by another comparison among our four quadrants. In contrast to the
quadrant we have called Suppressed Talent, where the odds for reporting
trust in leaders are 4.3 to 1, the odds for employees reporting high trust in
leadership in the Strong Foundation quadrant are 14.7 to 1 – over three
times higher.

The finding for trust clearly shows how the Place can modify and even
suppress the behaviour of employees even when those behaviours
strengthen the opportunity for employees to build trust between themselves
and leaders in the organization.

Is this simply the effect of better leaders demonstrating the behaviours

needed to build trust? In part, the answer to that question is yes. The impact
of good leadership is irrefutable. That said, good leadership needs good
followership to be effective and that is shown by comparing the odds of
reporting higher trust in our quadrants for Virtuous Alignment and Strong
Foundation – the odds of reporting high trust in leadership are almost four
times higher for the Virtuous Alignment quadrant where the Place and the
People are stronger.
So, what were the behaviours that we found underpinned trust, wellbeing
and employee voice and that make up our indices for the People and the
Place? Figure 16.5 summarizes the six behaviours from the CultureScope
framework that, between the People and the Place, we found drive higher
employee trust in leaders, higher employee wellbeing and substantially
higher levels of comfort in speaking up and sharing opinions and ideas at
work.

FIGURE 16.5 The six behaviours that drive trust, wellbeing and
employee voice

An illustration with two parts shows a summary of the six

behaviours from the CultureScope framework.

Figure 16.5 details

Two of the behaviours are common and one, Expressive, will seem
immediately intuitive since the willingness among employees to share
opinions and the encouragement of the workplace to do so is an obvious
element in supporting the sharing of employee voice.

The other common behaviour, Active Learning, may seem a little less
intuitive but a reading of the research literature shows that obtaining value
from encouraging employee voice does depend to a large extent on whether
that voice leads to more effective ways to deal with issues and problems,
better ways of working and meeting stakeholder needs, and diffusion of
effective working practices across the organization. Those benefits point to
learning and a key question for any organization wanting to build and
sustain effective employee voice is whether the people they recruit and
promote are active learners, and whether the workplace encourages active
learning through experimentation and managed risk.

Four of our six behaviours operate as complementary pairs; again, one pair
is highly intuitive in that voice is more likely to be received positively
when it is shared in a way that is seen to be focused on building and
strengthening partnerships within the workplace rather than just for the
benefit of the individual sharing their voice.
This is how we see Collective operating in our predictive model and how it
plays through the actions of leaders to build trust among employees. That
behaviour from the perspective of the People is complemented and
strengthened in workplaces by demonstrating an awareness of the impact of
actions on others when executing tasks and achieving objectives, which the
CultureScope framework captures through People Focus.

Our other complementary pair can be understood if we turn to trust in the

workplace where we found the behaviour of both the People and the Place
had a synergistic impact. Trust depends on expectations of others and
whether others can be expected to act in ways that are predictable. That
view of trust in part explains the contribution of Moderate from the
perspective of the People – how employees behave through their expression
of voice – and Conformity from the perspective of the Place.

Conformity in the CultureScope framework does not just talk to playing by

the rules, though that is important in employees having faith that there is
justice in the workplace.10 It also speaks to the manner in which leaders
receive employee voice, and we believe that goes some way to explaining
the higher levels of trust in leaders that we saw in this organization’s data
where Conformity in the Place was higher.

So, where is this organization on their journey to strengthen employee

voice and to make that voice more effective in driving organizational
transformation and change? Armed with a clear and manageable focus on
six behaviours, they have started a series of cascading conversations
through their organization framed by those six behaviours. Those
conversations are not just about how to have better conversations in the
workplace. They include HR processes such as performance appraisals to
ensure that those key behaviours are being reinforced through one of the
key forums through which the People and the Place interact.

Those conversations also include organizational structures and workflows

to examine where the form of the organization is getting in the way of the
function of employee voice.

Through further segmentation of the data and using the six behavioural
clusters as shown in Figure 16.5, we used the quadrant analysis method to
know ‘where’ in the organization these behaviours are present and where
absent. Teams within the Virtuous Alignment quadrant have these key
behaviours embedded and the outcome has manifested. Hence they can
learn from teams in the other three remaining quadrants where their
investment needs to happen, and by virtue of the different quadrants they
will know if the interventions should be focused on the People (capability-
type intervention) or the Place (opportunity- and motivation-type
interventions), or both.

How we crunched the numbers

Our data sets were taken from a CultureScope behavioural diagnostic
deployment in the last calendar quarter of 2020 and four waves of client
employee surveys that occurred quarterly from the end of 2019 through to
the end of 2020. While a large portion of survey questions appeared in all
four surveys, some appeared only annually. Our sample sizes were 9,121
for CultureScope measures and 6,912 for employee surveys across all four
quarters.

Quarterly employee attitude surveys were based on stratified random

sampling across the organization by level (status), geography and business
function. Employees responding to each quarterly attitude survey varied
quarter on quarter, with only a small percentage of employees completing
the survey more than once.

To manage this, we conducted a series of analyses within each quarter as

well as with data aggregated across all calendar quarters for which data
were available. For those employees sampled more than once, we used the
average of their survey responses to each question.

At the aggregated data level, maximum likelihood factor analysis with

oblique rotation of the five survey questions – two for trust, two for
speaking up and one for wellbeing – yielded two factors that offered a good
fit to the data. Those factors were defined by the trust questions and the
wellbeing question with cross-loadings for the two speak up questions.
This suggested to us a model in which speaking up (employee voice) was
the dependent or outcome variable while trust and wellbeing operated as
causal factors or independent variables influencing whether employees felt
comfortable in sharing opinions or speaking up when they saw something
wrong.

We tested this through a series of path models, rotating each of the

attitudinal variables – trust, wellbeing and speak up – as dependent
variables. The model in which speaking up served as the dependent
variable offered a stronger fit than when wellbeing served as the dependent
variable, and equivalent fit when trust served as the dependent variable.

We then used a series of hierarchical regressions to explore the

relationships between CultureScope behavioural measures for the People
and the Place and trust, wellbeing and speaking up. Again, we explored
these models within each calendar quarter and for the data aggregated
across calendar quarters. We used a fully saturated model with all 15
CultureScope dimensions as our baseline to test more restricted and nested
models with fewer CultureScope dimensions to arrive at our final predictor
set of six CultureScope behaviours.

The path model presented was also tested using path models analysis and
yielded the equivalent to a Multiple R of 0.50 (accounting for 25 per cent
of the variance in the data) in predicting employee voice.

The effect size obtained from our model in predicting employee attitudes
for nine months outranks at the 80th percentile of effect sizes observed in
research on employee attitudes.11

Notes
1 M Bazigos, A De Smet and C Gagnon, Why agility pays, McKinsey,
1 December 2015, www.mckinsey.com/capabilities/people-and-
organizational-performance/our-insights/why-agility-pays (archived at
https://perma.cc/GV4F-A4GB). This research was based on 1,000
companies and explored the relationship between 37 management
practices and organizational agility, defined as the ability of an
organization to adjust to changes and new ways of doing things.
Associated research had shown that organizational agility is related to
the share value and EBITDA of for-profit organizations. The other
three management practices identified alongside the involvement of
employees in shaping organizational vision were financial
management, financial incentives and capturing external ideas.

2 W A Kahn (1990) Psychological conditions of personal engagement

and disengagement at work, Academy of Management Journal, 33, pp.
692–724

3 This statistic was drawn from a number of sources, including:

Ipsos (2012) Mental well-being in the workplace. Cited by M L

Frazier, S Fainshmidt, R L Klinger, A Pezeshkan and V Vracheva
(2017) Psychological safety: A meta-analytic review and
extension, Personnel Psychology, 70, pp. 113–65

M Kirrane, D O’Shea, F Buckley, A Grazi and J Prout (2017)

Investigating the role of discrete emotions in silence versus
speaking up, Journal of Occupational and Organizational
Psychology, 90, pp. 354–78

F J Milliken, E W Morrison and P F Hewlin (2003) An

exploratory study of employee silence: Issues that employees
don’t communicate upwards and why, Journal of Managerial
Studies, 40, pp. 1453–76

K D Ryan and D K Oestreich (1991) Driving Fear Out of the

Workplace: How to overcome the invisible barriers to quality,
productivity and innovation, Jossey-Bass, San Francisco

Towers Watson (2014) Employee Voice: Releasing voice for

sustainable business success, Towers Watson, London

4 The distinction between employee voice and employee silence is a

topic of debate amongst academics in this area. Some such as
Morrison see voice and silence as opposites on a continuum. Others
such as Van Dyne see voice and silence as distinct behaviours driven
by similar but distinct employee motivations. For more on this debate,
see:

E W Morrison (2011). Employee voice behaviour: Integration

and directions for future research, The Academy of Management
Annals, 5, pp. 373–412

E W Morrison (2014). Employee voice and silence, Annual

Review of Organizational Psychology and Organizational
Behavior, 1, pp. 173–97

L Van Dyne, S Ang and I C Botero (2003). Conceptualizing

employee silence and employee voice as multidimensional
constructs, Journal of Management Studies, 40, 1359–1392

5 Financial Conduct Authority. Transforming culture in financial

services, March 2018, www.fca.org.uk/publication/discussion/dp18-
02.pdf (archived at https://perma.cc/2TSQ-6CS8)

6 For examples of research on the antecedents of employee voice, see:

M A Chamberlain, D W Netwon and J A LePine (2017) A meta-

analysis of voice and its promotive and prohibitive forms:
Identification of key associations, distinctions, and future
research directions, Personnel Psychology, 70, pp. 11–71

M L Frazier, S Fainshmidt, R L Klinger, A Pezeshkan and V

Vracheva (2017) Psychological safety: A meta-analytic review
and extension, Personnel Psychology, 70, pp. 113–65

J A LePine and L V Van Dyne (1998) Predicting voice behaviour

in work groups, Journal of Applied Psychology, 83, pp. 853–68

7 ‘Form follows function’ is an axiom attributed to the architect Louis

Sullivan and states that the purpose of a building should be the starting
point for its design. We borrow from this axiom to emphasize that the
purpose of any processes and structures implemented by an
organization to encourage employee voice should focus on the
behaviours that support effective employee voice and should be
evaluated in terms of whether a process or a structure is encouraging
those behaviours.

8 Key researchers and research in this area include:

S E Asch (1956) Studies of independence and conformity: A

minority of one against a unanimous majority, Psychological
Monographs, 70, pp. 1–70

S Milgram (1963) Behavioral study of obedience, Journal of

Abnormal and Social Psychology, 67, pp. 371–78

M E P Seligman (2006) Learned Optimism: How to change your

mind and your life, Vintage Books, New York

9 One limitation of this project was the framing of the survey question
asked of employees about their wellbeing at work. The specific
question used asks whether the manager is invested in the employee’s
wellbeing rather than the wellbeing the employee generally
experiences at work. However, we feel that it is a reasonable inference
that employee wellbeing is correlated with the extent to which an
organization, through its managers, is concerned about and invested in
employee wellbeing.

10 Models of organizational justice show that perceptions of justice in

the workplace depends on whether employees see that process by
which decisions are reached are fair (procedural justice), whether the
distribution of rewards is fair (distributive justice), whether the
involvement of people is fair (relationship justice) and whether the
sharing of information is fair and reasonable (informational justice).

11 Many researchers use Jacob Cohen’s correlations effect size

benchmarks to gauge the strength of relationships. Those benchmarks
are 0.2 for a small effect, 0.3 for moderate effect and 0.5 for a large
effect. The results of our model for predicting employee voice would
benchmark as a large effect size (R=0.496). More recent research
suggests that the size of an effect varies depending on what is being
studied and that different benchmarks reflecting different fields of
study are required. We obtained our 80th percentile benchmark for
research on employee attitudes from F A Bosco, H Aguinis, H, K
Singh, J G Field and C A Pierce (2015) Correlational effect size
benchmarks, Journal of Applied Psychology, 100, pp. 431–49
17
Driving organizational change
through workplace culture and
behavioural analytics
Addressing the root causes of why
change programmes fail
Transformation has become the new buzzword, replacing the
perhaps less dramatic word ‘change’. Organizations use the two
terms interchangeably, but are they really the same? When it
comes to change versus transformation, what are the
differences? I am sure to spark a big debate here, but I will
simply offer my opinion: change can be seen as incremental,
perhaps driven by a response to a situational event or events.
However, transformation implies a drastic shift from one state
to another. Naturally, transformation could include multiple
and significant changes. It could be the difference between
evolution and revolution.
This case study will be in three parts as it will address the
most common reasons why organizational change programmes,
often associated with transformation, fail.
The first reason is simply not knowing which behaviours are
required as an ally to achieve desired outcomes. Most
organizations have no problem defining their destination, but
few have the tools to identify and implement the corresponding
behaviours required to attain their desired destination.
Indeed, identifying your destination is the first step.
However, you will need to identify and focus on the workplace
culture journey and lead drives to ignite change.
The second reason is failing to tailor the change approach. As
we have already shown in Part 1 of this book, each
organizational culture is unique; going even further with this
thought, different people perceive and react to change in
different ways and the change you are implementing is going to
mean different things to different people and teams. Many
organizations cascade one-size-fits-all approaches that some
buy into, but many resist.
The third reason is relying on assumptions, rather than a
data-driven approach. Often, we see transformation strategies
based on abstract theories rather than data-driven
interventions.
So, before you embark on your transformation journey,
gather precise data to ensure that any culture and behavioural
intervention is tailored to the unique context and capability
needs of your organization. If not, you might be aiming at
outcomes that are irrelevant, distracting or even harmful to
your organization.
In this three-part case, I will focus on the behavioural aspects
of why culture change fails and how to increase your chances
for success.
Part 1: Do your change champions have what it
really takes to effect change?
The statistics have remained stubbornly negative for decades.
Whether it is the bleaker 70 per cent of change programmes
failing to achieve their objectives1 or the more benign statistic
of 50 per cent success for Forbes 1,000 companies2, at best that
gives change programmes a one in two chance of succeeding.
Change has become the new normal, alongside increased
economic and social uncertainty as well as the growing impact
of technology. Yet, the question that continues to plague leaders,
managers and those that support organizations through change
is ‘Why are the odds stacked so heavily against change
programmes succeeding?’
More specifically, and given that the problem is not that
organizations are incapable of launching change programmes,
why is it that change programmes so often fail to land
effectively and deliver the outcomes they were launched to
deliver?
The prevailing wisdom is that change programmes must have
a compelling purpose for change, must be supported by
effective processes for managing the programmes and must
have clear commitment and support from organizational
leaders.3 But we have known all of these must-haves for
decades, and yet those damning statistics on the failure of
change programmes remain negative or 50:50 at best.
What is new is the growing interest in exploring the
application of behavioural psychology to improve the success of
change programmes.4 This is the focus of this section. Though
the application of behavioural psychology to change
programmes is still at a relatively early stage, we explore how
behavioural psychology allied with data analytics addresses
three of the root causes for why many change programmes fail.

Why workplace culture change programmes fail

In essence, change programmes are about changing behaviour,
whether that is how an organization functions once new
structures and processes have been introduced or how
individuals and teams interact among themselves and with
those outside the organization. It is about changing how the
workplace behaves and how employees behave.
With behavioural change in mind, our experience shows that
there are three behaviourally related reasons for why change
programmes fail. We do not claim that these are the only
reasons, but they do point to common blind spots that all too
often undo the well-intentioned and significant efforts devoted
to change programmes. Those three reasons stem from a lack of
knowledge about the organization’s people and how the
workplace influences employee behaviour.
When we talk in this chapter about the People, we mean the
deep-seated behavioural profiles of employees at any level and
in any location, and those behavioural profiles include their
appetite for personal change. Yet, few organizations have
detailed insight and data on the behavioural profiles of their
employees, and this includes the ones they nominate or who
offer themselves to be champions for a change programme.
So, the first reason that we see for culture change
programmes failing is that organizations either do not know or
lack insight into whether their change champions have what it
takes to be effective in delivering change programme
objectives. While it seems an obvious factor to consider, it
remains a significant blind spot for most organizations.
The second reason that our experience has shown for why
programmes fail is a lack of insight into the size of the change
task – the ‘ask’ – that those programmes demand. From a
behavioural perspective, improving the odds of achieving a goal
depends partly on how difficult challenging that goal is, and
partly on the context in which that goal is set.
This is where insight on the Place (our shorthand for the
workplace) or, to be more accurate, the Places that make up the
organization becomes important. The question here is not the
natural behavioural profiles of the People that occupy those
Places, but how those Places act to frame and influence
employee behaviour. As research in social psychology has
shown, the Place exerts a strong influence on how we behave,
so strong that the Place can override how a person would
naturally behave outside that work context.5
One way in which a lack of data and insight into the Place
undermines the success of change programmes is in the setting
of common change objectives for the entire organization. That
supposes that every business function, location, team and
individual employee is facing the same challenge in achieving
those change goals, whether that is the specifics of what needs
to change, or the size of the change needed to achieve the goal
of the change programmes. Yet, we understand that
organizations are not that homogenous, just as their target
clients or customers and the markets that different parts of the
organization serve are also rarely homogenous.
This mistake of adopting a one-size-fits-all approach plays out
in at least two ways. It may be that change targets are imposed
on parts of the organization where the gap between where the
organization wants to go behaviourally and where it is today is
either small or non-existent. These are parts of the organization
that are already where the change programme wants to get to.
These are also potential hubs of good practice that may be
hidden from those designing change programmes and that offer
resources that could support learning elsewhere in the
organization. But that potential is all too frequently squandered
through a one-size-fits-all approach to change.
Conversely, there may be parts of the organization where
there is a substantial behavioural gap between how the Place
behaves today and where the organization wants to be
tomorrow. Closing that behavioural gap at the same pace as
other parts of the organization may well be, and in our
experience often is, too much of an ask.
The third reason we see for behavioural programmes failing
is a lack of a behavioural roadmap and, specifically, a data-
driven behavioural roadmap. We are not saying that the
judgement and intuitions of leaders, managers and employees
do not count, but we are saying that those judgements and
intuitions are best supported by valid and objective data ahead
of launching a change programme.
So, what is our notion of a behavioural roadmap? Essentially,
that roadmap combines data of the People and the Place to
identify where in the organization the People are more likely to
embrace change and deliver the goals of the organization with
data on the Place in terms of how big an ask is being made and
where specifically different parts of the organization should
focus. It also serves to identify where in the organization the
desired behavioural profile may already exist, and how the
resources for learning can be marshalled to support the odds of
change programmes succeeding.
We will tackle each of these three reasons that we see for
change programmes failing in this chapter. We will come back
to how to size the behavioural ask of employees and how to
construct a behavioural roadmap in the second and third parts.
For the remainder of this first part, we will address the
question: ‘Do your change champions have what it really takes
to deliver change effectively?’

A metric for whether employees are really up for

change
Just how willing are employees to change their behaviours?
That was the question that emerged from discussions with this
large organization in the financial sector preparing to launch a
major change programme. The CultureScope behavioural
diagnostic was deployed, and the organization agreed that we
add a simple sentiment type question to the diagnostics and
asked respondents to tell us how willing they were to change
how they as individuals behaved.
Only 20.3 per cent, or around one in five, employees across
the organization responded positively. With 79.7 per cent saying
no to the question, that suggested that the starting odds in
terms of employee buy-in to changing their own behaviours
were a resounding 4:1 against. That also suggested that the
change programme the organization was about to embark on
was a poor bet in terms of success.
The question that we posed was that, given that the
disposition to change of individuals is more of a continuum
than a binary change vs no-change condition, could we identify
the behavioural profiles of those more open to personal change
and develop a metric to help this organization understand what
to look for in those who were more vs less likely to commit to
change?
To develop that metric, we used the data on the People (those
deeply ingrained behaviours we all have as individuals)
provided by CultureScope. We will come to the specific
behaviours that drove that metric and what that metric offered
the organization a little later in this chapter.
Figure 17.1 shows how that metric differentiated the
disposition to personal change in significantly more detail than
the binary yes/no question initially posed by the organization.
 FIGURE 17.1 Behaviours metrics versus probability of
change

NOTE *20 per cent of sample responded yes

Figure 17.1 details

What are you seeing here? Well, we know that the base rate
(the yes answers to that question) is around 20 per cent or one
in five employees. But that base rate masks an underlying
metric that clearly distinguishes between very low disposition
to personal change (0 on the horizontal axis) and a very high
disposition to personal change (5 on that horizontal axis).
What Figure 17.1 also shows is that those who are highest on
the underlying metric (5) are 23 times more likely to commit to
behavioural change than those lowest on that underlying
metric (0).
 37 per cent of employees were identified as showing positive
likelihoods of being able to change their own behaviours (3 and
above on the metric). That still showed that the odds in favour
of employees having a positive disposition to personal change
were 1.7:1 against, but that was a significant improvement on
odds of 4:1 against.
The good news for the organization was that those
behavioural profiles could be identified and the metric could be
used to identify not only those with a substantially higher
likelihood of being effective change champions, but also to
identify those who, with the right support from those change
champions, would be able to embrace the change programme.

The behavioural profiles driving that metric

Our modelling showed that there were eight behavioural
factors that, together and either positively or negatively, drove
that metric. Figure 17.2 summarizes those behavioural factors.6
 FIGURE 17.2 The behaviours for the change metric

Figure 17.2 details

Those employees who were more likely to embrace behavioural
change and to commit to change programmes, were those who
were more likely to focus on the impacts of change on people,
as well as how change programme goals could be incorporated
in the way teams work more effectively together. They were
also those who were more likely to adapt their approach to the
demands of different situations and challenges, and were more
likely to share their opinions and views on how best to
approach change and whether change programmes were
achieving their goals or not.
In contrast, those who were less likely to embrace
behavioural change and to commit to change programmes were
those with a stronger tendency to focus on process for the sake
of process rather than the impacts on other employees. They
were also more likely to focus on what change programmes
offered for them as individuals and were also less likely to voice
their opinions on whether change programmes were achieving
their goals or not.
Putting the metric into practice
Unpacking these behavioural profiles offers value in two ways.
One flows from data analytics, the other flows from providing
more qualitative insight to change programme leaders as well
as equipping change champions to deliver on change
programme goals.
From a data analytics perspective, the metric enables further
analysis by employee segment (seniority, tenure, gender and
age) as well as by business function, department, team and
location to identify where effective change champions are more
likely to be found in the organization. Simply put, why look to
find champions where they are less likely to be committed to
personal change and to the change programme, particularly
when the starting odds are 4x against any employee selected at
random fitting the bill of an effective change champion?
At a more qualitative level, the behavioural profiles pose
clear questions that need to be addressed to ensure that change
initiatives are being landed. For example, how can we promote
employee voice to flag issues and course correct as change
programmes unfold? How can we promote team goals that
support change programme goals and reward and recognize
team rather than individual successes in achieving change
goals? How do we frame change programme communications
to ensure that they reflect the people aspect of change as well as
the achievement of process and structural improvements?

How we crunched the numbers

The question about disposition to change was framed as a
closed ‘yes’ or ‘no’ question. Given that the response was binary,
we used logistic regression to identify which behavioural
dimensions predicted whether employees were more likely to
respond positively and with a yes.
Logistic regression was developed to predict binary
outcomes. Logistic regression models serve to identify which of
a set of non-binary and binary predictors are most strongly
related to an outcome such as a ‘yes’ response in this particular
application. The result of a logistic regression provides the
building blocks for constructing metrics that measure the
likelihood of the outcome being observed (e.g. a ‘yes’ response
to the personal change question). Those metrics can then be
used to predict the odds or probabilities of that outcome being
observed.
We tested a number of nested models to arrive at the most
efficient model that offered the strongest predictions of a ‘yes’
response to the question posed to employees of whether they
would be willing to change their behaviours. The strongest
predictors of that ‘yes’ response formed the basis of the
behavioural profile described in the main body of this piece.
In terms of model fit and for those more familiar with linear
methods such as correlations and least squares regression
models, two indicators of fit allied to those obtained from linear
methods are available from logistic regression – as explained by
Cox and Snell.7 Taking the square root of these fit indices yields
indicators that approximate the multiple R obtained from
linear regression or the correlation between two variables.
The multiple Rs associated with our final model for predicting
a yes response to the survey question from behavioural factors
were 0.43 (Cox & Snell) and 0.53 (Nagelkerke). The Place size of
the relationships found in our model at or above the 75th
percentile for research into employee attitudes and intentions
such as commitment to change programmes.8
Part 2: Sizing the ‘ask’ of your organization when
you ask it to change
In this part, we turn to the work environment, or Place, and
how to use behavioural diagnostics to understand the size of
the challenge that a change programme may present in asking
the organization to operate and behave differently.

The growing realization that effective change

means understanding the People and the Place
Recent research published by McKinsey shows that using
behavioural diagnostics to guide change programmes delivers a
fourfold increase in their odds for success.9 That research also
shows that behavioural diagnostics is not just about
understanding the behaviours of employees. The success of
change programmes also requires an understanding of how the
workplace influences employee behaviour.
Why? The work context or Place influences how employees
act by providing a strong set of stimuli which frames the
behaviour of employees. Employees tend to act to satisfy what
they see as the expectations of the Place even if those
expectations go against how they would normally behave in
non-work settings.
Another benefit from using behavioural diagnostics to guide
change programmes flows from the simple axiom that to know
where you are going, you need to know where you are starting
from. That starting point might be very different across an
organization. To borrow from a sports analogy, while
organizational leaders might see change as a 200-metre sprint,
it may feel more like a long-distance haul for some parts of the
organization while other parts of the organization might
actually be confounded by the need for a change programme as
they see themselves as already standing at the finishing line.
This variation in where different parts of an organization
start their change journeys also points to an all-too-often-
missed opportunity to proactively identify where the desired
behaviours may already exist, offering opportunities to tap into
effective behaviours that already exist in the organization to
promote learning and support. This is not just setting up groups
through which experiences can be shared and diffused. This is
about clarity in terms of specific strengths and weaknesses in
different parts of the organization and using those specifics to
design tangible and effective exchanges between employees
and workgroups targeted on specific behaviours.

A case in point
I will briefly describe a project with this organization in the
financial sector to illustrate the power of behavioural insights
into the Place. This organization was embarking on a change
programme focused on strengthening three key pillars of its
organizational culture – Customer Focus, Simple is Best and
Succeeding Together.10 They wanted to ensure they had a clear
understanding of where they were, their start position, and
commissioned behavioural diagnostics using CultureScope to
gain that understanding.
Delivering actionable insight from behavioural diagnostics
requires more than just a survey and analytics know-how; it
also requires a deep understanding of the context. To gain that
understanding, I ran a series of workshops with senior
members from the organization using the CultureScope
behavioural taxonomy of 30 factors.11
The process for the workshops began with an open-ended
question: ‘How would you know that Customer Focus, Simple is
Best and Succeeding Together were being truly lived in the
organization?’ To surface the specific organizational context,
debate and even disagreement were encouraged while
maintaining appropriate respect for the opinions of others. The
workshops then progressed to mapping the 30 CultureScope
behavioural factors to each of the three organizational pillars to
establish tangible behavioural markers for the success of the
organization’s change programme.12
Those mappings framed how data from the CultureScope
diagnostic were used to create metrics and establish where the
Place was acting to promote desired behaviours and where the
Place was acting against those desired behaviours.13 So, what
did the data show?

How variation in behavioural profiles of the Place

provides insight into the ask of change programmes
While organizations strive to promote consistent values and
behaviours, our experience shows again and again that the
‘Place’ may actually promote very different employee
experiences depending on where employees work. This was the
case with this organization.
We identified eight clusters from the data on how employees
saw their workplace. These eight clusters were obtained with
the help of K-means clustering using indices to measure the
between-cluster differences as well as within-cluster variation.
The between-cluster differences indicate the extent to which
the clusters represent distinct profiles across the three pillars.
The within-cluster variation indicates the extent to which the
overlap between clusters is low and, therefore, the extent to
which the boundaries between clusters are distinct.
As a quick introduction, K-means clustering is a type of
unsupervised learning, which is used when you have
unlabelled data (i.e. data without defined categories or groups).
The goal of this algorithm is to find groups in the data, with the
number of groups represented by the variable K. The algorithm
works iteratively to assign each data point to one of K groups
based on the features that are provided. Data points are
clustered based on feature similarity. The results of the K-
means clustering algorithm are:
the centroids of the K clusters, which can be used to label
new data
labels for the training data (each data point is assigned to
a single cluster)
Rather than defining groups before looking at the data,
clustering allows you to find and analyse the groups that have
formed organically.
We ran a series of cluster analyses ranging from two to 12
clusters; all cluster solutions bar one showed significant
differences for metrics for the three pillars between clusters.
The within-cluster variation reduced with increasing number of
clusters and reached a lower limit at eight clusters. As such,
eight clusters were chosen as the most parsimonious solution
meeting the criterion for differentiating behavioural profiles
with lower overlap in cluster membership.
 Using those clusters, we could show the organization how
much variation there was in terms of the starting point for
different groups of employees in their change journey.14
Figure 17.3 shows the variation for behavioural alignment for
the three change pillars. The data demonstrated that a one size-
fits-all approach to change was less likely to succeed. As we will
see a little later, a given cluster might show strengths in some
areas and need for improvement in others. Adopting a one-size-
fits-all approach would have ignored the nuances in the cluster
profiles as well as the simple fact that all three pillars act
together to define the context in which any given group of
employees work.
The organization saw that a more tailored approach offered a
better opportunity to guide change based on the differences in
the starting points for different clusters and specifics on what
behaviours those clusters should focus on.
FIGURE 17.3 The three organizational change pillars
versus behavioural alignment
 Figure 17.3 details
Figure 17.4 shows the resulting profiles of the eight clusters and
why a deeper dive into the data supported a more nuanced and
tailored approach to launching and supporting the change
programme.
 FIGURE 17.4 The profiles of the eight clusters and the
alignment by pillar

Figure 17.4 details

That deeper dive into the data also showed that there were
parts of the organization where strong alignment with the goals
of the change programme already existed and where stronger
alignment could be leveraged to support learning at the very
earliest stages of the change journey.

Using behavioural diagnostics to support and

strengthen the learning
Let’s explore those eight clusters in Figure 17.4 in more detail.
The left side of Figure 17.4 shows a ranking of the clusters
relative to their general strength of alignment across the three
pillars of Customer Focus, Simple is Best and Succeeding
Together, the focus of the change programme. This takes into
account where each cluster stands across the three pillars and
shows the highs and lows in terms of alignment for each
cluster.15
 The bar charts to the right of Figure 17.4 shows how each
cluster scored on each pillar in terms of alignment and conveys
the start positions for each cluster. Clearly, the ask of Cluster 6
and Cluster 2 is substantial. In contrast, Cluster 8 and Cluster 5
show strong alignment, suggesting that these clusters offer
potential for tapping into tangible day-to-day experiences that
are benefitting the organization today and that can be
leveraged to the benefit of other parts of the organization in
supporting stronger alignment for tomorrow.
The clusters identified from behavioural data on the Place
provide a frame for data analytics to identify which Places (e.g.
business function, location, team) are likely to show stronger or
weaker alignment with the change programme’s goals, and
answer the question of where the ask of the change programme
is more achievable in the shorter term and where meeting that
ask is more likely to be a longer-term effort.
I am not suggesting that those clusters showing weaker
alignment and a bigger ask should be ignored. However, in the
immediate term, these may represent Places within the
organization that pose a risk to current performance in the
organization. They clearly need to be monitored.
What I am saying is that a failure to recognize these as Places
where the ask is considerably more challenging is also a failure
to proactively address potential risks to the success of a change
programme.

Understanding the ask of change is a key factor but

not the only factor to consider
We have focused on how behavioural diagnostics of the Place
can help an organization to understand the ask demanded by
change programmes and provide insight to enable an
organization to think ahead of the launch of a change
programme by knowing where that programme is more likely
to gain traction and where that programme is more likely to
struggle.
My experience tells me that data on the Place is not enough to
build a robust plan of action that supports the success of change
programmes. Building that robust plan also required data on
the People and their appetite for change – as I have already
demonstrated in Part 1.
Bringing together behavioural diagnostics on the People and
the Place is what we will explore in Part 3 of this case study.
Part 3: Building a behavioural roadmap to drive
effective change
Let’s address how organizations can use cultural diagnostics
and behavioural analytics to avoid failure of change
programmes, the unfortunate outcome that decades of research
documents as the fate of those programmes.
In my view and that of others,16 behavioural analytics is
emerging as a critical tool in enabling organizations to improve
the odds that their change programmes will deliver. In Part 1,
we explored how behavioural analytics on the people in the
organization helps organizations to understand which
employees have what it takes to embrace and respond
positively to change. In Part 2, we explored how behavioural
analytics on the work environment helps organizations to
understand the size of the challenge that change programmes
pose in asking an organization to operate and behave
differently.
To create impact, I must show how insights into the People
and the Place can be combined to develop a behavioural
roadmap for proactively planning and managing change. That
behavioural roadmap not only gives organizations a better
means of implementing change strategy and tactics, but also the
means to promote the organizational learning that serves to
strengthen the achievement of change programme goals.
But how do you persuade people to engage in learning and to
change their behaviour? Or, more specifically, how do you
design programmes and interventions that increase the
likelihood that a desired and beneficial behaviour will be
adopted and sustained? These questions have led to the
emergence of a new thinking in the field of design, namely
persuasive design which will target capability building.
The principles that underpin persuasive design represent a
mode of thinking that those charged with designing and
implementing change programmes would do well to adopt.
Framed by a roadmap developed using behavioural analytics,
we see this mode of thinking as a step change in how
organizations approach the challenge of change. So, let’s briefly
explore those principles of persuasive design before we move
onto a case study that describes the application of a
behavioural roadmap for guiding change.

Applying the principles of persuasive design to

organizational change programmes
The first step in applying persuasive design is to identify a
target behaviour. This is no different to the first step in
designing a change programme. Whether it is the introduction
of new systems and practices or improving the performance of
existing systems and practices, success lies in changing People’s
behaviour and setting and reinforcing new norms for that
behaviour. That, in turn, requires a clear, tangible and
commonly understood definition of the target behaviour.17 We
gave an example in Part 2 of how an organization used cultural
diagnostics, specifically CultureScope, to define the target
behaviours they set as goals for their change programme.
Once that target behaviour is identified, the next step in
persuasive design is to understand what might prevent people
from performing the target behaviour. Is it a lack of motivation
– are people unwilling or simply uninterested in adopting the
target behaviour? Is it a lack of capability or ability – do people
lack the means or resources to adopt the target behaviour, or,
correctly or not, see the target behaviour as too ambitious for
them to perform? Or is it a lack of a well-timed trigger to
perform the behaviour (the opportunity) – is the organization
asking the wrong people at the wrong time to perform the
target behaviour?
Persuasive design advocates clear thinking about the goals set
for people, as well as how those goals improve motivation and
promote the sense that people are able to achieve them. Much
of that effort is centred on promoting incremental learning and,
with learning as a central focus, persuasive design advocates an
incremental approach to behavioural change.
Applied to change programmes and the level of individual
and groups of employees such as work teams, persuasive design
advocates for a careful sizing of the goals set for individuals
and teams, whilst also providing a clear understanding of their
motivation for change and their ability to achieve the change
that the organization is asking for. At the programme level,
persuasive design advocates for smaller-scale experiments
conducted in targeted parts of the organization and where the
opportunities for learning are stronger. Ultimately, the success
of change programmes depends on knowing where in the
organization work practices are stronger and exploiting
opportunities for the diffusion of those stronger work practices
across the wider organization.18
Whilst I see the thinking represented in persuasive design as
offering a step change in how organizations approach change,
implementing that thinking raises a challenge and a question –
how do you identify where in the organization those
opportunities for learning are stronger? This is where the
behavioural roadmap that we will describe next comes into
play.
In order to understand where those learning opportunities
are more likely to be stronger, we have argued that you need to
have two pieces of intelligence on an organization. First, you
need to know which employees are more likely to be more open
to change and more likely to adopt the behaviours a change
programme seeks to promote.19 Through the lens of persuasive
design, addressing the question of ‘who’ through diagnostics on
the People addresses the issue of motivation.
Second, you need to know how big an ask a change
programme is making of the organization and, more
specifically, where in the organization that ask is more likely to
be achievable and where it isn’t. Through the lens of persuasive
design, addressing the question of ‘what’ through diagnostics on
the Place addresses the issue of ability.
That, from the perspective of persuasive design, leaves the
issue of well-timed triggers and, from a change programme
perspective, that raises the question of where an organization
should think about starting their change journey and where the
opportunities for learning are strongest.
This is where a behavioural roadmap offers value to change
programmes. By combining insight into the what (the Place)
and the who (the People), organizations can have a much better
stab at identifying where they are likely to yield better returns
on starting their change journey. So, let’s explore what we mean
by a behavioural roadmap for guiding change.
Building a behavioural roadmap from insights on
the People and on the Place
Let’s briefly reprise where we landed in Part 1 and Part 2. In
Part 1, I described how we built an index that measured
employee propensity for change. Those employees who ranked
highest on whether they would voice an opinion, be more
flexible, be more people- and team-focused were 3.5 times more
likely to champion change than the average employee,
irrespective of whether those employees are afforded the title
of change champion. That behavioural index defines the
horizontal axis of our behavioural roadmap and speaks to the
inherent motivation of employees to engage with change
positively.
In Part 2, I showed how analytics applied to the Place
surfaced eight clusters or profiles in terms of the gap between
where they were at the start of the change process and where
the change programme wanted the organization to be. This data
defines the behavioural roadmap and speaks to the size of the
challenge that different parts of the organization face in
adopting and demonstrating the behaviours promoted by the
change programme.
Figure 17.5 summarizes how combining insight into the who
(the People) and the what (the Place) helps determine where
opportunities to start the change journey lie in an organization.
The question of where is answered in terms of where change is
more or less feasible, and feasibility is driven taking into
account the size of the ask from change across the organization
and combining that insight with knowledge of the propensity of
employees to engage positively with change.
 FIGURE 17.5 Combining the insight of the Place
defining the size of the ask vs the People defining the
propensity for change

Figure 17.5 details

Back to our case study, where the organization had to come to
the realization that effective change is dependent on learning,
that an incremental approach was required to enable the
learning, and looked to behavioural analytics to provide the
intelligence it needed to tell it where in the organization the
change journey was more feasible and where starting that
journey could result in failure.

Putting a behavioural roadmap to work

Let’s see what the starting position for that organization looked
like when we mapped the organization, by combining analytics
on the People and the Place. Note that once the map is defined,
we can dig deeper into each cluster and segment by employee
demographic to understand what membership of each cluster
represents in terms of employee age, gender, tenure, job level,
business function, work team and location.
 FIGURE 17.6 The eight-cluster mapping analysis
within the size of the ask (Place) vs propensity for
change (People) metrics

Figure 17.6 details

Refer to Figure 17.6, which maps where the eight-cluster lay
within the size of the ask (Place) vs propensity for change
(People) metrics. This mapping tells us that around 20 per cent
or one in five employees represented by Cluster 8 already
showed strength in the behaviours the organization wanted to
strengthen.20 There was still scope for improvement for this
cluster, but note that this cluster ranks among the lowest for
propensity for change (basically, present at Place, absent for
People). That tells us that Cluster 8 was a potential resource for
the organization to draw on in understanding what it is at that
Place in that cluster that is going right and that could then be
shared with other clusters to facilitate organizational learning.
Whilst Cluster 6 shows the highest propensity for change, it
sits in the mapping as one of the clusters facing the largest
challenges in the Place adopting the target behaviours. That
told us that while employees and parts of the organization
represented by that cluster might be willing, it was highly
unlikely that they would be able to achieve the goals of the
change programme if the Place continues to be the inhibitor
Perhaps not the best Place to start the change journey if you
cannot change the Place, or in other words, if the size of the ask
is too difficult to change the Place; again, it’s about making
informed targeted choices to build on data and evidence.
What we can see from the mapping in Figure 17.6 of the
organization is a series of clusters towards the top right that
represent employees and parts of the organization where there
is scope for improvement, where change is feasible and where
the propensity for change is relatively strong. These are Cluster
5, 4 and 1, and here is where we begin to turn our map into a
roadmap as per our Figure 17.7.
 FIGURE 17.7 The translation of target clusters as the
starting point of the change programme into the three
pillar benchmarks

So, what is Figure 17.7 actually showing? Let’s start with the
doughnut charts at the bottom of that figure, which show
internal benchmarks for each of three target behaviours drawn
from analytics on the Place: B1 represents behavioural markers
for Customer Focus, B2 represents behavioural markers for
Simple is Best and reflects a desire to remove internal barriers
to organizational performance, and B3 represents behavioural
markers for Succeeding Together and an aspiration to break
down internal silos and promote shared commitment to
achieving organizational goals.21
 As I have already described, Cluster 8 represents a profile
where there is scope for improvement but lower inclination to
change. My view was that this is a segment of the organization
that offers a resource to support other segments given its
relatively strong profile across all three target behaviours. They
may not be inclined to change themselves, but they do offer the
opportunity to understand what is driving their current
strength to facilitate learning elsewhere in the organization.
There is also a reciprocal benefit to leveraging Cluster 8 in this
way. By observing the learning in other clusters, it is reasonable
to assume that this will reduce any reticence to change shown
by those in Cluster 8 at the start of the organization’s change
journey.
My suggested direction of travel at the start for this
organization’s change journey is centred on Cluster 5, Cluster 4
and Cluster 1 as these represent employees and parts of the
organization with a stronger propensity for change and where
there is specific rather than general scope for strengthening the
target behaviours.
Diving into the detail of the doughnut charts for these three
clusters shows where they can share learning from their
respective strengths to address target behaviours where a
cluster is weaker; for example, compare the bench strength of
Cluster 1 on Succeeding Together with the bench strength of
Cluster 4, and the bench strength of Cluster 4 on Customer
Service and Simple is Best with the bench strength of Cluster 5
for these target behaviours.
Iterating with and learning from these clusters offers a bridge
to other parts of the organization as the change programme
demonstrated that change is feasible, builds success stories and
leverages shared learning across and beyond these clusters.
And where would that change journey move next? Armed with
those tangible examples, leveraging the strong propensity for
change of those in Cluster 6 becomes not only feasible but also
significantly more achievable.

Addressing three root causes for the failure of

change programmes
More than 83,000 books on change management can be counted
on Amazon,22 and that number is growing I am sure, yet despite
all of that published wisdom, the statistics on the success of
change programmes continue to disappoint.
I do believe that organizations can beat the odds of failure for
change programmes by addressing three root causes driving
that failure:
1. A lack of insight into the employees who are the true
champions of change and have what it takes to be
effective in delivering change programme objectives.
2. A lack of insight into the size of the change task – the ‘ask’
of employees and work teams – that their change
programmes demand and how much the size of that ask
varies across an organization.
3. A lack of tools and specifically a behavioural roadmap that
enables them to promote shared learning through change
and to understand where in the organization they are
most likely to gain early traction with their change
initiatives.
So, the key learning is not to leave change to chance!
Notes
1 B Ewenstein, W Smith and A Sologar. Changing change management, McKinsey &
Co, 2015, www.mckinsey.com/featured-insights/leadership/changing-change-manag
ement (archived at https://perma.cc/298Y-WVE9)

2 P Strebel. Why do employees resist change? Harvard Business Review, May–June

1996, https://hbr.org/1996/05/why-do-employees-resist-change (archived at https://p
erma.cc/XKT7-BT84)
3 B Gleeson. Leading change: 6 reasons change management strategies fail, Forbes, 7
December 2016, www.forbes.com/sites/brentgleeson/2016/12/07/leading-change-6-r
easons-change-management-strategies-fail/ (archived at https://perma.cc/VC9Y-3CN
M)
4 E Lawson and C Price. The psychology of change management, McKinsey, 1 June
2003, www.mckinsey.com/capabilities/people-and-organizational-performance/our
-insights/the-psychology-of-change-management (archived at https://perma.cc/Q94
M-WTHL)
5 For an example of how the impact of context such as the workplace is now being
recognized in research on organizations see G Johns (2006) The essential impact of
context on organizational behaviour, Academy of Management Review, 31, pp. 386–
408
6 These behavioural dimensions are consistent with broader research on the
characteristics of people who are more open to personal change and more likely to
commit to organizational change programmes. For an example of this research see
M Vakola, I Tsaousis and I Nikolaou (2004) The role of emotional intelligence and
personality variables on attitudes towards organizational change, Journal of
Managerial Psychology, 19, pp. 88–110

7 D R Cox and E J Snell (1989) Analysis of Binary Data (2nd ed), Chapman & Hall
8 For details of benchmarks associated with different lines of organizational
research, see F A Bosco, H Aguinis, H, K Singh, J G Field and C A Pierce (2015)
Correlational effect size benchmarks, Journal of Applied Psychology, 100, pp. 431–
49
9 S Keller and B Schaninger. Getting personal about change, McKinsey Quarterly, 21
August 2019, www.mckinsey.com/capabilities/people-and-organizational-performa
nce/our-insights/getting-personal-about-change (archived at https://perma.cc/4G4J-
FE4Z). This paper shows that executives in organizations that use insights into the
mindsets of employees to guide their change programmes are four times more
likely to rate those programmes as successful.
10 To maintain organizational confidentiality, the titles used in this piece are
approximations of the actual titles used by this organization.
11 These workshops follow the principles outlined by Kurt Lewin for action research
with a specific focus on the ‘Changing’ aspect of change. That is, diagnosing the
impetus for change and exploring and testing new models of behaviour.

12 These behavioural markers act as pre- and post-intervention indicators or Key

Development Indicators (KDIs) that enable the organization to know where they
are starting from and how much variation there is in those starting points across
the organization, as well as the basis for metrics for measuring where the change
programme has achieved its objectives and where more investment is required to
achieve the change desired.
13 CultureScope provides data on both the People and the Place. In this chapter, we
focus on the data from the Place. In the third piece in this series, we show how
data on the People and the Place are used to develop a behavioural roadmap to
guide change programmes.

14 The metrics shown in this chapter are calibrated in percentiles using an internal
benchmark based on this organization’s employee responses to the CultureScope
behavioural diagnostic.
15 This plot shows the spread between the highest and lowest scores for any pillar.
The box for any cluster shows the main concentration of scores across pillars while
the vertical lines indicate where the highest and lowest scores for any pillar lie for
each cluster.
16 S Keller and B Schaninger. Getting personal about change, McKinsey Quarterly, 21
August 2019, www.mckinsey.com/capabilities/people-and-organizational-performa
nce/our-insights/getting-personal-about-change (archived at https://perma.cc/X34U-
3HZY). This paper shows that executives in organizations that use insights into the
mindsets of employees to guide their change programmes are four times more
likely to rate those programmes as successful.

17 There may be more than one target behaviour that sets the objectives and goals of
a change programme. In Part 2 of this chapter, we show how three business
objectives for the organization were distilled through a systematic process into a
set of behavioural markers that served to provide a clear and tangible definition of
each behaviour. Those behavioural markers also served to provide the
organization with benchmarks against which it could see where it was prior to
implementing its change programme and for measuring progress towards its
business aims.
18 For an example of how an incremental approach to change was adopted with
measurable impact using ‘implementation intentions’ (having a plan to achieve a
goal – an ‘implementation intention’ – increases the chance that people will follow
it up with actions), see the work supported by the Behavioural Insights Team in
supporting the people back to work programme. This (pages 7 to 11) and other
examples are provided in the Behavioural Insights Team: Update Report 2013-
2015, www.behaviouralinsights.co.uk/wp-content/uploads/2015/07/BIT_Update-Rep
ort-Final-2013–2015.pdf
19 As this part argues, a key factor in anyone adopting a target behaviour is the
obvious one of how goals are set for them to achieve the behaviour in question.
Irrespective of how challenging a target behaviour may be for an individual or
group of employees, adopting or changing behaviour is only possible if that
behaviour is defined in a clear and tangible way, and when there is consensus on
how that behaviour is defined and experienced.

20 Cluster numbering simply follows the numbering of the clustering programme we

used rather than any ranking or ordering by cluster number.
21 To maintain organizational confidentiality, the titles used in this chapter are
approximations of the actual titles used by the organization.

22 R Ashkensas. Change management needs to change, Harvard Business Review, 16

April 2013, https://hbr.org/2013/04/change-management-needs-to-cha (archived at
https://perma.cc/LZ8D-F5C3)
18
Are your HR processes
supporting alignment between
culture and strategy?
Using culture and behavioural
analytics to find whether the answer
is ‘Yes’ – or where to intervene when
the answer is ‘No’
As any wise organizational leader knows, organizations
perform better when they acquire, develop and retain the best
talent, but simply claiming ‘best practice’ HR processes is not
enough. This is because value grows fastest when those
processes align with business strategy.
As the global economy becomes increasingly knowledge-
based, firms need to acquire and nurture the best and brightest
human capital available and keep these investments in the
firm.1
HR processes improve operational effectiveness and financial
performance2 when they build skills, motivate employees and
empower employees to use their skills and motivations to
achieve organizational outcomes.3
 When you align culture with strategy, you transcend the
limits of traditional business conduct. You still need to crunch
numbers, but the data holds a more profound meaning – one
that informs multiple departments working together toward
cultivating something greater than the company itself.4 In fact,
in one study focused on the five-year benefits for alignment
between culture and strategy vs lack of alignment, has revealed
a positive impact of 17 per cent on financial profit and a 30 per
cent positive impact on growth:5 ‘The particular strategy a
company employs will succeed only if it is supported by the
appropriate cultural attributes’, yet over 56 per cent of the
surveyed leaders in the study reported low to moderate
alignment between workplace culture and strategy.6
The plethora of HR metrics in use today do not ask or answer
this question: ‘Are our HR processes strengthening the
alignment between strategy and culture?’ Or, as another
research study asks, ‘Is your talent strategy rooted in your
business strategy? Culture can’t just be an assortment of well-
meaning HR practices… There can be no talent strategy without
a compelling business strategy.’7
Organizations ignoring this question risk failing to anticipate
operational issues and diluting or destroying the value that they
seek to create.
Most HR metrics are focused on the following:
attract and acquire
onboard and develop
motivate and retain
Ultimately, the alignment between workplace culture and
strategy will be a strong predictor for the above metrics, as this
case study will shortly demonstrate.
 To create alignment between workplace culture and strategy,
organizations should have:
a valid approach for measuring their organizational
culture
a valid method for measuring their intended outcomes
a metric to measure alignment between culture and
strategy as well as the impact of the adopted HR process
an action plan to inform ongoing HR investments to
achieve targeted culture for serving the strategy
These approaches and methods bring together effective
behavioural diagnostics and culture analytics, showing
organizations to what degree their HR processes are aligned to
strategy, and, critically, where to intervene to improve the
alignment.
You may be asking, ‘Why this approach?’, and I will show you
why by making the following key points:
1. The most tangible evidence for an organization’s culture is
observable behaviour – the day-to-day actions of the
organization and its employees, and how they behave in
this organization.
2. Metrics for alignment must address two questions:
What behaviours drive the success or failure of
business strategy?
How aligned are the expectations of the organization
with the behaviour of its employees?
3. Culture and behaviour analytics must deliver actionable
insight into:
the measurable impact of HR processes on alignment
 which processes to focus on to ensure investments do
strengthen alignment
4. It is not enough to know which HR processes to focus on;
any action plan must also identify specific employee
segments where action will deliver the dividends from
strengthening alignment between workplace culture and
strategy.
So how do we get to understanding fit, misfit and impacts of HR
processes? Data on organizational expectations for behaviour
(the Place) and how employees naturally behave (the People)
provide clear metrics for levels of fit, misfit and misalignment,
which can be categorized as follows:
Virtuous Alignment: Expectations of the Place and the
behaviour of the People are aligned with strategic goals.
Double Negative: Clear lack of alignment (misfit) between
culture and strategic goals for both the People and the
Place.
Strong Foundation: Distinct type of misalignment,
pointing to the need for capability-type interventions for
the People.
Suppressed Talent: Another distinct type of
misalignment, pointing to the need for opportunity- and
motivation-type interventions for the Place.
Figure 18.1 illustrates these categorizations using the quadrant
distribution method by showing the behavioural alignment
between Place and People.
 FIGURE 18.1 The behavioural defined alignment
categorization zones

Figure 18.1 details

Using this categorization, organizations can discover how HR
processes affect alignment and invest appropriately and
effectively.
Figure 18.1 is only showing the behavioural alignment, which
on its own may be meaningless. However, with analytics you
can now add HR outcome metrics just as I described earlier. I
will now provide metric definitions as follows:
Attract and acquire: Hiring process, methods and
conversion rates and metrics. Are you attracting and
hiring talent for your target culture in line with the critical
behaviours you will need in the organization?
Onboard and develop: Where there is misalignment,
what should you focus on? Is it the Place or the People?
 Are there any particular employee segments that you need
to focus on?
Motivate and retain: Is staff turnover weakening the
alignment between culture and strategy? Are there key
points in the employee journey that you need to improve?
Let’s get into the details of how behavioural measurement and
culture analytics actually make sense of workplace culture
alignment with strategy. The process can be articulated as
follows:
1. Define: What are the intended outcomes metrics that you
have – are they valid? Do you have any gaps? Review your
strategy and define or translate it to outcomes that you
can measure.
2. Measure: Deploy valid behavioural diagnostics
organization-wide, with a target to achieve a statistically
relevant sample size.
3. Identify: Through the predictive analytics and path
analysis process already outlined in Part 1 of this book,
you will get to the behavioural leading indicators needed
to drive your thematic outcomes and achieve your
strategy.
4. Design, plan and implement targeted interventions:
You will now be able to examine the behavioural
alignments across the entire organization and design, plan
and implement interventions by demographic segments.
5. Validate: As before, remeasure at sensible intervals
measuring the behaviours targeted, the alignment
between Place and People and the outcomes.
Let us now test this approach on our next case study. This
organization is a large professional services business that was
interested in hiring and retention success in line with five
behaviours that predicted all their targeted outcomes and
strategy. The behaviours are: Collective, Empower,
Standardized, Conformity and Innovate.
Figure 18.2 shows a multiple countries analysis where
alignment to the targeted behaviours of recent hires is lower
despite stronger available talent pools. Is this due to a blind
spot in hiring practices, complacency, or both?
 FIGURE 18.2 Multiple countries analysis hiring
against the targeted five behaviours

Figure 18.2 details

Analysing Figure 18.2 reveals that there is a three times
variance across the countries for talent acquisition. At the top
end, one in two candidates have the behaviour Virtuous
Alignment; that drops to one in six at the other end of the scale.
Combining behavioural alignment data between Place and
People with additional employee metrics for tenure, gender and
job level (seniority) reveals the following trends:
One-year tenure is a key pivot point in terms of alignment,
with longer tenure correlated with Suppressed Talent or
Double Negative.
Gender is correlated with distinct patterns of
misalignment, with females more likely to represent
Strong Foundation and male employees Suppressed
Talent.
 Job level (grade) correlated with misalignment
(Suppressed Talent) and misfit (Double Negative).
Figure 18.3 shows the above demographic type insights versus
the alignment zones and the likelihood of occurrence.
 FIGURE 18.3 Employee demographic insights versus
the alignment zones and the likelihood of occurrence

Figure 18.3 details

For clarity, the narrative for Figure 18.3 goes like this
(remember, the alignment refers to the five critical behaviours
that predict their targeted outcomes and ultimately their
strategy):
For Virtuous Alignment (the target zone), I am almost two
times more likely to be an employee outside of Europe,
with a tenure of one year or less and within a lower job
grade.
 For Strong Foundation (focus interventions on People
capability), I am four times more likely to be female, with
a tenure of one year or less and within a lower job grade.
For Suppressed Talent (focus interventions on opportunity
and motivation), I am four and a half times more likely to
be male with a tenure of two years or more.
For Double Negative (focus interventions on capability,
opportunity and motivation), I am almost two times more
likely to be a manager in Europe with a tenure of two
years or more.
In addition to answering the question for this organization ‘Are
our HR practices strengthening our alignment to strategy?’, the
culture fit metrics also provide the analytics framework for
identifying actions and interventions where the answer is
‘maybe’ or ‘no’.
Employee behavioural profiles associated with levels of fit
provide clarity as to what investments are required for HR
practices to strengthen alignment and where those investments
should be targeted. Those profiles enabled the organization to
design and implement interventions that are tailored to specific
employee segments and to set up actions with clear goals and
measures of success.
Having gained the behavioural actionable insights along with
the demographic segmentations, the following were
recommended as key action points for this organization:
Attract and acquire:
Action: Key organization-wide issues are centred
around the variation in the impact of the hiring
practices across countries, manifesting in the lower
 fit of hires in countries with stronger talent pools;
immediate review of the selection process; perhaps
learn from the one country that got it right.
Action: Key employee segment is the managerial
hires with higher levels of Double Negative and
where hiring and promotion criteria will benefit
from immediate review.
Onboard and develop:
Action: For male executives, onboarding should focus
on leveraging their talents and soft skills to build
stronger alignment between the Place and the
People.
Action: For female employees, development
programmes should focus on the positives of the
Place and address soft skills gaps among employees.
Motivate and retain:
Action: Pivotal retention point is just after one year
of tenure, with alignment weakening for longer
tenures; is there an issue with employee experience
over time? The trick is to learn what is happening in
the first year of tenure that is favourable and find a
way to maintain it after the first year in the employee
journey.
Action: Risks of employee turnover in Europe are
driven by management employees who feature in the
Double Negative zone. Immediate
programmes/interventions should be designed and
implemented to work on capability, as well as
 opportunity and motivation; these interventions
must target the five critical behaviours.
Additionally, those demonstrated profiles defined by
demographics also serve to challenge working assumptions that
may be misplaced and fail to surface operational issues and
risks.
Before we move on, I want to address an interesting question
related to tenure that I often get asked. That is, ‘Can the
behavioural alignment or tension between People and Place be
a predictor of tenure?’ To answer the question fully, I would
develop a full saturated regression model resulting in restricted
nested behaviours, with odds ratio, based on tenure to answer a
primary question, ‘Can the behavioural profile predict which
tenure group a person is likely to belong to?’ The answer for
one of the organizations that asked this question is
demonstrated in Figure 18.4.
 FIGURE 18.4 The tenure odds ratio for five
behaviours: Disenfranchise, Delivery Focus,
Collective, Consolidate and Conformity

NOTE N=8,588 and R=0.359

Figure 18.4 details

Analysing the model in Figure 18.4 reveals that the five
behaviours are the predictor of tenure; these are
Disenfranchise, Delivery Focus, Collective, Consolidate and
Conformity. You can see that the longer the tenure, the more
likely that employees will have these behaviours as present. In
fact, employees that have been with the organization for 11 to
16 years are 11 times more likely to have these behaviours as
present, and employees that have been with the organization
for up to two years are nine times less likely to have these
behaviours as present.
Having developed the model, we can now look at the tension
between Place and People for the five behaviours. That yields a
dramatic insight for two of the behaviours:
Disenfranchise tension for lower tenures is at 40 per cent
between People and Place, decreasing to 23 per cent for
higher tenures
Consolidate tension for lower tenures is at 38 per cent,
decreasing to 10 per cent for higher tenures
Basically, the higher the tenure, the more likely employees are
aligned to the Place for the two behaviours.
This organization also examined employee departure historic
data, showing that employees are 38 times more likely to leave
when the tension between People and Place, for the two
behaviours, is above 18 per cent.
Now for the complication with this case. Two of the predictive
behaviours that are part of the model for the intended
outcomes to achieve the strategy are opposite to the retention
model – Empower vs Disfranchise and Innovate vs Consolidate.
This organization is actually hiring talent fit for their strategy;
however, this talent is not likely to stay as the issue is with the
Place. So action is needed in two ways:
1. Intervention design targeted at the Place for the desired
two behaviours, Empower and Innovate (dichotomous to
Disenfranchise and Consolidate) for all tenure groups, to
address the need for opportunity and motivation.
2. Intervention design targeted at the People for the desired
two behaviours, that is Empower and Innovate
 (dichotomous to Disenfranchise and Consolidate) for
tenure of five years and above, to address the need for
capability.
This second case tells an interesting story about how you simply
cannot drive change by just hiring. You will need to create the
environment as well, as the two will work in tandem.

Notes
1 T R Crook, S Y Todd, J G Combs, D J Woehr and D J Ketchen (2011). Does human
capital matter? A meta-analysis of the relationship between human capital and
firm performance, Journal of Applied Psychology, 96, pp. 443–56
2 B E Becker and M A Huselid (1998) High performance work systems and firm
performance: A synthesis of research and managerial implications, Research in
Personnel and Human Resources Management, 16, pp. 53–101
3 K Jiang, D P Lepak, J Hu and J C Baer (2012) How does human resource
management influence organizational outcomes? A meta-analytic investigation of
mediating mechanisms, Academy of Management Journal, 55, pp. 1264–94
4 W Craig. What happens when you align culture with strategy, Forbes, 29 May 2018,
www.forbes.com/sites/williamcraig/2018/05/29/what-happens-when-you-align-cult
ure-with-strategy/ (archived at https://perma.cc/B4LZ-JQF9)
5 B Jaruzelski, J Loehr and R Holman. Booz & Company Global Innovation 1000: Why
culture is key, strategy+business, 25 October 2011, www.strategy-business.com/arti
cle/11404 (archived at https://perma.cc/46UQ-RWSG)
6 Institute for Management Development (2017) IMD World Talent Rankings 2017

7 B Taylor. 5 questions to ask about corporate culture to get beyond the usual
meaningless blather, Harvard Business Review, 1 June 2017, https://hbr.org/2017/0
6/5-questions-to-ask-about-corporate-culture-to-get-beyond-the-usual-meaningless-
blather (archived at https://perma.cc/EP94-KDY7)
19
Going beyond diversity metrics
to actionable insights on DE&I
It’s the behaviours that matter
While the business case for diversity, equity and inclusion
(DE&I) has been made by those such as Dixon-Fyle, Dolan, Hunt
and Prince (2020),1 doubts are growing about whether DE&I
programmes are having a real and positive impact. In an online
BBC article, Nathoo (2021) cites research showing that the kinds
of training that organizations tend to favour the most are short,
one-shot sessions that can be completed and the requisite
diversity boxes ticked, are unlikely to make a difference in the
long-term behaviour of participants.2
Friedersdorf (2023) in an article in The Atlantic comments:
‘While its advocates claim that “diversity workshops can foster
better intergroup relations, improve the retention of minority
employees, close recruitment gaps and so on” ... in practice
there is little evidence that many of these initiatives work. And
the type of diversity training that is currently in vogue –
mandatory trainings that blame dominant groups for DEI
problems – may well have a net-negative effect.’3
These doubts are supported by a US survey conducted by
WebMD Health Services (2023) showing that, while 89 per cent
of respondents report their organization has a DE&I
programme and while 72 per cent report wanting to work for
an organization that supports it, 46 per cent report that their
organization is not doing enough, with around a half of
respondents reporting experience of work situations
inconsistent with a DE&I culture.4
Could the root problem be behaviours and organizational
culture?
From their survey, WebMD concludes ‘the DE&I focus should
be on shifting workplace culture so that the programs meet the
needs of employees and are supported through all levels of
management’. I agree, but the notion of ‘culture’ is a rich one
with a wealth of theories and models (e.g. Denison; Hofstede;
Schein; Trompenaars and Hampden-Turner5). To frame our
approach to understanding culture, let’s consider the notion of
‘unconscious bias’ that many DE&I programmes aim to address.
Unconscious bias, by definition, is bias that we are unaware
of. The former US Secretary of Defence Donald Rumsfeld
captured this when he stated ‘Reports that say that something
hasn’t happened are always interesting to me because, as we
know, there are known knowns; there are things we know we
know. We also know there are known unknowns; that is to say
we know there are some things we do not know. But there are
also unknown unknowns – the ones we don’t know we don’t
know. And if one looks throughout the history of our country
and other free countries, it is the latter category that tend to be
the difficult ones.’6
So, how can organizations get to something that is an
‘unknown-unknown’? In our view, maybe they can’t. What they
can do is determine what people should consciously experience
day-to-day that evidences a real commitment to DE&I. In that
spirit, our approach to culture can be summed up as clusters of
behaviours that reflect the beliefs and values of individuals and
organizations, and the questions we seek to answer are: what
behaviours demonstrate a true commitment to DE&I, and to
what extent are those behaviours present in an organization
today? From the answers to those questions, we then work back
to explore how targeted behaviours can be developed to
provide a platform for broader organizational learning and
change.
A similar definition has been adopted by the UK’s Financial
Conduct Authority (FCA), important to the context of this case
study, who define culture as ‘the habitual behaviours and
mindsets that characterize an organization’.7 The FCA goes on
to state: ‘Simply put, culture is “the way things are done around
here” ... This includes the norms, values and practices which are
revealed by how people think and behave ... as well as our
behaviour when no-one is looking.’

What’s the elephant still in the room? Maybe it’s

real commitment to DE&I programmes!
Recently we have been hit with a tsunami of DE&I-related news
items that begs several questions about whether we have
achieved a truly DE&I-friendly workplace culture. This is the
elephant that may still be in the room. Whilst DE&I
programmes are seen as a must-do, are organizations really
gaining traction with those programmes? From that question
follows another for the analytics community – how is the
investment in analytics helping organizations understand the
impact of their DE&I programmes, and what do organizations
need to do to strengthen that impact?
 The ask of DE&I programmes is not an easy one; if
organizations want diversity, then they have to be able to
embrace difference, even divergence, and recognize the value
in doing so. If they want inclusion, then they have to accept the
sharing of opinions even if, at first sight, that may appear to
bring the additional complexity of including others in making
decisions and in acting on those decisions.
Let’s take a step back from demographics and take a moment
to think about what DE&I programmes are really asking
organizations and their employees to do. A little reflection takes
us inevitably and logically to a simple proposition: if the proof
for whether DE&I programmes are gaining traction lies in how
people behave rather than the demographic they represent,
then surely organizations need to understand whether their
culture encourages or discourages the behaviours aligned with
DE&I initiatives.
Whilst most organizations are frantically chasing the
diversity metrics, and there are so many to focus on, they often
miss that one must primarily and critically create the
environment of acceptance and equality, else all that diversity
will leak out. It’s a bit like trying to eat soup with a fork!
Before we move on, I want to provide basic definitions for
diversity, equity and inclusion:
Diversity: being mindful of all dimensions of human
differences at organizations. It often translates to a
deliberate process for how many different types of people,
by demographics, are hired and promoted.
Equity: having an egalitarian environment and the
policies that service equality where people of all
differences, genders, races and ethnicities are treated in
 an equal manner. This results in equal opportunities for
all.
Inclusion: the act of including people regardless of their
differences. This act leads to an environment where
everyone feels respected and valued.

Seven behaviours of a DE&I-friendly workplace

culture
How do you know if the culture of your organization is DE&I
friendly? As part of the research I conducted, outlined in
Chapter 4, I managed to extract seven behavioural dimensions
for the Place that talk to how inclusive an organization is:
1. Collective: Goals achieved through partnership and
alliances.
Inclusion impact: Shared obligations are clear. Goals
are set and agreed within the organization and with
external stakeholders in a spirit of partnership and
through alliances.
2. Empower: Decisions are delegated to lower levels with
supportive guidance.
Inclusion impact: Decision-making is a cascade, not a
directive. They are effective at delegating decisions to
lower levels within the organization and provide
supportive guidance on how those lower down
should exercise judgment.
3. Expressive: People do not hesitate to express their
feelings and openness is valued.
 Inclusion impact: They want to know what
employees really think and feel. Employees are
encouraged to share and express their opinions and
concerns.
4. Team Focus: Work is delivered through collaborative
effort and team results are prioritized over individual
results.
Inclusion impact: Teamwork is valued as much as
individual contributions. Managers recognize that
goals are achieved through collaborative effort.
5. Active Learning: Active steps are taken to improve
employees’ skills and careers, and employee growth is
considered an integral part of the job.
Inclusion impact: They ask the question ‘How are
people growing through the work they do?’
Employees grow through their jobs and their work is
actively recognized in reward and in career
progression.
6. Conformity: People try to deal fairly with each other
based on clear rules, values and mutual obligations.
Inclusion impact: The rules for dealing with
differences are clear. Employees are encouraged to
resolve issues based on clear expectations for
interpersonal behaviour and mutual obligations.
7. Achievement: People are valued for the contribution that
they make, irrespective of status.
Inclusion impact: Accountability is based on
contributions made. People are valued for the
 contribution they make, irrespective of the status
they hold in the organization.
Essentially, once the CultureScope diagnostic is deployed
organization-wide, measuring the above listed behaviours, an
inclusion index can be obtained. However, as I am regularly
reminding all organizations, workplace culture is unique – so
one should not rely purely on that index. Through the addition
of DE&I outcome measures to the analytics, each organization
can obtain their own inclusion behavioural index.

The context for the case study

The organization is a private financial institution subject to the
regulatory framework for the financial sector in the UK, and
specifically the Financial Conduct Authority (FCA).8
An audit by the FCA triggered by concerns over the
company’s risk exposure and volatility in its core markets had
identified a number of additional concerns related to working
practices that they felt reflected issues with the organization’s
culture. As a consequence, the FCA instructed the company to
commission a report by what the FCA refers to as a ‘skilled
person’, which they define as ‘appearing to the regulator to
have the skills necessary to make a report on the matter
concerned’.
In short, the organization was required to respond to the
FCA’s concerns in a meaningful way and to demonstrate that it
understood what it needed to fix by producing a tangible plan
for doing so.
Whilst the organization had an existing employee survey, it
and the FCA felt that this survey did not provide clarity around
employee engagement and did not provide the basis for
tangible actions. As an example of the type of negative
workplace culture indicator referred to by the FCA, and for
which the existing employee survey did not provide clear
diagnostics, one data point shared by the organization with me
before CultureScope was deployed indicated potential DE&I
issues.
While the employee split across the organization’s workforce
by gender was approximately 50:50, the average tenure of
female employees was 2.2 years compared to an average tenure
for male employees of 6.8 years, more than three times lower.
The CEO had some key questions: ‘What behaviours exist in the
organization, and how do these specifically relate to their DE&I
agenda? And what are the relevant outcomes to consider?’
Thus, with the FCA’s approval, CultureScope was
commissioned at a time when the company appointed a new
CEO to lead the transformation of the organization. That CEO, in
turn, created a new position of Communications and Culture
staffed by a C-suite appointee with a strong background in
corporate communications.
These appointments signified a broader intent of the
company in deploying a behavioural approach to change. While
we have referred to the FCA’s interest in organizational
workplace culture underpinning the effective conduct of
financial firms and therefore, in their view, business
performance and customer confidence, the new CEO with
Board approval had also established a need to develop and
implement a new people strategy.
To respond to the FCA’s concerns, the organization had
identified a clear need to undertake a business transformation
to address rebalancing its portfolios and its systems for
managing its customer transactions as well as tighter controls
over its day-to-day finances. To achieve that business
transformation and to demonstrate to the regulator that it could
address what might be referred to as the hard and soft factors
key to successful transformation, the organization was also
conscious of the need to ensure it had a stable supply of talent
to deliver the transformation. This included retaining the talent
it had as well as attracting strong talent where it needed to
strengthen its workforce. A new people strategy was therefore
seen as key to the overall success of a broader business
transformation.
The project was supported by the appointment of an
independent transformation consultant to work alongside my
team. In line with the principles of change management
described earlier, the appointment of the transformation
consultant served to combine a data-driven approach with the
insights offered by someone experienced in working at Board
and C-suite levels, in business change and transformation as
well having experience at a senior level in financial services
including commercial and retail banking, brokerage and
Fintech.
The focus of the transformation consultant’s role was to work
with the organization’s leaders in translating behavioural and
actionable insights from CultureScope into achievable goals, to
work with human resources and organizational development
specialists in the organization to help shape communications in
support of achieving those goals, as well as working alongside
my team in identifying where good practice already existed in
the organization that could be used to provide concrete
examples for sharing and strengthening desired behaviours
across the organization.
This specific case for DE&I began with the organization
asking our team to investigate broader issues related to an
organizational transformation programme which was part of
an organizational development (OD) initiative.
One of the OD initiatives was to explore barriers to employee
engagement, seen as essential to shifting the client organization
from where it had been to a healthier inclusive culture and
demonstrating a commitment to this shift in culture to the
financial regulator.
To truly answer the ‘why’ of why DE&I was not working, and
just before CultureScope diagnostics deployment, I worked with
the OD consultant to design specific outcome measures. We
decided to add in our deployment two sentiment-type questions
related to the feeling of belonging, as belonging is one key
outcome of DE&I. The questions were:
Outcome 1: I feel I am a valued member of our organization.
1. Strongly agree
2. Agree
3. Neither agree nor disagree
4. Disagree
5. Strongly disagree
Outcome 2: I feel I am a valued member of my team.
1. Strongly agree
2. Agree
3. Neither agree nor disagree
4. Disagree
5. Strongly disagree
Insights generated
The CultureScope diagnostic was deployed, and data collection
yielded a completion rate of 1,565, representing a 75 per cent
response rate within three weeks, providing a substantial effect
size. The high response rate reflected the company’s
commitment to culture measurement, including a video by the
CEO explaining the purpose and importance of the project,
daily reminders with links to the survey and dedicated time to
complete the survey allotted to employees as part of their work
schedules.
To develop the full insights for DE&I, I used two approaches:
firstly using the seven behaviours inclusion index, and secondly
developing an outcome-based model then delivering insight on
absent and present behaviours.
Using the seven behaviours inclusion index, the data was
analysed by gender and ethnicity with breakdowns for various
parts of the organization, notably for client-facing ‘front office’
operations. The results showed the following:

Substantial variance by gender

To illustrate the difference at the Place, and to provide another
diagrammatic method to present analytics, I will use radar
graphs. With this type of graph you will be able to see the
targeted behaviour and its dichotomous pair.
Figure 19.1 shows the master inclusion index as well as the
behavioural factors driving the results of the overall index. For
each behavioural dichotomous pair, the target behaviours are
underlined for ease of reference and clarity. The inclusion
index includes a percentage ruler, which is the outermost rim,
and, for each benchmark, an inner strength rim denoting in
percentage terms how inclusive the organization is. Anything
beyond 50 per cent for a benchmark denotes that inclusion is
present, and less than 50 per cent indicates that for the specific
benchmark inclusion is absent.
 FIGURE 19.1 Radar graph and inclusion index for the
whole organization, organizational behaviours

NOTE Total n=1565, M=811, F=754. ODNA = Organizational Culture Behavioural

DNA

Figure 19.1 details

The overall gender split may look great at 52 per cent male vs
48 per cent female employees. The gender diversity agenda is
working! However, the issue is with inclusion. Figure 19.1
shows significant variance by gender of the observed (at the
Place) behaviours. To provide contextual analysis for two of the
behavioural factors as an example:
Conformity organizationally is absent for females, so
fairness and rules based on equality are not working for
me if I am female. However, if I am male then Conformity
is relatively present.
Team Focus is organizationally absent if I am female,
meaning I am not invited to team collaborative work and I
am expected to work alone. This is very different if I am
male, when teamwork is happening at the Place.
The above behaviours are a Place issue. What I am articulating
is that the opportunity and motivation for Team Focus and for
Conformity are present for males, and absent for females.
The Place is working against providing an environment for
inclusion, resulting in much higher female employee turnover
(female average tenure is 2.2 years vs male average tenure of
6.8 years) and lower engagement, explaining the ‘why’ of
poorer performance. By now, this example should illustrate
how inclusion by behaviour is a leading indicator for tenure,
engagement and performance.
I find it very interesting that most organizations tell me that
they want to focus on performance. My response would be:
performance is the outcome, and you need to focus on the
culture and behaviours that drive the desired performance.
This example goes further, as you will need to focus on
behaviours that drive inclusion before you can influence
performance.
To demonstrate the ‘where’ as an additional insight, we
examined the Front Office since that is the function that is the
most client-facing. Their behaviour organizational data (the
Place) demonstrated a concerning variance.
Figure 19.2 shows a significant difference across a number of
organizational behaviours. In the ‘Place’, in fact, for the Front
Office, inclusion is working totally against females.
 FIGURE19.2 Radar graph and inclusion index for
Front Office, organizational behaviours

NOTE Total n=538, M=281, F= 257

Figure 19.2 details

Whilst I was participating in a number of focus groups
organized by the OD consultant with female employees to
explore this further, the qualitative information obtained from
these focus groups’ substantiated CultureScope results which
had also identified an unhealthy management environment in
one of the customer-facing operations (Front Office).
When this unhealthy environment was uncovered, the Chief
Executive Officer asked me why the female employees had not
spoken up and expressed these significant issues. I was quick to
share with the executive the females’ own behavioural
measurement for Expressive, as it is absent and shows
significant difference from their male colleagues. See Figure
19.3 for the radar graph: behaviour peaks in the present zone
for males for Expressive and is absent for females. This
behavioural People absence is also combined with the
organizational (Place) absence for Expressive along with other
gender- and Place-related behavioural factors, namely
Conformity, creating an environment in which female
employees were unlikely to speak up.
Simply, in the Front Office, the environment does not support
gender equality for speak up, and that has fully manifested in
its absence for females.
 FIGURE 19.3 Radar graph for Front Office, personal
behaviours

NOTE Total n=538, M=281, F= 257

Figure 19.3 details

Substantial variance by ethnicity and racial

discrimination uncovered
Let’s examine ethnicity with the same behavioural lens. The
data analysis showed a positive result for only one ethnic group
tagged as ‘White’ (51 per cent) with another ethnic group in
particular showing a concerning variance tagged as ‘Black’ (22
per cent). A further 27 per cent reported a different ethnicity
and were classified for analysis as ‘Mixed’. Interestingly,
ethnicity diversity numbers again may look OK, but the
inclusion index tells a different story.
This illustrated a particular ethnic cohort with significantly
absent organizational behaviours for the inclusion index,
suggesting that the organization treats this group differently.
These differences conjure up contrasting styles of interaction
depending on the ethnicity of the employee, but what does data
on the Place suggest for behaviours encouraged by the
workplace? Figure 19.4 shows Black employees and employees
classified as Mixed see the workplace as discouraging for
Collective, Team Focus, Expressive, Active Learning and
Conformity when compared to White employees. Particularly
concerning is the significant behavioural difference between
the White ethnicity and Black ethnicity employees at the Place,
resulting in the inclusion index being absent for the Black
ethnicity and by contrast somewhat present for the White
ethnicity.
 FIGURE 19.4 Radar graph showing the organization
(Place) behavioural differences and the inclusion
index for different ethnic groups for the whole
organization

NOTE Total n=1565, White=792, Mixed= 421, Black=352

Figure 19.4 details

Going back to the outcome measures to understand

‘how’ to fix inclusion
As with any organization and research project, the process
through which I developed the model shared in this book
followed a number of iterative steps. Ultimately, choosing a
model comes down partly to what the data says. So, to develop
an inclusion index and behaviours specific to this organization
as leading indicators, let us now run an analysis for the two
outcome measures: ‘I feel I am a valued member of our
organization’ and ‘I feel I am a valued member of my team’. The
outcomes are broadly seen as indicative for the feeling of
belonging.
 At the aggregated data level, I completed a maximum
likelihood factor analysis with oblique rotation of the two
outcome questions. I then used a series of hierarchical
regressions to explore the relationships between CultureScope
behavioural measures for the ‘People’ and the ‘Place’ for each
outcome model. I used a fully saturated model with all 15
CultureScope dimensions as the baseline to test more restricted
and nested models with fewer CultureScope dimensions to
arrive at our final predictor set of CultureScope behaviours.
To illustrate the strength of each model, I further calculated
the odds ratio for each outcome as odds ratio is an easy way to
represent the likelihood of an outcome based on the absence
and/or presence of nested behaviour models to show how good
the predictive model is.
To apply the odds ratio method, we have arranged the
behaviours into four quartiles, and count the number of
outcome occurrences per nested model quartiles.
If it occurs that the odds ratio is lower than 1, this indicating
an odds against relation, hence the negative reciprocal of this
value, and therefore it would be plotted on the negative y-axis.
Based on the methodology above, let’s explore the first
outcome:
Outcome 1: I feel I am a valued member of our organization.
1. Strongly agree
2. Agree
3. Neither agree nor disagree
4. Disagree
5. Strongly disagree
 FIGURE 19.5 Regression analysis odds ratio for the
first outcome, employees being valued by the
organization

NOTE n=1,565. Probability of behaviours’ presence or absence. Behaviours are

Conformity, Expressive and Team Focus

Figure 19.5 details

Figure 19.5 shows that the absence or presence of three
organizational behaviours, Conformity, Expressive and Team
Focus, predict the first outcome with a multiple R=0.5614. For
this effect size that is beyond the 80th centile for predictive,9
and results in the following odds ratio: employees are 19 times
more likely to feel valued at the organization when Conformity,
Expressive and Team Focus are present at the ‘Place’, and two
times less likely to feel valued when the said behaviours are
absent.
Again, based on the methodology above, let’s explore the
second outcome:
Outcome 2: I feel I am a valued member of my team.
1. Strongly agree
2. Agree
3. Neither agree nor disagree
4. Disagree
5. Strongly disagree
 FIGURE 19.6 Regression analysis odds ratio for the
second outcome, employees being valued by their
team

NOTE n=1,565. Probability of behaviours’ presence or absence. Behaviours are

Conformity, Expressive, Achievement and Team Focus

Figure 19.6 details

Figure 19.6 shows that the absence or presence of four
organizational behaviours predict the second outcome with a
multiple R=0.6211, that is Conformity, Expressive, Achievement
and Team Focus, resulting in the following odds ratio:
employees are 36 times more likely to feel valued by their
teams when Conformity, Expressive, Achievement and Team
Focus are present at the ‘Place’, and two times less likely to feel
valued when the said behaviours are absent.
 Having explored both models, you can now see that three
behavioural factors exist in both models. In fact, with simple
path analysis it can be deduced that the organization will not
achieve outcome 2 if they can’t solve outcome 1, and since
outcome 1 relates to being valued at the organization, I think
that makes perfect sense. If belonging is not driven across the
entire organization, then teams are unlikely to deliver it.
So, what does this mean? Focusing on the inclusion context, it
means that regardless of gender, ethnicity, seniority and so on,
the Place must provide all employees with the opportunity to
express their opinions and views. The rules must apply fairly
and equally to all (this will be a critical checkpoint for any
organizational policies, as some can work against inclusion),
and team contribution should be encouraged and welcomed as
every team member matters (it’s about what you do rather than
who you are).
Having built a cluster of the behaviours for outcomes 1 and 2
respectively, we can further investigate how significant the
issue is by gender and ethnicity and then focus on the absent
behaviours for intervention design and planning.
Figure 19.7 shows the three behavioural clusters generated by
the first outcome model applied to the gender benchmark, and
this illustrates that at the Place, interventions are needed for
opportunity and motivation for the female population. The
inclusion index for these three behaviours provides further
evidence on the gap, showing how differently females are
treated by the Place compared to their male colleagues.
 FIGURE 19.7 Gender behavioural radar graph and
inclusion index for model one outcome, employees
valued at the organization

NOTE Total n=1,565

Figure 19.7 details

Figure 19.8 shows the four behavioural clusters and inclusion
index for the second outcome generated model; that is, being
valued by teams within the organization. As the second model
encompasses all three behaviours from the first model, it will
not come as a shock that variance between the genders is
significant; however, the fourth behaviour, which is
Achievement, does not have such a stark difference. This does
simplify matters as interventions directed at the three
behaviours that exist in both models will go a long way to solve
the inclusion issues.
For completeness, let us apply the same lens for the generated
behavioural cluster models for both outcomes to generate
ethnicity benchmarks.
 FIGURE 19.8 Gender behavioural radar graph and
inclusion index for model two outcome, employees
valued by their teams

NOTE Total n=1,565

Figure 19.8 details

Figures 19.9 and 19.10 show the ethnicity benchmarks for the
outcome 1 and 2 models respectively, exhibiting organizational
behaviours and the inclusion index. These figures indicate that
similar interventions are needed as per the gender insights, and
should be directed at the Place for three behaviours:
Conformity, Team Focus and Expressive. The key demographic
that would benefit from such interventions is the ‘Black’ ethnic
group.
FIGURE 19.9 Ethnicity behavioural radar graph and
inclusion index for model one outcome, employees
valued at the organization

NOTE Total n=1,565

Figure 19.9 details

 FIGURE 19.10 Ethnicity behavioural radar graph and
inclusion index for model two outcome, employees
valued by their teams

NOTE Total n=1,565

Figure 19.10 details

A better approach to illustrate the ‘where’ is by using
quadrant dispersion analysis, so let us apply the quadrant
dispersion method for the model 1 and model 2 behaviours. It
can be seen from the analysis in Figure 19.11 that the three
behaviours can be learnt from the Treasury and IT functions,
and in particular if we can understand the environmental
difference between the two functions and Front Office, we can
design interventions that are likely to work. If they are working
for you already, then the question becomes: how can we
replicate that?
 FIGURE 19.11 Quadrant distribution of the three
targeted behaviours: Conformity, Team Focus and
Expressive

Figure 19.11 details

From a return-on-investment perspective, it is interesting to
note that the targeted three behaviours resolve inclusion and
boost all diversity demographics.

What actions were taken?

Turning insights into action for this organization involved both
quantitative and qualitative elements, and the behavioural
quadrant alignment matrix as shown in Figure 19.12 developed
for this organization was used to guide the OD transformational
consultant in conducting further interviews and workshops.
That alignment matrix identified which business functions
showed stronger alignment and which, like the Front Office,
showed weaker alignment, giving the OD transformational
consultant a map to work from to compare employee
experience where alignment was stronger and weaker.
 The deeper dives led to a number of targeted interventions
ranging in scope from companywide communication to those
focused on the company’s executive committee’s actions and to
how existing use of technology, such as collaboration platforms,
in one business function could strengthen DE&I behaviours in
other functions.
 FIGURE 19.12 Behavioural quadrant alignment
matrix and descriptions per quadrant

Figure 19.12 details

These deeper dives were led by the OD transformation
consultant and identified that where alignment was higher
between the People and the Place for Expressive, there was a
stronger capability among leaders to make the time to engage
in two-way communication. Furthermore, that willingness to
encourage open communication whilst showing respect also
enabled trust in the use of technology for more open
communication between employees. In contrast, and where
alignment was lower, the contrasting pole of Neutral was more
likely because leaders and managers felt they did not have the
time or the physical space for them to be Expressive, and those
leaders and managers also felt uncomfortable about where the
limits were to be Expressive, indicating an issue of
psychological safety.
Further deeper dives led to interventions aimed at promoting
a sense of psychological safety from the top down in the
organization, to using corporate communications to provide
examples of how Expressive and other behaviours could be
incorporated and encouraged in day-to-day interactions, and
how issues of physical space could be addressed through using
the virtual space provided by technology.
Here are three examples of the interventions introduced by
the organization using data to guide the qualitative inquiry
pursued by the OD transformation consultant:
Companywide communications. Review of existing
communications showed that they rarely touched on the
three DE&I behaviours. The Head of Communication and
Culture took the remit with her team to establish
guidelines to frame future communications and reinforce
the seven behaviours, singly or together as the nature of
the communication allowed. These guidelines were first
used in company-wide messaging and then, once the
guidelines had been concretely modelled, cascaded to
business unit-level communications.
Executive Committee. One example of interventions at this
level was framed around psychological safety and
included workshops as well as one-to-one coaching
focused on two behaviours, Team Focus and Expressive, to
build understanding among leaders on how these
behaviours foster trust and their role in modelling those
behaviours for the wider organization.
 Use of Microsoft Teams channels. This was already used
effectively in a back-office function to promote and
reinforce three of the behaviours in the normal course of
work activities; Conformity (what fairness and applying
the rules fairly to all means and how to apply it day-to-
day), Expressive (encouraging sharing of opinions and
feelings) and Team Focus (fostering collaboration). The
deeper dive showed that training and usage of Teams was
variable across the company, and few in some functions
knew of or even thought they could use the chat function
to explore ideas and opinions with their colleagues.

So, did the needle move?

CultureScope measurement was redeployed as planned after 10
months from the first measurement to check if the planned
interventions were working. To what degree were they having
a positive effect on the intended behaviours, and had the
outcomes improved?
Having analysed the newly collected data with n=1,586, the
following can be observed:
Expressive has increased by 22 per cent and is now a
present behaviour for females. Interestingly it has also
slightly improved for males, and we now have alignment
at the Place regardless of gender. Expressive presence is
also showing alignment for different ethnic groups.
Personal behaviours also showed an improvement by 19
per cent and 21 per cent.
Conformity has had a significant change from very absent
for females to present with a 32 per cent change, and is
 now showing alignment with male colleagues for the
Place. Conformity at the Place has also improved to only
present for the ‘White’ ethnicity to present for all
ethnicities.
Team Focus has also shown an improvement from
somewhat absent to somewhat present for females, and
although not totally aligned with male colleagues yet, it’s
not far behind. Ethnicity also shows similar improvement
at the Place.
Achievement has increased by 20 per cent for females and
is now aligned with males, which has also impacted
ethnicity by 16 per cent improvement.
Although these demonstrate a moderate gain, they are
encouragingly in the right direction.

So what? What about the outcome?

A further deployment of the organization’s employee survey in
the same timeframe also showed a positive shift in employee
outcomes. Table 19.1 compares the percentage variance for
employees responding favourably (agree or strongly agree) to
the two outcome items discussed previously by gender and
ethnicity with favourable responses to the items increasing
substantially for Female employees and for ethnically Black
employees and those classified as Mixed, with little or no
change for Male and ethnically White employees (total n =
1,565).
 TABLE 19.1 The two outcomes comparison before
and after interventions by gender and ethnicity
demographics
Skip table

1st Measurement (before 2nd Measurement (after

interventions) n=1565 interventions) n=1586
Valued at the Valued by the Valued at the Valued
Place People Place People
Gender: Male 68% 71% 72% (var 4%) 74% (v
positive response
Gender: Female 33% 36% 59% (var 26%) 56% (v
positive response
Ethnicity: White 75% 80% 75% (var 0%) 78% (v
positive response
Ethnicity: Mixed 52% 57% 66% (var 14%) 67% (v
positive response
Ethnicity: Black 48% 38% 61% (var 13%) 59% (v
positive response

The behavioural and outcome measurement data shows

improvement, but the organization still has a way to go, now
that they have the evidence that their targeted interventions
are working.
The data was also shared with the financial regulator (the
FCA), and the big question is: what did the regulator think when
this data and the interventions designed by the OD
transformation consultant were shared with them? The FCA
was satisfied that the company had identified where the gaps in
its workplace culture were and had taken concrete actions to
address those gaps, which is a great result considering the
starting point for this organization.
 Additionally, the organization has also reported an
improvement in employee openness and wellbeing. It is also
interesting to note the organization effectiveness measures
have also improved, and in particular at the Front Office; such
measures included customer workflow effectiveness and issue
resolutions. Clearly an inclusive organization will be creating
an inclusive Place and brand for their employees and
customers alike! As the saying goes, ‘Look after your employees
and they will look after your customers.’

Implications for the research approach focused

on inclusion
As those cited in the introduction to this chapter either state
directly or imply, current DE&I practice appears to have run
ahead of research on what interventions are more likely to
promote a lived experience of DE&I in the workplace. So, what
does this case study say about where future research on DE&I
might focus?
First, I suggest it offers one way to respond to the
observations made by WebMD Health Services on the back of
their 2023 survey in investigating cultural factors underpinning
employee experience of DE&I as it is and as it could be.10
Second, I hope it offers a model for how both quantitative data
using an empirically tested behavioural framework can be
combined with qualitative lines of inquiry to build a deeper
understanding on where there are levers to promote a stronger
sense of DE&I among employees and where there are barriers
to employees having a positive sense of an organization’s
commitment to DE&I.
 From the perspective of the practitioner, this case study is
offered as one example of how a behavioural framework
drawing on practical research can be applied to generate
testable hypotheses about what promotes or hinders employee
experience of DE&I within a given organizational setting,
hypotheses that can be tested in terms of whether there is a
consensus about what the tangible experience of DE&I should
be and whether that consensus translates into meaningful and
substantive predictions of whether employees feel valued or
not.
The case study is also offered to show how quantitative data
analytics can be combined with qualitative inquiry to better
understand the employee experience as it is and as it could be,
but also to leverage behavioural frameworks by translating
them into practical interventions that can, in turn, be tested as
to whether they are having an impact and moving the needle.
Many DE&I practices appear to have erred on the side of
efficiency at the expense of meaningful applications of research
on behaviours and behavioural change. I really hope that I have
provided an example of how behavioural science and culture
analytics can not only help promote meaningful support to
DE&I initiatives through organizational applied research and
data-guided insight, but also achieve tangible results efficiently
through an evidence-based approach.

Notes
1 S Dixon-Fyle, K Dolan, V Hunt and S Prince. Diversity wins: How inclusion matters,
McKinsey, 18 June 2020, www.mckinsey.com/~/media/mckinsey/featured%20insight
s/diversity%20and%20inclusion/diversity%20wins%20how%20inclusion%20matter
s/diversity-wins-how-inclusion-matters-vf.pdf (archived at https://perma.cc/F37S-Z
WP7)
2 Z Nathoo. Why ineffective diversity training won’t go away, BBC Worklife, 17 June
2021, www.bbc.com/worklife/article/20210614-why-ineffective-diversity-training-w
ont-go-away (archived at https://perma.cc/BC4E-LUJ8). This article cites a 2019
qualitative review of 30 studies of interventions aimed at reducing implicit bias (C
FitzGerald, A Martin, D Berner et al. Interventions designed to reduce implicit
prejudices and implicit stereotypes in real world contexts: a systematic review,
BMC Psychology, 7 (29), https://doi.org/10.1186/s40359-019-0299-7 (archived at http
s://perma.cc/VFX2-4D6C)). These authors found inconsistent impacts from these
interventions and comment on the lack of substantive empirical research in this
area.
3 C Friedersdorf. The paradox of diversity trainings, The Atlantic, 18 January 2023, w
ww.theatlantic.com/newsletters/archive/2023/01/diversity-training-paradox-intoler
ance/672756/ (archived at https://perma.cc/96YS-L9E8)
4 WebMD Health Services. Efforts to promote diversity, equity, inclusion, & belonging
in the workplace falling short, new WebMD Health Services survey finds, PR
Newswire, 2 February 2023, www.prnewswire.com/news-releases/efforts-to-promo
te-diversity-equity-inclusion--belonging-in-the-workplace-falling-short-new-webmd
-health-services-survey-finds-301736551.html (archived at https://perma.cc/K6NC-C
7DA)
5 See examples of some research by these authors:

N Fondas and D Denison (1991) Corporate Culture and Organizational

Effectiveness, The Academy of Management Review, 16, p. 203, https://doi.org/10.
2307/258613 (archived at https://perma.cc/2VQM-9232)
G Hofstede (2011) Dimensionalizing cultures: The Hofstede model in context,
Online Readings in Psychology and Culture, 2 (1), https://doi.org/10.9707/2307-091
9.1014 (archived at https://perma.cc/LJA4-G8GT)
E Schein (1990) Organizational Culture, American Psychologist, 45, pp. 109–19, h
ttps://doi.org/10.1037/0003-066X.45.2.109 (archived at https://perma.cc/B6S5-LW
29)
F Trompenaars and C Hampden-Turner. Riding the Waves of Culture, January
1998, www.researchgate.net/publication/238710832_Riding_the_Waves_of_Cultu
re (archived at https://perma.cc/HXG2-5CWU)

6 Transcript – US Department of Defense News Briefing – Secretary Rumsfeld and

Gen Myers, 12 February 2002, https://web.archive.org/web/20160406235718/http://a
rchive.defense.gov/Transcripts/Transcript.aspx?TranscriptID=2636 (archived at htt
ps://perma.cc/R5EV-XFHG)
7 FCA, Culture and governance, www.fca.org.uk/firms/culture-and-governance
(archived at https://perma.cc/U6NK-9YKB)
8 For more information about the role and mandate of the FCA, see www.fca.org.uk/p
ublications/corporate-documents/regulatory-framework-reforms (archived at http
s://perma.cc/HGS6-AKCV)
9 F A Bosco, H Aguinis, K Singh, J G Field and C A Pierce (2015) Correlational effect
size benchmarks, Journal of Applied Psychology, 100, pp. 431–49
10 C Muldoon. The future is inclusive: What a culture of diversity, equity, inclusion &
belonging looks like, WebMD Health Services, 14 February 2023, www.webmdhealt
hservices.com/blog/the-future-is-inclusive-what-a-culture-of-diversity-equity-inclus
ion-belonging-looks-like/ (archived at https://perma.cc/X9PS-PV5K)
20
A tale of two organizations: Why
innovation needs the letter ‘C’
Day in, day out, we read, hear and see breakthroughs that are
changing the way we live and work. The Apple iPhone, the
Internet of Things and robotic treatments that have
dramatically increased the survival rates from disease, not to
mention the ubiquitous impact of big data and artificial
intelligence. So, we may think that innovation is alive and well.
But is it?
Companies are still spending heavily on research and
development (R&D), with PwC reporting that the 1,000 largest
corporate R&D spenders globally increased their R&D spend by
3.2 percent in 2017 to $702 billion (over £500 billion).1 As that
article states, there is still a shared belief among executives that
‘innovation today is a key driver of organic growth for all
companies – regardless of sector or geography’.
The latest report for the global R&D market stipulates that
R&D is poised for substantial growth in the coming years,
driven by increasing investments in various industries and
technological advancements.2 The predicted Compound Annual
Growth Rate (CAGR) for this market ranges between 14.1 per
cent and 17.2 per cent from 2023 to 2027. This growth is
attributed to several factors, including the emergence of
cutting-edge technologies, the rise of innovative startups, and
the increased focus on R&D by governments and private
organizations worldwide. The pharmaceutical and life sciences
sector accounts for a significant share of the R&D market, given
the growing demand for novel treatments and therapies.
Additionally, the information technology and electronics sectors
are also witnessing considerable R&D investments due to rapid
advancements in areas such as Artificial Intelligence (AI), the
Internet of Things (IoT) and 5G technology.
Clearly the budgets and investments are there, but return on
investment is still the elephant in the room. So, is innovation
getting harder or are we just getting worse at innovation? This
was the question posed in a Harvard Business Review article as
to whether those investments in innovation are seeing a
sustainable return.3 The data suggests not, with returns to
companies’ R&D spending declining by 65 per cent since the
late 1980s – a decline that mirrors the worrying decline in
productivity that continues to vex economists.
Is it simply that innovation has become harder as scientists,
engineers and those engaged in R&D struggle to find ever more
incremental ways to add value to the processes and
technologies already in place? Or is it that some companies are
better at innovation than others? And, if that is the case, what is
it about those companies that enables them to be better than
others at innovation?
A deeper dive into the data indicates that the decline in
returns from R&D spend is not necessarily true of all
companies. The not-so-good news is that many companies are
making it harder for themselves to achieve a return from the
spend on R&D, and they are just not very good at innovation.
It’s not just about smarts and process: Why the
letter ‘C’ is important
Whilst returns from R&D spend have been declining,
organizations, private and public, have expended enormous
efforts in processes for managing R&D, not least in change
management. The problem with these efforts is that all too
often they have ignored one fundamental factor in successful
change – workplace culture. And, yes, that’s where the letter ‘C’
comes in – it’s not about yet another change management
programme.
This became clear at a recent meeting of 100 CEOs organized
by IBM’s Ginni Romerty. Representing 17 different industries
and some $2 trillion (£1.5 trillion) in revenues, the meeting was
confident that technology was about to see a new wave of
disruption by helping companies leverage their core expertise
through more effective management of data. Upbeat though
this meeting was, Time magazine found a challenge shared
among CEOs in realizing this opportunity: ‘The biggest problem
they face is not technology, but rather creating a workplace
culture that can embrace and adapt to technological change’.4
This isn’t surprising when you consider that innovation is
principally about new futures – new products, new services and
new ways of doing things. It is about future states. While those
charged with innovation are pursuing those future states, their
colleagues are focused on business as usual, the current state
they have been trained and socialized to understand and work
with. Bridging these two states – current and future – is the crux
of the problem recognized by those CEOs attending the IBM
event.
Context and methodology
Following the 2008 financial crisis, banking competition was
increased by the opening of the financial market. Banking now
operates in an increasingly competitive environment, which
makes innovation an important strategy for organizational
survival.
A combination of regulatory changes and public
dissatisfaction with traditional banking practices led to the
proliferation of new entrants of various structures and
offerings into the banking market.5
These new entrants into the market have been grouped into a
single category by the media: challenger banks. Therefore,
customers now have increased choices about where they can do
their banking and are no longer limited to the big traditional
banks for their banking needs.
Has regulation improved innovation by virtue of introducing
competition? This case study aims to give a snapshot of both a
challenger and a traditional bank’s journeys with innovation,
and can also show whether regulation enables or hinders
innovation. Both banks underwent a qualitative and
quantitative cultural analysis to understand the characteristics
in each bank that enable and/or inhibit innovation.
I will aim to demonstrate the connection between
organizational culture and innovative capabilities driving the
competitive advantage.
To understand and analyse the workplace culture of these
two banks, CultureScope behavioural diagnostic and analytics
was used. However, for the purposes of this study only seven
behaviours driving innovation were analysed, which will I
refer to as the culture innovation index. This index was
developed during the significant research outlined in Chapter 4
as well as recent ongoing research. As I have articulated in
previous chapters, we should never rely exclusively on a
predeveloped framework, and you must research your own
workplace culture and the outcomes it drives to build your own
unique model and framework. However, I will use this index
for this research as the insight into the cultural differences and
impact on innovation returns.
You may be wondering why I am doing so despite the ‘health
warning’ I just provided. As the purposes for this case study
suggest, I only want to create the connection between R&D
investment, workplace culture and the return on investment,
hence any organizations seeking to drive change should do
their organization-specific research.
For qualitative analysis, semi-structured interviews were
conducted with employees from both the banks. In the
traditional bank, participants from various departments
including the innovation centres, asset management,
investment banking, systems and sales were interviewed. In the
challenger bank, participants from the departments of
recruitment, private banking, regional banking, innovation,
learning and development, and retail management were
interviewed.

The seven behaviours defining the innovation

index
Before we dive into the differences between the two
organizations, it would be good to define those seven
behaviours in the context of innovation as follows:
1. Flexible: People value variability and deal with each
situation afresh, vs Standardized: People value working
within clear processes and systems.
2. Innovate: New and creative ideas are pioneered;
intelligent risk taking is encouraged and praised, vs
Consolidate: Predictability and control are preferred; risk
taking is avoided until it is absolutely necessary; tried-and-
tested methods are encouraged.
3. Radical: Continuous improvement and evolving ideas are
valued; there is an emphasis on responding differently to
different situations, vs Moderate: Individuals value
working steadily within assigned and streamlined
processes favouring stability.
4. Synchronous: People see the past, present and future as
interwoven periods; they often work on several projects at
once, in a fast-paced manner, and view plans and
commitments as flexible, vs Sequential: People like events
to happen in order; they place a high value on punctuality,
quality, planning and staying on schedule; in this culture,
‘time is money,’ and people don’t appreciate it when their
schedule is thrown off.
5. Outer Focus: People believe that their environment
controls them; they want to work within their
environment to achieve goals; at work or in relationships,
they focus their actions on others, vs Inner Focus: People
believe that they can control nature or their environment
to achieve goals; this includes how they work with teams
and within organizations.
6. Conformity: People place a high importance on laws,
rules, values and obligations; they try to deal fairly with
 people based on these rules, but rules come before
relationships, vs Nonconformity: People believe that each
circumstance, and each relationship, dictates the rules
that they live by; their response to a situation may change
based on what is happening in the moment, and who is
involved.
7. Strategic: Emphasis is on long-term delivery and results,
and focus is on the wider impact, vs Tactical: The
emphasis is on short-term delivery and results.

The traditional bank (TB)

This is a well-established global bank that has been trying to
drive innovation particularly in their retail and business
banking. The focus has been on researching, developing and
delivering innovative products and solutions that are
technology-based. For that purpose, a research and
development (R&D) function was established that included the
creation of an innovation board to manage a significant billion-
dollar investment – a well-funded R&D centre.
This innovation centre was established within the bank’s
normal operational environment and is governed and managed
the same way as all other ‘business as usual’ functions.
Unfortunately, the bank had very little success to show off!
Hardly any innovative products or solutions have been
delivered, and any product improvements that did go live had
significant issues, leading to a barrage of customer complaints.

Qualitative analysis
Based on 38 semi-structured interviews conducted with
employees at the traditional bank, and after analysing the data
through a bottom-up approach, a thematic map was generated
to describe the organizational culture. The resulting map has
been crystalized in three themes: power and hierarchy,
importance of internal risk management, and external
moderators to innovation.
Let’s dive into each theme.
THEME 1: POWER AND HIERARCHY
Top-down decision-making: The bank is highly
bureaucratic and values the importance of power and
hierarchy. Top-down decision-making is the normal course
of action in deciding what innovation means for the
organization and who decides and implements functions
related to it.
Information exclusivity: Another output of the
hierarchical structure is that very senior executives had
exclusive knowledge of how innovation is propagated and
maintained in the organization, and the teams driving
innovation were simply information-blind about what is
next and the direction of travel or even strategy.
Tradition: The bank was often described as ‘old’ and
‘slow’ by employees. Participants spoke of an organization
where ‘everything moves slowly’ and referred to the
technology and ways of thinking as ‘outdated’ or ‘passé’.
The value TB places on tradition and ‘the way things are
done’ also seemed to inhibit the implementation of new
ideas.
 FIGURE 20.1 Power and hierarchy theme

See Figure 20.1 which provides a diagrammatic summary for

this theme.
THEME 2: IMPORTANCE OF INTERNAL RISK MANAGEMENT
Pockets of experimentation: It is within the bank’s
innovation functions that employees are encouraged, by
their remit, to experiment and are provided this space to
try different approaches, occasionally creating a ‘two-
speed’ organization.
Playing it safe: Although employees are encouraged to
speak up if they have new ideas, these were usually
contained to small changes that would not risk the
efficient running of the organization as a whole.
Sharing and collaboration: Being a large organization,
various communication channels are needed to make
relevant employees aware of the potential risks and
benefits of new products or ideas. ‘Tech fests’ and
‘innovation showcases’ are places where information
about new technology can be shared amongst members of
the organization. Ideas are then reported upward and
implemented if management believes they can provide
value for the business.
 FIGURE 20.2 Importance of internal risk
management theme

Figure 20.2 provides a diagrammatic summary for this theme.

THEME 3: EXTERNAL MODERATORS TO INNOVATION
Competitor effects: If competitor banks have done it,
then we can – it’s about borrowing innovation. Certain
innovative solutions can also be progressed quickly, but
only if there is a perceived threat to the bank’s reputation
or market share.
Regulatory roadblocks: A highly competitive
marketplace is a key accelerator for innovation; however,
there was a perception by employees that regulatory
roadblocks and security concerns, such as cybersecurity
and customer data protection, inhibit or limit the level of
innovation.
Security concerns: Security concerns are a key factor that
could stop the business from innovating in the ways it
would like. Historically, TB has garnered media attention
for their involvement in various scandals that created a
loss in reputation and a resulting loss in stakeholder trust
and market share. The bank seeks to protect both its
reputation and existence as an organization by
prioritizing cyber-security and allowing innovation to
‘take a back seat’ in the day-to-day functioning of the
organization. The focus was total risk avoidance rather
than risk awareness.
 FIGURE 20.3 External moderators to innovation
theme

Figure 20.3 is the diagrammatic summary for this theme.

Quantitative analysis
In the thematic qualitative analyses, we built a picture of what
is hindering innovation. Let’s validate further by now adding
the behaviours to articulate ‘why’ the qualitative outcomes are
so.
Figure 20.4 shows the master innovation index as well as the
radar graph showing the seven behaviours that combine to
drive the index. As for each behavioural dimension, both
dichotomous factors are shown; however, the targeted
behaviours are underlined for ease of reference.
 FIGURE 20.4 The innovation index indicator and a
radar graph for the seven behaviours that drive it for
the Place

Figure 20.4 details

Analysing the innovation index yields the first headline news,
essentially that the organization is working against innovation.
Why? The Place is about Consolidating, being Moderate, being
Standardized, and without analysing further we can see this is a
culture for preserving the status quo and business as usual. A
workplace like this is unlikely to support multiple
experimentation or pioneering ideas, or support doing things
differently. We just met the ‘enemy’, and it’s the Place!
What about the People? The bank has been hiring a team
dedicated to innovation and R&D – has the bank got its hiring
strategy right? To answer this question, we can now look the
behavioural insights using the same index for the People hired
to be dedicated to R&D.
Figure 20.5 looking at the innovation index indicates that the
bank is indeed hiring talent aligned to a culture for innovation.
Why? Again, look at the Innovate, Radical, Flexible and
Synchronous behaviours to name a few. However, the
innovators are incubated in an environment that would seem
to inhibit what they are hired for, hence significant tension. I
can tell you that this tension has two outcomes: these
innovators have a maximum tenure of 16 months, and as the
Place is acting as the inhibitor, although the programme is very
well funded, there are no tangible outcomes and return on the
bank’s investment.
 FIGURE 20.5 The innovation index indicator and a
radar graph for the seven behaviours that drive it for
the Place and the People – traditional bank

Figure 20.5 details

The key action is to create a separate environment to incubate
the innovators away from the business-as-usual operations
within the bank.

The challenger bank (CB)

This is a relatively new bank, having gained their license in
2018/19. The bank is all about driving innovation in both
technology and modus operandi. Smaller than a typical
traditional bank, with significantly lower financial muscle
power available to invest in R&D for innovation, this bank is
winning awards for product innovation and is making
significant gains in new customer acquisition, customer
retention and organizational growth.
Qualitative analysis
Based on 32 semi-structured interviews conducted with
employees at the challenger bank, and after analysing the data
through a bottom-up approach, I generated a thematic map to
describe the resulting organizational culture. To compare like
with like, I managed to establish three themes: however, they
are very different themes to the traditional bank, which is to be
expected if they do have a fundamentally different workplace
culture.
The themes are: creating the best customer experience,
importance of collegial relationships, and external perceptions.
Let us dive into each theme as follows.

THEME 1: CREATING THE BEST CUSTOMER EXPERIENCE

I can summarize the qualitative evidence for this theme
resulting from the semi-structured interviews as follows:
(words in bold show relevance to the thematic Figure 20.6):
All participants interviewed spoke about innovation at CB
as their commitment to customer service. Employees
believe the banking experience should be simple and
friendly and employees are encouraged to go above and
beyond for the customer.
Customers are also encouraged to give feedback both
online and face-to-face on their banking experience,
which is then incorporated into further innovative
offerings.
Part of creating the best customer experience also depends
on providing top-of-the-line new technological offerings
to customers. Due to financial and human resources
constraints, CB does not have the ability to provide in-
house created offerings. The organization reconciles this
challenge by partnering with third-party vendors.
 FIGURE 20.6 Creating the best customer experience
theme

Figure 20.6 is the diagrammatic summary for this theme.

THEME 2: IMPORTANCE OF COLLEGIAL RELATIONSHIPS

Again, I can summarize the evidence for this theme as follows:
Employees place a high value on customer relationships,
but they also place equal value on collegial relationships,
stating that ‘our colleagues are our customers’. The
resulting work culture is friendly, nurturing and
structurally not hierarchical.
CB’s employees are actively encouraged to come up with
new ways to make the organization better, whether it’s a
new idea to solve an internal problem or a new customer
proposition.
This informal ‘roll up your sleeves’ attitude is also
solidified in the formal performance review process,
where employees are asked to recount what tangible
actions they had taken to make sure the organization is
gradually improving.
Tying the collegial relationships theme together is a
shared responsibility to make the organization better,
with everyone taking ownership to make improvements
in the organization. When asked where new ideas come
from, answers were similar: ‘The onus is on everyone, we
all play a part’, and ‘Everyone is responsible’.
 FIGURE 20.7 Importance of collegial relationships
theme

Figure 20.7 is the diagrammatic summary for this theme.

THEME 3: EXTERNAL PERCEPTIONS

The theme’s demonstrated narrative summary is as follows:
The bank is perceived as new by the public and has an
untarnished reputation due to the fact that it was
established after the financial crisis. When asked about
how the bank’s ability to innovate is affected by this, one
employee answered: ‘People trust you as a brand. It’s not
necessarily perceived as being as risky as it might be for
a [traditional bank] that have got legacy issues.’
Therefore, the bank feels enabled to provide new
customer technologies and innovations without fearing
that customers may be wary of such a change.
The bank also believes that it is perceived by the public
to be risk-averse, due to its low lending rates and purely
retail-based offering. Employees believe customers view
CB as a ‘safe and stable lender’ and, thus, a trusted brand.
There are certain technology affordances that are
available to the bank that stem from the fact that CB was
created with technology already embedded in its
structure, unlike traditional banks that have had to add on
technological structures as society has moved into a
technology age.
Having a modern ‘tech stack’ enables innovation at CB
since technological changes can be implemented quickly
and widely throughout the bank.
 FIGURE 20.8 External perceptions theme

Figure 20.8 is the diagrammatic summary for this theme.

Quantitative analysis
Buckle up and hold on to your seat as I will now show the
quantitative behavioural insights explaining the ‘why’ of the CB
workplace culture and the great outcomes that they are
achieving.
Figure 20.9 shows the master innovation index as well as the
radar graph showing the seven behaviours that combine to
drive the index. I have combined the benchmarks for both the
People and the Place in one figure, and what a difference
compared to the traditional bank! The innovation index shows
the behavioural presence and alignment between the Place and
the People ‘ready made’ to serve and drive effective innovation.
 FIGURE 20.9 The innovation index indicator and a
radar graph for the seven behaviours that drive it for
the Place and the People – challenger bank

Figure 20.9 details

The Place is enabling the behaviours of Innovate to be
pioneering, Flexible to be able to respond to operational change
and Synchronous providing the ability to work on multiple
innovative trials and experimentations, to name a few
behaviours. The People are aligned to Place and, if anything,
there is some positive mild tension as the Place is ahead in
some of the behaviours.

Do we need the letter ‘C’ in Innovate?

Well, I can’t change the English language, but I really hope I
have illustrated the importance of workplace culture to drive
Innovate. For this case study, I can also argue that regulation is
not a blocker – both banks are operating in the same regulatory
environment, and you can see from the behavioural diagnostic
that both have Conformity as a present behaviour, hence you
can still innovate while abiding by the regulations and rules.
Another key takeaway is that pouring money into innovation
is not the only solution. One needs to create the environment
though workplace culture; in fact, I would go as far as saying do
not fund it until you are sure the environment exists, or you are
working to create it. Remember what I previously called this
approach: ‘Culture by Design’.
For large, well-established organizations, innovation can
become a survival issue, and I would argue that creating a
completely separate environment away from the day-to-day
business will be important. Creating such an environment
means a different culture, with different operating model, agile
organizational structure, different policies and processes and
ways of working. Just think of it as creating and funding your
own competitor to your current organization, and the good
news is that it still belongs to your organization and brand.
Another question that I posed at the beginning of this case
study: is innovation getting harder? It would seem that it is for
many organizations as they are making it harder for
themselves by either ignoring the importance of culture or
because they lack the insight they need into whether their
workplace culture is enabling or disabling their efforts to
innovate.
That’s why I believe that effective innovation needs the letter
‘C’, and the letter ‘C’ is for culture.

Notes
1 PwC. The Global Innovation 1000 study: Investigating trends at the world’s 1000
largest corporate R&D spenders, strategy&, https://www.strategyand.pwc.com/gx/e
 n/insights/innovation1000.html (archived at https://perma.cc/DVN8-6SG6)
2 Research And Development in Global Market Overview 2023–2027. Market
Overview Report – August 2023
3 G O’Connor. Real innovation requires more than an R&D budget, Harvard Business
Review, 19 December 2019, https://hbr.org/2019/12/real-innovation-requires-more-t
han-an-rd-budget (archived at https://perma.cc/37B9-BMU6)
4 A Murray. Get ready for the new disrupters, TIME, 2 November 2017, https://time.co
m/5006970/get-ready-for-the-new-disrupters/ (archived at https://perma.cc/27CF-3Q
8Q)

5 R Bennett and R Kottasz (2012) Public attitudes towards the UK banking industry
following the Global Financial Crisis, International Journal of Bank Marketing, 30
(2), pp. 128–47, https://doi.org/10.1108/02652321211210877 (archived at https://per
ma.cc/KS5D-JLS4)
21
Safety culture and risk
management
Nuclear power generation stands at the intersection of
advanced technology, energy production and safety
considerations. In this context, cultivating a robust safety
culture is paramount to ensure the reliable and secure
operation of nuclear power plants.
Safety culture in nuclear power generation refers to the
shared values, beliefs and behaviours that prioritize safety
above all else. Given the potential risks associated with nuclear
energy, maintaining a strong safety culture is not just a
regulatory requirement but a moral and operational
imperative. A robust safety culture is built on a foundation of
accountability, transparency and continuous improvement.
Key attributes for nuclear generation safety workplace
culture:
Leadership commitment: Safety culture starts at the top.
Leaders in the nuclear industry must demonstrate an
unwavering commitment to safety, setting the tone for the
entire organization. This involves fostering a proactive
approach to identifying and addressing safety concerns.
Employee empowerment and accountability: An
effective safety culture empowers every member of the
organization to actively participate in ensuring safety.
Employees should feel comfortable reporting concerns,
 suggesting improvements and contributing to a culture
where safety is everyone’s responsibility.
Continuous training and learning: The dynamic nature
of the nuclear industry necessitates ongoing training and
education. Keeping employees abreast of the latest safety
protocols, technological advancements and lessons
learned from past incidents contributes to a culture of
continuous improvement.
Open communication: Open and transparent
communication channels are fundamental to a strong
safety culture. Employees should feel encouraged to
express safety concerns without fear of reprisal. Regular
safety meetings, reporting mechanisms and feedback
loops play a crucial role in maintaining an open dialogue.
Regulatory bodies play a pivotal role in ensuring that nuclear
power plants adhere to strict safety standards. Regulatory
oversight helps maintain accountability and provides an
external layer of scrutiny to complement internal safety
measures. For that purpose, the World Association of Nuclear
Operators (WANO) has established the Nuclear Leadership
Effectiveness Attributes.1
Given the global nature of nuclear energy, international
collaboration is essential. Countries and organizations share
best practices, conduct joint safety assessments and participate
in forums to collectively enhance safety measures and address
common challenges.

Context
For this case study, we will visit a nuclear power generation
organization as we address the important topic and theme of
safety culture.
The organization’s aim is to prove their ability to demonstrate
to the World Association of Nuclear Operators (WANO) that
they have established a workplace culture for safety. In line
with the key attribute of the safety workplace culture for the
nuclear operations, I worked with the organization to design
the following key outcome questions (this will act as the lagging
outcome indicators).
Psychological Safety:
We hold each other to account and challenge
appropriately.
I am always able to speak up and challenge deviations.
Inclusion:
I feel like I can be myself at work.
Questioning Attitude:
During an unexpected event, my line manager encourages
stopping and seeking advice.
When faced with an unclear procedure, I pause work and
try to resolve the issue.
I am expected to consider undesired consequences before
starting work.
Safety Communication:
I always have briefings to capture safety issues and
concerns when handing over work.
Line manager communicates desired and undesired safety
behaviours with examples.
Continuous Learning:
We often seek support from OPEX to enhance safety
before carrying out work.
My department proactively conducts self-assessments to
improve nuclear safety.
My department proactively benchmarks to improve
nuclear safety.
I proactively and regularly check that I am compliant with
training to be qualified and experienced to do my job.
The training I receive enables me to perform at my best.
Accountability:
The decisions I make affect our performance.
I am consistently held to account for the delivery of
actions.
Leaders use the formal accountability process to monitor
performance and delivery.
Leaders hold each other to account for delivery and
performance standards.
There are clear consequences for non-delivery or
displaying poor behaviours.

The generated actionable insights

The CultureScope behavioural diagnostic was deployed in
October 2023 for the entire organization. After two weeks of
data collection, a 48 per cent completion rate was established,
providing a large enough sample size for statistical and
analytics validity as described in Chapter 6.
After closing data collection, the analytics for actionable
insights work commenced with a clear objective of establishing
a behavioural (leading indicator) predictor for each theme and
for each outcome within a theme. This was achieved by
combining the outcome measures to the saturated 30 factors (15
dimensions) behavioural measurement and conducting the
hierarchical regression analysis to establish the best
parsimonious model for a restricted nested behavioural model.
The odds ratio method was then used to articulate the model
bench strength.
Having established the leading indicators for each outcome
and theme, a path analysis was established using the
methodology described in Chapter 10.
I will provide you with outcome examples for each theme to
demonstrate the findings before we put the entire picture
together using the path analysis method.

PSYCHOLOGICAL SAFETY
Figure 21.1 shows that people are 19 times more likely to be psychologically safe when
the behaviours of Active Learning, Standardized, Collective and Sequential are in the
upper quartile of presence.2
FIGURE 21.1 The odds ratio for Psychological Safety
where Active Learning, Standardized, Collective and
Sequential behaviours are present

n=303. Probability of behaviours’ presence or absence

Figure 21.1 details

INCLUSION
This is predicted and driven by Active Learning, Conformity, Relationship and People
Focus. When these behaviours are in the upper quartile of their presence, this
organization is 34 times more likely to gain inclusion at their Place.
Referring to Figure 21.2, if these behaviours are absent, this organization is five times
less likely to gain inclusion at the workplace.
FIGURE 21.2 The odds ratio for Inclusion at work
where Active Learning, Conformity, Relationship and
People Focus behaviours are present

NOTE n=357. Probability of behaviours’ presence or absence

Figure 21.2 details

QUESTIONING ATTITUDE
Three behaviours are the leading indicators for questioning attitude: they are Active
Learning, Sequential and Team Focus. Where the People have these three behaviours as
present, they are 61 times more likely to have a questioning attitude, as Figure 21.3
shows.
If that was not compelling enough, consider this – employees are two times less likely
to have a questioning attitude when these three behaviours are absent.
 FIGURE 21.3 The odds ratio for Questioning Attitude
at work where Active Learning, Conformity, Sequential
and Team Focus behaviours are present

NOTE n=303. Probability of behaviours’ presence or absence

Figure 21.3 details

SAFETY COMMUNICATION
For this organization, as Figure 21.4 indicates, they are 54 times more likely to achieve
safety communication outcomes where four behaviours are present; they are
Standardized, Sequential, Conformity and Team Focus.
FIGURE 21.4 The odds ratio for Safety Communication
outcomes at work where Standardized, Sequential,
Relationship and Team Focus behaviours are present

NOTE n=303. Probability of behaviours’ presence or absence

Figure 21.4 details

CONTINUOUS LEARNING
For this theme, it’s good to see the very first behavioural leading indicator is Active
Learning. Additionally, Strategic, Team Focus, Sequential and People Focus are key
behaviours, and where these five behaviours are in the upper quartile of their presence,
this organization is 24 times more likely to achieve a workplace culture for continuous
learning.
Referring to Figure 21.5 you can also see that this organization is six times less likely
to achieve a culture for continuous learning where the five behaviours are absent.
FIGURE 21.5 The odds ratio for driving a Continuous
Learning culture at work where Active Learning,
Strategic, Team Focus, Sequential and People Focus
behaviours are present

NOTE n=357. Probability of behaviours’ presence or absence

Figure 21.5 details

ACCOUNTABILITY
For this last theme, we have a very interesting odds ratio as Figure 21.6 reveals. This
organization is seven times more like to drive accountability where the behaviours of
Strategic, Conformity, Team Focus and Sequential are in the upper quartile of their
presence.
The flip side of this story provides for an interesting insight as well; if these four
behaviours are absent, this organization is 17 times less likely to drive a culture of
accountability, and even if these four behaviours are only slightly absent, this
organization is 9 times less likely to drive the desirable culture of accountability.
 FIGURE 21.6 The odds ratio for driving a culture for
Accountability at work where Strategic, Conformity,
Team Focus and Sequential behaviours are present

NOTE n=357. Probability of behaviours’ presence or absence

Figure 21.6 details

Let’s just take a breather and think about what this is telling us;
if the People do not think and act in line with the long-term
needs and action consequence, if they do not follow the rules
and regulations, they do not plan accurately and precisely and
let’s not forget team orientation (it seems accountability in
nuclear generation is a team sport!), even the slight absence of
these behaviours can spell disaster.
Already we can see a few behaviours existing in many of the
predictive models, but for good measure we should now turn
our attention to the path analyses to see if all the themes hang
together and what is the path to greatness.

Path analysis
The next step is to use the structural equation modelling
approach as described in Chapter 10 to discover if these
outcomes are connected, and in what order.
The result of the analyses has provided two paths. I have
named these paths for easy of reference. Path 1 is the
Accountability path, and I will present it in two ways, simplified
as demonstrated in Figure 21.7 and a more detailed path as
shown in Figure 21.8.
FIGURE 21.7 The Accountability path, also showing
the behaviours’ presence needed to drive that path

Figure 21.7 details

 FIGURE 21.8 The full outcome details for the
Accountability path and the R value illustrating the
strength of the relationship between the outcomes

Figure 21.8 details

The starting point behaviours that are critical for driving the
Accountability path are Strategic, Sequential, Conformity, Team
Focus and Outer Focus. The first two behaviours are present at
the Place and for the People, so no interventions are needed.
However, Conformity, Team Focus and Outer Focus are absent
behaviours, so intervention design is needed as we cannot drive
down that path if the starting point is not achieved. Once that is
achieved, we can target People Focus and Empower to achieve
the rest of the path outcomes. Note that Active Learning is
needed, but it is already a parent behaviour.
To clarify what this means: the focus now must be on clear
accountability. Currently this organization has issues with
abiding by the rules and regulations. Accountability at team
level is not happening, so each team member will end up
making lone decisions rather than team decisions, and I do
want to remind you that in critical environments team
decisions are better for risk management. Finally, the
organization is not considering the impact on others as part of
their accountability.
Let’s turn our attention to the second path, which is Proactive
Safety Management. The simplified path is illustrated in Figure
21.9 and the detailed path with the outcomes is shown in Figure
21.10.
FIGURE 21.9 The Proactive Safety Management path,
also showing the behaviours’ presence needed to
drive that path

Figure 21.9 details

 FIGURE 21.10 The full outcome details for the
Proactive Safety Management path and the R value
illustrating the strength of the relationship between
the outcomes

Figure 21.10 details

It is interesting to note that inclusion is part of the Proactive
Safety Management path, something that has often been cited
by researchers as critical for good safety and risk management.
I support this theory as it makes good common sense – if a
certain cohort of people do not feel that they matter with no
sense of belonging, they are unlikely to take part in the safety
culture. If we analyse the behaviours of the second path, you
can see that the starting point is yet to be achieved as the
Relationship behaviour is absent.
In order to demonstrate the behavioural intervention design
needed and the return on investment, we should bring the
entire system together as illustrated in Figure 21.11.
 FIGURE 21.11 The behaviours that need intervention
design, planning and implementation as well as the
odds ratio for impacting the outcomes’ return on
investment

NOTE Each outcome box shows the likelihood of this outcome manifesting from the
intervention (i.e. 46 X = 46 times more likely to manifest if you have the presence of
behaviours and other outcomes

Figure 21.11 details

To enable change, this organization should design two
sequential intervention programmes. The first should focus on
the following behaviours:
Conformity (valuing clear rules and laws, and treating
people fairly according to these): The Place is seen to not
consistently value rules, laws and obligations, whereas the
People value rules and obligations at work. How can we
create opportunities to clarify what the rules and laws are
 and ensure employees consistently follow these within the
organization?
Outer Focus (focusing on the impact of work on the wider
environment): The Place is seen to be more internally
focused and does not consider the wider impact of work
on the environment in which it operates. How can we
create opportunities to focus on the wider impact of work
on the external environment?
People Focus (focusing on people rather than purely on
tasks and outputs): The Place tends to focus more on tasks
and outputs than on the people – leading to the
unintended consequence of making people feel like they
don’t matter. How can they ensure the organization and
its ways of working elevate the impact of work on people
rather than only focusing on tasks and output?
Empower (delegating decisions to lower levels with
supportive guidance): The Place is encouraging
hierarchical ways of working where leaders tend to retain
decision-making power. How can we create more
opportunities and motivation for better delegation and
empowerment?
Team Focus (collaborating group efforts and team results
are prioritized over individual results): The Place is seen
as encouraging high individual performance rather than
valuing achieving goals through partnerships and
alliances. How can we create opportunities to focus on
collective efforts and team results within the
organization?
I am sure you noticed that the above behaviours targeted for
change are all related to the Place. The interventions should be
focused on opportunity and motive.
The deliberate focus on driving these behaviours will get this
organization off the blocks for their starting points in their
respective paths, and will directly and significantly impact six
outcomes as shown in Figure 21.11.
The second intervention programme should work on the
following behaviours:
Relationship (valuing building strong relationships to
meet work objectives): The Place doesn’t value building
relationships as part of achieving work objectives. People
believe that they can work without having good
relationships. How can we create more opportunity and
motivation for building relationships to get work done?
Collective (valuing building strong relationships to meet
work objectives): The Place doesn’t value building
partnerships and networks to achieve goals. It seems to
elevate winning and individualism over collective effort.
How can we create more opportunity and motivation for
building partnerships and alliances to achieve work?
The first behaviour in the second intervention should target the
People and the Place, hence interventions should work on
capability, opportunity and motivation. The second behaviour is
related to the Place only, and the interventions should focus on
opportunity and motivation.
As with some previous case studies, you may well be
wondering what the ‘size of the ask’ for change is. As they
mostly need to target the Place, the best and quickest way to
demonstrate that is using the quadrant distribution graph.
Figure 21.12 illustrates the size of the entire ask, and the critical
seven absent behaviours.
 FIGURE 21.12 Quadrant distribution for the entire
sample size for the seven absent behaviours,
Conformity, Team Focus, Collective, Relationship,
Outer Focus, People Focus and Empower

Figure 21.12 details

Only 31 per cent of the entire organization exists in the top
right-hand side of the quadrant – the is where both the People
and the Place have these behaviours as present. This is low, but
at least we have a Place to learn these behaviours from. They
are the champions, if you like.
The lower right-hand side of the quadrant should be an
interesting target. Just over 51 per cent of the total sample exists
here, and since the People have the capability, if the
organization targets the Place for change they are onto a
winner. That population will move on up to the top right-hand
side of the quadrant, and the result will be 82 per cent of the
entire sample size in that quadrant, relatively low-hanging, and
just like that they would have significantly improved their
chances to gain 18 outcomes. Another demonstration of the
power of targeting the leading indicators for change.
 Behaviours form the bedrock of conduct and safety
workplace culture at nuclear generation organizations. The
World Association of Nuclear Operators is very aware of this
fact, and has dedicated much research to it. Applying the
correct behavioural interventions and, critically, knowing
which behaviours each organization should uniquely target,
will significantly increase the chances for proactive risk and
safety management while maximizing effectiveness.
The best evidence I can point to for this approach is the 1986
Chernobyl disaster which is still negatively impacting the
environment today. The lack of transparency and a culture of
authority caused a disaster and inhibited learning.3 The root
causes can be summarized as:
A fear of authority and the need to please political masters
resulted in a fear- and compliance-driven culture.
A pervasive culture of secrecy and denial bred in part by a
desire to preserve national pride obstructed the
acceptance of external help, stalling recovery efforts and
exacerbating the disaster’s effects.
A lack of psychological safety, resulting in operators not
speaking up about their concerns, was fundamental to the
causal chain of events that led to the disaster. Had
operators, decision-makers and politicians felt safe and
able to raise concerns and make decisions, the disaster
may not have occurred, or had it occurred, the impact
could have been mitigated.

Notes
1 WANO. Nuclear leadership effectiveness attributes, December 2019, www.wano.inf
o/resources/nuclear-leadership-effectiveness-attributes (archived at https://perma.c
c/3Z92-6C7R)
2 Refer to Figure 11.1 as a reminder of how to read this graph.

3 T Geraghty. Chernobyl: Safety cultures and secrecy, Psychological Safety, 23 June

2023, https://psychsafety.co.uk/chernobyl/ (archived at https://perma.cc/QME9-VN2
H)
22
What about agile culture? Does
that need the letter ‘C’?
Should you be embracing agility by nurturing an agile
organizational culture? If so, why? What is agile going to do for
you, and how do you even achieve agile?
In today’s fast-paced and dynamic environment, the concept
of agility has emerged as a critical factor for organizational
success. An agile organizational culture is characterized by
adaptability, collaboration and a continuous learning mindset.
Typical key questions for seeking an agile organization are
the benefits of an agile organizational culture, the key elements
that contribute to its development, and how organizations can
successfully transition to and then sustain agility.
Let us focus on the characteristics to articulate what each
means.

Flexibility and adaptability

At its core, an agile organizational culture is marked by the
ability to quickly respond to change. This involves being
flexible, adaptive and embracing uncertainty as an opportunity
rather than a threat. Agile organizations are adept at adjusting
strategies and operations in response to shifting market
dynamics.
A very useful exercise is to develop an opportunistic
behavioural model, as behaviours are the leading indicators, by
doing a behavioural mapping exercise similar to what you have
learnt in Chapter 4. This can be done as follows:
‘Ability to quickly respond to change’: The mapped
behaviour is Flexible (people value variability and deal
with each situation afresh) rather than Standardized
(people value working within clear processes and
systems).
‘Embracing uncertainty as an opportunity’: The mapped
behaviour is Radical (continuous improvement and
evolving ideas are valued; there is an emphasis on
responding differently to different situations) rather than
Moderate (individuals value working steadily within
assigned and streamlined processes favouring stability).
‘Adjusting strategies and operations in response to shifting
market dynamic’: The mapped behaviour is Outer Focus
(people believe that their environment controls them; they
want to work within their environment to achieve goals;
at work or in relationships they focus their actions on
others); rather than Internal Focus (people believe that
they can control nature or their environment to achieve
goals; this includes how they work with teams and within
organizations).
And remember to keep reading this characteristic title to
validate the mapping. In this instance the three mapped
behaviours, Flexible, Radical and Outer Focus, are a good match
for flexibility and adaptability.

Collaboration and cross-functional teams

Collaboration is a cornerstone of agility. Agile organizations
break down traditional silos and encourage cross-functional
teams that bring diverse skills together. This fosters a
workplace culture of shared responsibility and collective
problem-solving.
Let’s do the mapping exercise as follows:
‘Break down traditional silos and encourage cross-
functional teams’: The mapped behaviour is Team Focus
(work is delivered through collaborating group efforts and
team results are prioritized over individual results) rather
than Self Focus (individualism and independence are
encouraged and rewarded).
‘This fosters a workplace culture of shared responsibility
and collective problem-solving’: The mapped behaviour is
Collective (goals are achieved through partnerships and
alliances) rather than Individual (maintaining high
individual performance and winning prevail over
working relationships).

Iterative and incremental approach

Agile workplace culture embraces an iterative and incremental
approach to work. Rather than waiting for a perfect, fully
developed solution, agile teams deliver incremental value
regularly, allowing for continuous improvement and
adjustment based on feedback.
The behavioural mapping exercise gives us:
Synchronous (people see the past, present and future as
interwoven periods; they often work on several projects at
once in a fast-paced manner and view plans and
 commitments as flexible) rather than Sequential (people
like events to happen in order; they place a high value on
punctuality, quality, planning and staying on schedule; in
this culture ‘time is money’ and people don’t appreciate it
when their schedule is thrown off).
Radical – again, which indicates the importance of that
behaviour (continuous improvement and evolving ideas
are valued; there is an emphasis on responding differently
to different situations) rather than Moderate (individuals
value working steadily within assigned and streamlined
processes favouring stability).

Customer-centricity
Agile organizations prioritize customer needs and feedback.
This customer-centric focus ensures that products and services
are developed with a deep understanding of user requirements,
fostering stronger customer satisfaction and loyalty.
The behaviours are:
Outer Focus (people believe that their environment
controls them; they want to work within their
environment to achieve goals; at work or in relationships,
they focus their actions on others) rather than Internal
Focus (people believe that they can control nature or their
environment to achieve goals; this includes how they
work with teams and within organizations).
Strategic (the emphasis is on long-term delivery and
results, and focus is on the wider impact) rather than
Tactical (the emphasis is on short-term delivery and
results).
To bring the mapping into one view, see Table 22.1.
 TABLE 22.1 The resulting behavioural map for agile
organizational characteristics
Skip table
Collaboration and Iterative and
Mapped Flexibility and Cross-Functional Incremental Custom
Behaviours Adaptability Teams Approach Centri
Flexible X
Radical X X
Outer Focus X X
Team Focus X
Collective X
Synchronous X
Strategic X

Seven behaviours drive the four agile characteristics, and it is

interesting to note that Innovate is not one of those behaviours.
I note this as generally people tend to link agile with
innovation, but that’s not always true – you can drive agile with
or without the need to innovate for new products and services.
Hence the importance of understanding the difference, and this
is exactly why I did this mapping exercise as an illustrative
exercise.
Although these four characteristics and the behaviours are a
good base model, never purely borrow a ‘straight out of the
box’ model for your organization. As with previous examples,
always develop your own model for your own unique situation
and desired outcomes. I will shortly demonstrate why with the
next case study.
Before we turn to the case study, I would like to address the
‘why bother?’ question. What can agile do for you? Well that
depends on what you want to achieve that your current modus
operandi is perhaps not delivering, and possibly even working
against. However, general benefits can include:
Faster time to market: With a focus on iterative
development and quick response to changing
requirements, agile organizations can deliver products
and services to the market more rapidly, gaining a
competitive edge.
Enhanced customer satisfaction: By prioritizing
customer needs, gathering feedback regularly and
adapting products or services accordingly, agile
organizations are better positioned to meet and exceed
customer expectations, leading to improved satisfaction
and loyalty.
Improved efficiency: By simplifying complex processes
and procedures while being risk aware rather than risk-
evasive.
Increased creativity: Although innovation is not directly
linked, you can design an agile organization to benefit
from faster creativity as the environment will encourage
experimentation and learning from both successes and
failures.
Improved employee engagement: Agile cultures promote
a sense of purpose and autonomy among employees,
contributing to higher levels of engagement and job
satisfaction. When individuals feel empowered and
valued, they are more likely to contribute meaningfully to
the organization.

Context
A global oil and gas organization driving a major
transformation programme wanted to target an agile
methodology as part of their transformation. The organization
targeted 2,652 employees for this approach to evaluate the
effectiveness before a wider rollout.
The targeted employees were selected from several countries
to fully validate a global agile approach. The countries included
were the USA, Canada, Mexico, Trinidad and Tobago, the UK,
Azerbaijan, Georgia, Oman, the United Arab Emirates, Egypt,
Angola, India, Indonesia, Malaysia, China, Australia and New
Zealand. The total sample size per county is shown in Table
22.2.
 TABLE 22.2 The sample size per county targeted for
the agile methodology
Skip table

Country Sample Size

USA 356
Canada 192
Mexico 172
Trinidad and Tobago 156
UK 295
Azerbaijan 126
Georgia 113
Oman 120
United Arab Emirates 151
Egypt 108
Angola 98
India 137
Indonesia 110
Malaysia 120
China 186
Australia 123
New Zealand 89

The organization developed their own theoretical concept for

an agile characteristics framework. This was developed via a
collaborative effort utilizing a number of qualitative workshops
that included a significant representation from the above
targeted participants per country. The framework output from
these collaborative workshops included the current as well as
the targeted characteristic, a very useful exercise for all
participants to understand and evaluate the ‘size of the ask’
which can be compared against the actual quantitative
measurement.
 The participants also completed a behavioural mapping
exercise which cemented ‘how’ we get there, now that they
know the ‘what’. As with previous case studies, I will shortly
reveal the ‘why’ as I share with you their targeted outcomes.
Table 22.3 shows the developed agile characteristics, and
Table 22.4 shows the behavioural mapping.
TABLE 22.3 The organizational current state vs the
target agile characteristics
 TABLE 22.4 The behavioural mapping for both the
current organizational state as well as the targeted
agile state
Skip table
From individualism
CultureScope Hierarchy & Holding Back Limited Fixed M
Dimensions Siloed Thinking Ownership
Disenfranchise Vs Disenfranchise Disenfranchise
Empower
Active Learning Passive
Vs Passive
Learning
Collective Vs Individual Individual
Individual
Innovate Vs Consolidate Conso
Consolidate
Conformity Vs Nonconformity
Nonconformity
Team Focus Vs Self Focus
Self Focus
Achievement Vs Status Status Status
Status
People Focus Vs Delivery Focus Delive
Delivery Focus
Neutral Vs Neutral Neutra
Expressive
Moderate Vs Moder
Radical

To build a behavioural predictive model and a path to achieving

an agile workplace culture relevant to their target framework
in Table 22.4, the organization designed a number of outcomes
that will be used in the analytics to build the quantitative
behavioural model. These outcomes are:
 As a team we are transparent about how we work.
The organization is developing my skills for the future.
I am able to speak up if something is bothering me
without the fear of negative consequences or retribution.
I feel supported by my line manager to push back when I
face challenges related to workload.
I trust and respect the management team.
Often my colleagues and I wait until things have gone
wrong before acting to manage the impacts.
I apply my own judgement to the situation while following
risk policies and processes.
My manager emphasizes the importance of doing the right
thing even when working under pressure.
When faced with a difficult situation, I have confidence in
using my own judgement for decision-making.
Our policies are centred around helping me achieve my
goals and be effective.
I have regular opportunities to ask questions, give my
point of view and get my voice heard.
These outcomes have been tracked on a quarterly basis using a
Likert scale to measure them, up to the full deployment of the
CultureScope behavioural diagnostic.
The CultureScope deployment yielded a completion rate of 68
per cent with all participating countries achieving at least 38
per cent, hence a statistically valid sample.

Analytics and actionable insights

CultureScope data collection closed within four weeks,
achieving a total sample of n=1,803. The analytics then
commenced to establish a behavioural (leading indicator)
predictor for each outcome. This was achieved by combining
the outcome measures to the saturated 30 factors (15
dimensions) behavioural measurement and conducting the
hierarchical regression analysis to establish the best
parsimonious model for a restricted nested behavioural model.
The odds ratio method was then used to articulate the model
bench strength.
I can share with you some examples of the predictive models
as follows.

TRANSPARENCY
Outcome: As a team we are transparent about how we work.
Figure 22.1 shows the resulting regression analysis and odds ratio. The organization
is 60 times more likely to be transparent where Team Focus, Conformity, Empower and
People Focus behaviours are in the upper quartile of their presence.
Interpreting this model provides an interesting insight: transparency will happen
when all abide by the common rules fairly for teams and the individuals within each
team, and empowerment will play a significant part in that journey.
 FIGURE 22.1 The odds ratio for the transparency
outcome where Team Focus, Conformity, Empower and
People Focus behaviours are present
A bar graph shows the odds ratio for the transparency outcome at the
workplace where behaviours, team focus, conformity, empower and people focus
are present.

NOTE Multiple R=0.42, n=1,803

Figure 22.1 details

GROWTH MINDSET
Outcome: The organization is developing my skills for the future.
The resulting regression and odds ratio is shown in Figure 22.2. This significant model
shows that skills development for the future is 43 times more likely to occur where
Active Learning, Team Focus, Conformity and Strategic are in the upper quartile of their
presence. Additionally, the absence of these behaviours will mean that the organization
is 13 times less likely to achieve a growth mindset.
The organization was very pleased to see both Active Learning and Strategic
featuring in this model. Essentially, the organization has to create equal opportunities
for all, and learning and development must serve the long-term strategy while also
serving the team’s needs.
 FIGURE 22.2 The odds ratio for the growth mindset
outcome where Active Learning, Team Focus,
Conformity and Strategic behaviours are present
A bar graph shows the odds ratio for the growth mindset outcome at the
workplace where behaviours, active learning, team focus, conformity and
strategic are present.

NOTE Multiple R=0.55, n=1,803

Figure 22.2 details

You may have already noticed that the behavioural leading

indicators are so far matching the behavioural mapping created
for the agile framework in Table 22.4. This is indeed good news
as it is validating the framework.

EMPOWERED TEAMS
Let us examine two outcomes.
Outcome 1: Our policies are centred around helping me achieve my goals and be
effective.
Four behaviours are the leading indicators; they are Active Learning, Conformity,
Strategic and Outer Focus. This organization is 21 times more likely to drive
effectiveness where their policies enable the presence of these behaviours, and 14
times less likely to drive effectiveness if their policies inhibit these behaviours.
 FIGURE 22.3 The odds ratio for the driving
effectiveness outcome where Active Learning,
Conformity, Strategic and outer Focus behaviours are
present
A bar graph shows the odds ratio for increasing effectiveness outcomes at the
workplace where behaviours, active learning, conformity, strategic and outer
focus are present.

NOTE Multiple R=0.49, n=1,803

Figure 22.3 details

Outcome 2: I am able to speak up if something is bothering me without the fear of
negative consequences or retribution.
Three behaviours are the leading indicators for psychological safety; they are
Conformity, Expressive and People Focus. This organization is 47 times more likely to
drive psychological safety where these behaviours are in the upper quartile of their
presence, and 9 times less likely to drive psychological safety when these behaviours
are absent.
Actually, it is reassuring to see Expressive behaviour being part of the leading
indicators for psychological safety. Additionally, Conformity means that the rules are
clear and applied fairly to all, and our people matter as they are the reason we will get to
our outcomes. Hence, People Focus is the third leading indicator.
 FIGURE 22.4 The odds ratio for the psychological
safety outcome where Conformity, Expressive and
People Focus behaviours are present
A bar graph shows the odds ratio for psychological safety outcomes at the
workplace where conformity, expressive and people focus behaviours are present.

NOTE Multiple R=0.47, n=1,803

Figure 22.4 details

As demonstrated, the multiple R for this effect size obtained for the developed models
outranks at the 80th percentile of effect sizes observed in the research for employee
attitudes.1 In fact, for all of the designated outcomes, the models yielded a multiple R
ranging from 0.41 to 0.55, providing significant models.

Path analysis
Having established the leading indicators for each outcome, a
path analysis was established using the structural equation
modelling methodology as described in Chapter 10.
The methodology delivered one integrated path for agile;
however, it can be described as two branches with a common
starting point as demonstrated in Figure 22.5. The upper path is
related to managing risks for agile workplace culture, and the
lower path is about effectiveness and achievement for agile
culture.
The most interesting insight for this path is that psychological
safety is a key starting point for this organization, without
achieving the starting point. Additionally it is fascinating to see
that the key behaviours of the starting point that will impact the
entire agile path are Team Focus, Empower, Conformity and
Expressive.
 Let us think about what this is telling us. Empowered teams
that are able to express their opinions openly and will abide
fairly by the rules and regulations will have a significant impact
on driving agile for this organization. This fact did ‘myth bust’,
as some leaders in the organization told me: ‘We bet you that
Innovate will be a key starting behaviour’. The dangers of
baseless assumptions!
 FIGURE 22.5 The full agile path analysis with the
leading behavioural indicators
A flow chart shows the integrated agile path analysis with
the behavioural indicators.

Figure 22.5 details

This is a good point in the analysis journey to compare the
evidence-based predictive behaviours versus the constructed
agile framework mapped behaviours this organization
constructed before the CultureScope diagnostic deployment.
Table 22.5 shows the original mapping for current state versus
future agile state; however, I have added a column for the
predictive behaviours which does show a reasonably good
match.
The organization needs to eliminate two behaviours from the
mapping as these are not part of the unique journey for agile.
The behaviours are Achievement and Radical, and they need to
include Strategic behaviour. As the predictive analytics have
specified, this behaviour should be mapped to ‘Empowered
Teams’ and ‘Accountability’ in their agile framework.
 TABLE 22.5 The behavioural mapping for the current
organizational state as well as the targeted agile state
A table lists the mapped behaviours for the current
organizational state and the targeted agile state.
NOTE This table includes the evidence-based predictive behaviours to achieve agile

Table 22.5 also indicates which behaviours are present,

indicated by a ‘tick’ icon next to any present behaviours, and
this illustrates that this organization is still at a very early stage
in their agile journey. In fact, we know from the path analysis
that the starting point is yet to be achieved; hence the left-hand
side of Table 22.5 is well populated with ‘tick’ icons and the
right-hand side of the table is rather bare.
To help this organization develop their behaviour-focused
intervention design and planning, I refer you to Figure 22.6
which shows a three-stage sequential intervention plan as
follows:
Stage one: For the People, Expressive capability building.
For the Place, opportunity and motivation to enable
Empower and Conformity.
Stage two: For the People and the Place, so capability,
opportunity and motivation, to enable People Focus,
Active Learning, Collective and Outer Focus behaviours.
Stage three: For the People and the Place, so capability,
opportunity and motivation, to enable Innovate
behaviour. This behaviour has the most drag as it is
significantly absent at the Place and for the People.
 FIGURE 22.6 The three-stage sequential intervention
plan and the odds ratio likelihood for the return on
investment
An illustration shows the sequential intervention plan in
three stages and the odds ratio likelihood for the return on
investment.
NOTE *Likelihood of this outcome manifesting from a behavioural intervention (i.e.
23 × more likely to manifest if you have the presence of behaviours and other
outcomes)

Figure 22.6 details

These findings were provided to the dedicated organizational
development function, and as I write this chapter, they are busy
with the design and action planning. Indeed, they have found
this to be both insightful and actionable as without this
approach, it would have been impossible to evaluate or validate
the mapped behaviours, or even know which behaviours they
need to focus on and where in the organization, by virtue of the
quadrant distribution method.
Another interesting learning in this case study is that I have
demonstrated how one can develop a multiregional
behavioural model yet still be able to provide a country-by-
country intervention action plan uniquely designed for where
each country is against the leading behaviours and the path
model.
I can understand that you may think agile will be harder to
deliver in large organizations and this can be further
exacerbated by having multiple regions and locations. The
methodology outlined in this case study should reassure you
with the art of the possible.
 As I have often stipulated, the uniqueness of each
organization is driven by different circumstances, events and
operating environments, so one must assess their needs for
agile and what agile will do for them. However, the evidence is
all around us that we are in a world where change is fast
becoming a constant, driven by technology improvements such
as artificial intelligence, shifting risk dynamics and talent
needs. Fostering an agile organizational culture of some sort is
not just a strategic choice; it may well be a necessity.
Revisiting the title for this chapter – I hope I have made my
case for why agile also needs the letter ‘C’.

Note
1 Many researchers use Jacob Cohen’s correlations effect size benchmarks to gauge
the strength of relationships. Those benchmarks are 0.2 for a small effect, 0.3 for
moderate effect and 0.5 for a large effect. The results of our model for predicting
employee voice would benchmark as a large effect size (R=0.496). More recent
research suggests that the size of an effect varies depending on what is being
studied and that different benchmarks reflecting different fields of study are
required. We obtained our 80th percentile benchmark for research on employee
attitudes from F A Bosco, H Aguinis, K Singh, J G Field and C A Pierce (2015)
Correlational effect size benchmarks, Journal of Applied Psychology, 100, pp. 431–
49
23
The crucial role that workplace
culture plays in shaping
employee wellbeing
I have come across a plethora of wellbeing initiatives all
claiming to improve employee wellbeing at work. Such
initiatives can address physical wellbeing, mental wellbeing,
social wellbeing and work/life balance. However, these
initiatives fail to address the primary root cause, which is
workplace culture.
Allow me to indulge you by providing an example. An
organization may provide its employees with a gym
membership claiming to improve both physical and mental
wellbeing. Sounds like a great initiative, doesn’t it? However,
this organization is not adjusting workload so employees can
actually take advantage of this the gym membership. In fact,
this organization may well be expecting that employees should
achieve more, now that they have provided them with an extra
motivational perk. I will go as far as to say that if any employee
complains about the amount that they are expected to achieve
against unrealistic targets, they will be quickly reminded of
how lucky they are to be part of an organization with such
great perks like the gym membership.
Needless to say, and as an unintended consequence, employee
wellbeing initiatives can be weaponized against individuals’
ability to speak up, affecting their psychological safety.
 In this case study, I want to demonstrate how workplace
culture for wellbeing is about behaviours as the leading
indicators for sustainable wellbeing, and should be your
primary focus ahead of any non-evidence based initiatives. Any
interventions that include employee perks should be planned
and designed in line with intended workplace culture and
behaviours.
There are several research studies connecting employee good
wellbeing with increased productivity, employee retention,
brand enrichment and so on, but what they do not focus on is
‘how’ to achieve an organizational culture for wellbeing.

Context
A competitive telecommunications operator, based in the
Middle East and part of a large regional group, had been
through a significant transformation programme that yielded
great financial results and improved new customer acquisition
and retention. This transformation and its results happened in
a record time of just seven months, resulting in improved
cashflow and profitability.
The CEO knew that this may have taken its toll on employees,
and wanted to see where they were at in terms of employee
wellbeing and how to improve this through workplace culture.
Specifically, the objectives were as follows:
Provide predictive analytics and a path analysis model on
achieving psychological wellbeing and work/life balance.
Design, plan and implement an actionable roadmap to
drive sustainable wellbeing.
To achieve this, I worked with this organization to design a
number of outcome measures which will help drive the
analytics and behavioural insights. They were:
I have regular opportunities to give my point of view and
get my voice heard.
Where I work, the same standards apply to everyone.
Where I work, people can state their opinion without the
fear of negative consequences.
In my team, I feel safe voicing problems and tough issues.
My manager emphasizes the importance of doing the right
thing even when working under pressure.
I feel supported by my line manager to push back when I
face challenges related to workload.
Our policies are centred around helping me achieve my
goals and be effective.
I feel my organization is sufficiently supporting my health
and wellbeing in the current environment.
I feel able to take time off work when I experience a
physical health concern.
I am often under constant strain at work.
I feel able to take time off work when I experience a
mental health concern or personal or family issues.
The outcome measure questions were deployed. At the same
time, a CultureScope behavioural diagnostic was launched
targeting the entire organization which totalled 756 employees.
The diagnostic period was two weeks, and this achieved a 52
per cent completion rate; that is a total n=395, representing a
good sample size for statistical and analytics validity.

Analytics and actionable insights

As soon as the CultureScope diagnosis was completed, the
analytics work started by combining the 11 outcome measures
with the saturated 30 factors (15 dimensions) behavioural
measurement and conducting the hierarchical regression
analysis to find the best parsimonious model to establish a
restricted nested behavioural model for each outcome. As with
previous case studies, the odds ratio method was then used to
articulate the model bench strength per outcome.
The achieved multiple R ranged between 0.38 and 0.59 for all
the outcomes which is an excellent model effect size.
I can share some example outcomes’ odds ratio to
demonstrate some of the behavioural predictive power as
follows.

WHERE I WORK, THE SAME STANDARDS APPLY TO

EVERYONE
Figure 23.1 indicates that where Conformity and Team Focus are in the upper quartile
of their presence, this organization is 25 times more likely to apply the standards fairly
to all.
Seeing Conformity as part of this model is not a surprise, and is in fact reassuring.
However, it’s interesting to see Team Focus behaviour rather than People Focus. So, for
this organization, having an egalitarian system should be applied at team level rather
than individual level.
Additionally, the absence of these two behaviours will mean that the organization is
38 times less likely to achieve this outcome. Note the significant multiple R of 0.59.
FIGURE 23.1 The odds ratio for applying the standards
outcome where Conformity and Team Focus
behaviours are present

NOTE Multiple R=0.59, n=395

Figure 23.1 details

I FEEL MY ORGANIZATION IS SUFFICIENTLY
SUPPORTING MY HEALTH AND WELLBEING IN THE
CURRENT ENVIRONMENT
Three behaviours predict this outcome: Conformity, Team Focus and Strategic. The
organization is 9 times more likely to achieve this wellbeing outcome where these
behaviours are in the upper quartile of their presence, and 22 times less likely to
achieve this outcome where they are absent.
As seen in Figure 23.2, it is interesting to note the significance of the achieved
multiple R of 0.55 for this model. It is not surprising to see Conformity here; however,
it’s interesting that Team Focus, yet again, staged an appearance in this significant
model. Perhaps this supports a recent study on the science of teamwork, which states
the importance of Team Focus for wellbeing.1
It is also good to see Strategic behaviour as part of this model, as wellbeing should be
part of the long-term and wider impact rather than a quick fix.
 FIGURE 23.2 The odds ratio for supporting employee
wellbeing outcome where Conformity, Team Focus and
Strategic behaviours are present

NOTE Multiple R=0.55, n=395

Figure 23.2 details

Path analysis for wellbeing

Having completed the predictive analytics for each of the
outcomes, the path analysis was constructed using the
structural equation modelling methodology. As a result two
paths were developed: path 1 for Psychological Safety, and path
2 for Wellbeing.
Figure 23.3 shows the first path. The behaviours for that path
are marked, and any present behaviours are also marked with
a tick.
 FIGURE 23.3 The first established path for
Psychological Safety along with the behavioural
leading indicators

Figure 23.3 details

Interesting to see that the starting point for this path is the
opportunity to be heard. However, the starting point
behaviours are absent, hence this organization has not yet
established the starting point for the first path.
The second path dedicated to Wellbeing can be seen in Figure
23.4. Although this path shares some behaviours with the
Psychological Safety path, it has its own distinct existence. As
can be seen, the starting point demonstrates exactly what I
mentioned at the beginning of this case study: leaders and
managers must think about workload as the root cause of
wellbeing issues, and any employee perks will not resolve this
issue.
Notice that two of the starting point behaviours for the
Wellbeing path, that is Expressive and Team Focus, are shared
with the Psychological Safety path, further emphasizing for this
organization the role that these two behaviours play in
achieving their optimal workplace culture.
 FIGURE 23.4 The second established path for
Wellbeing outcomes as well as the behavioural
leading indicators

Figure 23.4 details

The starting point behaviours for the Wellbeing path are all
absent, hence this organization has yet to establish the first
outcome in that path.
We can bring the entire system into one, as illustrated in
Figure 23.5. This shows that the organization needs to design
and deploy behavioural interventions in two stages. The first
stage will focus on the following behaviours:
Expressive: Capability-type interventions are needed as it
is People-related.
Conformity: Opportunity- and motivation-type
interventions are needed as this is absent at the Place.
Team Focus and People Focus: Capability-, opportunity-
and motivation-type interventions are needed to act on
the People and the Place as these two behaviours are
absent for both.
These interventions will get this organization on their way
down both paths, as these behaviours will work on both
starting points.
These second stage will need to focus on the following
behaviours:
Empower: As this behaviour is absent at the Place, the
inventions will need to be opportunity- and motivation-
focused.
Innovate: This behaviour is absent for the People and the
Place, so the interventions design will need to address
capability, opportunity and motivation.
I would like to point out that having the Innovate behaviour
appearing as part of the wellbeing model is very unusual and
unique to this organization. In this context it appeared as part
of the ‘the policies are centred around helping me to be
effective’ outcome model. That made sense to this organization
particularly as they want to implement innovative ways of
working to help employees be more effective and have a better
work/life balance.
 FIGURE 23.5 The absent behaviours, and the
proposed behavioural intervention stages for the
Psychological Safety and Wellbeing paths as well as
the return-on-investment likelihood

Figure 23.5 details

Innovate behaviour did also appear in the Psychological Safety
path, particularly for the ‘People at work are able to state their
opinion without the fear of negative consequence’ outcome.
Thought-provoking? I hope so; however, when I played this
back to the organization, they were very pleased. In their
context this is all about people’s ability to bring new ideas and
ways of working without fear. This type of speak-up culture is
known as promotive rather than prohibitive. I define the two
different types as follows:
Prohibitive is defined as the ability to report concerns
that may lead to negative outcomes. This could include
speaking up about safety issues or whistleblowing;
essentially, speaking up to prohibit undesired outcomes.
 Promotive is defined as the ability to speak up about ideas
that will help progress with the intention of supporting
desired outcomes.
I hope you will see why Innovate behaviour may well exist in
the second type of speaking up.

Intervention design and planning

Having delivered the analytics and actionable insights, I had the
opportunity along with one of my colleagues to get involved
with the first-stage behavioural intervention design and
planning, which I can share with you.
Before we get on with any design, and as I have regularly
articulated, we need to examine the organizational ‘where’.
This will help target our approach by knowing ‘where’ to learn
the behaviours from and ‘where’ to implement interventions.
To help demonstrate this approach, I have constructed a table
that shows the four targeted starting point behaviours as a
combined cluster and individually. You will be able to see the
percentage behaviour presence distribution pre-function. See
Table 23.1.
TABLE 23.1 The starting point behavioural
distribution per functional area with the organization
Skip table
Starting Point ALL Behaviours EXPRESSIVE TEAM FOCUS PEOPL
Percentages of 71% HR & Corp 43% HR & Corp 57% HR & Corp 12% H
employees falling Aff. Aff. Aff. Aff.
in the top-right 45% Engineering 33% IT 52% IT 15% Fi
green corner of 41% Commercial 22% Commercial 34% Commercial 10% C
the dispersion 41% IT 21% Engineering 33% Engineering 7% Eng
quadrant, where 30% Finance 10% Finance 15% Finance 0% IT
the behaviours
are present both
for the Place and
the People (learn
from these)
Percentages of 40% Finance 60% Finance 35% Finance 35% Fi
employees falling 23% Commercial 50% HR & Corp 28% Engineering 23% C
in the bottom- 20% Engineering Aff. 26% IT 19% E
right amber 18% IT 49% Commercial 23% Commercial 15% IT
corner of the 7% HR & Corp 48% IT 7% HR & Corp Aff. 7% HR
dispersion Affairs 47% Engineering
quadrant, where
the behaviours
are present for
the People, but
absent for the
Place (implement
here first to have
a conducive
environment)
EXCEPT FOR
EXPRESSIVE
where it’s the top-
left corner of the
quadrant
approach, hence ’I
don’t’, but the
Place ’does‘.
Let us examine Table 23.1. Look at the second row – this row
will show us the behavioural champions by function.
Expressive behaviour can be learned from HR & Corporate
Affairs, Team Focus behaviour can be learned from HR &
Corporate Affairs as well as IT. Conformity behaviour can be
learnt from IT, Finance, Engineering and HR & Corporate
Affairs, as well as Commercial. People Focus behaviour has no
obvious champions.
The third row of Table 23.1 would indicate ‘where’ we need to
focus our interventions within the organization.
We must also examine the potential gain from targeting the
interventions at the Place or the People using the quadrant
distribution method. See Figure 23.6. For the Expressive
behaviour, the idea is to move as many People as possible from
the top left to the top right of the quadrant. That movement is
shown by the superimposed arrow.
 FIGURE 23.6 The quadrant distribution for the
Expressive behaviour functionally across the
organization

Figure 23.6 details

As an example, if we drive the Expressive behavioural change
for the Commercial function, 49 per cent of that population will
move to the top right side, joining the 22 per cent already in
that zone, resulting in 71 per cent of the total population with
Expressive behavioural presence for People and Place.
Figures 23.7, 23.8 and 23.9 demonstrate the same approach
for Team Focus, People Focus and Conformity behaviours
respectively.
FIGURE 23.7 Quadrant distribution for the Team
Focus behaviour functionally across the organization

Figure 23.7 details

FIGURE 23.8 Quadrant distribution for the People
Focus behaviour functionally across the organization

Figure 23.8 details

 FIGURE 23.9 Quadrant distribution for the
Conformity behaviour functionally across the
organization

Figure 23.9 details

Looking at these distribution quadrants can really help to
demonstrate the gain from the intended interventions, but will
also illustrate the behaviours that may have the most
significant ‘drag’. Look at Figure 23.8 showing the People Focus
behaviour. Examine the bottom left zone – functions in that
zone have that behaviour absent for both Place and People, and
I am sure you will notice the percentages are by far the highest
in that zone for People Focus behaviour. That is what I mean by
the most ‘drag’.
This quadrant approach helped us to design several
investigative workshops and importantly to target these
workshops precisely for selected functions based on which
behaviours are present or absent, and helped us gain
significant insight for the context, the capability as well as the
opportunity enablers/inhibitors and motivation. Figures 23.10,
23.11, 23.12 and 23.13 show the output from these workshops
for Expressive, Team Focus, People Focus and Conformity
behaviours respectively. These outputs are very useful
templates that you can use for any intervention pre-design.
FIGURE 23.10 Workshop output for the Expressive
behaviour

Figure 23.10 details

FIGURE 23.11 Workshop output for the Team Focus
behaviour

Figure 23.11 details

FIGURE 23.12 Workshop output for the People Focus
behaviour

Figure 23.12 details

 FIGURE 23.13 Workshop output for the Conformity
behaviour

Figure 23.13 details

The outlined qualitative yet targeted approach derived from the
workshops outputs was used for the intervention design. These
interventions are listed in Table 23.2 and include the full
approach for targeting people, process, communications and
technology to drive behavioural change.
TABLE 23.2 All intervention designs
Skip table
Intervention Behaviour People Process
Employee Continuous Expressive Team
Initial training to New E
Improvement (CI) Tool Focus
introduce the tool contin
(Or Online Suggestion
and how to use it improv
Box)
and to drive proces
acceptance to emb
promo
system
Agend
include
Team m

TEAM Psychological Expressive Team All Level I & 2 employees

Needs
Safety Training Focus via experiential
to CO
workshops
compe
Introd
out’ fo
Safety
Intervention Behaviour People Process
Active Listening for Expressive
All Level 2 and Added
Leaders
above leaders leader
Initial training compe
Ongoing Coach
coaching/mentoring needs
writte
trainin

Team Reward System Team Focus

All Team members Rewar
need to be rated not or bon
just for individual must r
performance but Team K
also for Team achiev
performance results
individ
perfor

Time Management People Focus This is a self

Training on tool use
Training and Tool managemen
provide awa
where there
can be saved
the manager
team
Intervention Behaviour People Process
Leadership People Focus CEO & Board
Nil
Communication Around
People Being as
Important as Task in
Team Dynamics

Time Out for People – People Focus All – leader led

Daily t
Regular Time Each Day
People
to Focus on the How Not
manda
Just the What
meetin
where
stock

360 Leadership Reviews People Focus

All Leaders Clear m
Concentrating On How
the HO
Things Get Done Not
Integr
What
part of
annua

Leadership & Ethics Conformity Workshops

All Leaders
Training People Focus concentratin
leadership
 Intervention Behaviour People Process
Review Core People Focus
All Review
Competency/Leadership Expressive Team
trainin
Training Focus Conformity
compa
focus o
highlig
which
four de
behav

Rules & Responsibilities Conformity

All Conse
Comms Campaign
not fol
rules m
clear
Proces
to ens
Confo

In collaboration with the organizational leadership team, a

detailed implementation time plan was established to cover 12
months, which included a remeasure after six months. The
summary time plan can be seen in Figure 23.14.
 FIGURE 23.14 The interventions implementation time
plan

Figure 23.14 details

As a key takeaway, when it comes to wellbeing, do take a
detailed look at workplace culture first, and before you spend a
lot of cash on employee perks that may yield negligible results.
These perks may sound like a great idea, but there is limited
evidence for how they will work on the root cause of employee
wellbeing issues.

Note
1 S S Johnson (2021) The science of teamwork, American Journal of Health Promotion,
35 (5), pp. 730–32, https://doi.org/10.1177/08901171211007955a (archived at https://
perma.cc/C7BS-DGFD)
24
Can we make organizational
culture the hero it deserves to
be?
This question sounds so simple, yet it has been the most
significant driving force as to why I have dedicated myself,
since 2007, to finding a practical and actionable answer to this
exact question.
By virtue of the plethora of evidence available through
horrific cases, which are still unfortunately happening, the way
we have been thinking and actioning workplace culture has not
been working.
Why so? Simply, organizations have been badly served with a
lack of valid leading indicators measurement and at best the
reliance on lagging indicators. Outcomes and employee
engagement are important measures, but they are lagging
indicators, and by the time the outcome has manifested you are
already too late.
I do understand that the definition of workplace culture can
be complex, as it includes the shared beliefs, values, attitudes
and norms that exist within an organization. However, all of
these manifest in behaviours, which is the evidence-based
approach to measuring organizational culture and was the
result of the initial seven-year research I shared with you in
Chapter 4.
 This misconception is unsurprising as the narrative peddled
by countless consultancies and thought leaders is one that
confuses engagement, morale and culture as one and the same.
However, this couldn’t be further from the truth.
Organizational culture is, in its purest form, ‘how we do
things around here’. These are the unspoken norms and
behaviours that members of an organization instinctively
follow. When an individual joins a new company, they quickly
learn these, adjusting and adapting, often without even
realizing it. We quickly figure out these unwritten behavioural
rules and adapt our behaviour to fit in, or we may leave.
Let’s look at the three essential revelations every leader must
grasp to truly harness the power of culture.

1. To drive success, stop focusing on results

alone and focus on the behaviours that drive
these
Leaders get so caught up in wanting to drive outcomes such as
‘we need to increase performance’, ‘we need to enhance
wellbeing’, ‘this quarter we are going to drive engagement’, etc.
All of the focus, intervention planning and resources are put
into trying to enhance these outcomes, but little goes into
accurate, precise attention and focus on what moulds these
outcomes – behaviours.
To steer any vessel effectively, one must understand the
difference between the destination and the course. Outcomes,
like improved performance or heightened engagement,
represent the desired destinations. Behaviours, on the other
hand, are the courses charted – patterns of actions or
interactions that drive an organization toward these
destinations.
Critically, behaviours manifest in two distinct and
interconnected ways:
1. Individual behaviour: This pertains to how single
members of the organization act and respond. It is driven
by individual capability and the willingness to act in
certain ways. For instance, an employee’s dedication to
continuous learning or their responsiveness to feedback.
2. Organizational behaviour: This reflects the
organizational environment. It’s the behaviour that
employees observe around them and is a powerful driver
or inhibitor of individual behaviour. This pertains to
organizational processes, systems and established ways of
working. For example, the organization’s informal reward
systems that give accolades to employees who support
their colleagues.
Here’s why focusing on behaviours is transformative:
Behaviours are tangible and modifiable: Unlike
outcomes, behaviours are observable. They can be
identified, assessed and directly adjusted. By spotlighting
behavioural patterns, organizations can craft actionable
strategies to reinforce or rectify them.
Predictive power: Behaviours often serve as leading
indicators. Before any outcome materializes – be it good or
bad – there’s usually a consistent pattern of behaviours
leading up to it. Recognizing and addressing these patterns
early can help shape desired outcomes.
 Inclusivity and empowerment: While outcomes might
seem distant and broad, behaviours are immediate and
personal. By fostering a focus on behavioural standards,
every employee, regardless of their role, has a stake in
shaping the culture.
While outcomes present a snapshot of an organization’s current
state, it’s the behaviours that shape its future trajectory. For
leaders aiming for sustained success, the formula is
straightforward: prioritize behaviours as primary levers and
allow outcomes to naturally follow.
In essence, to truly influence results and cultivate a thriving
organizational culture, leaders need to shift their gaze from the
distant horizon and focus on the steps right in front of them.

2. Off-the-shelf models and values do not work

Delve into any leading management book, white paper or
keynote speech and you’ll likely stumble upon models or values
presented as the universal remedy for organizational
challenges. Yet, as my research revealed, the reality that
emerged was sobering: every organization’s behavioural
blueprint to drive success was different.
The idea that universal strategies can be transplanted
successfully across varied organizational contexts is a fallacy.
Each organization’s distinct context, strategy, people and
operating model sculpt a unique culture and behavioural
ecosystem.
Consider the interplay of behaviours and outcomes as an
intricate web, where every thread holds significance. This web
isn’t manufactured in bulk, but is a unique tapestry for each
organization, interwoven with its specific nuances and
contextual factors.
The implications of this revelation are two-fold:
Avoid copy-paste culture strategies: Just as a
neighbour’s key won’t fit your front door, their
organizational strategies won’t necessarily unlock success
in your company. Emulating external models without
adapting them to your unique context is a misstep.
Beware the off-the-shelf consultants: While external
guidance can provide invaluable insights, it’s crucial to
ensure consultants are tailoring strategies to your
organization, rather than deploying a generic playbook.

3. Leaders are the custodians of organizational

culture, but every employee is accountable for it
For all leaders, that will mean being behavioural champions;
however, it also means creating the workplace environment to
enable all employees to champion the pursued behaviours. This
is why you should adopt workplace Culture by Design.
If you don’t actively set the behavioural standards for your
organization, you might end up with negative behaviours
which can harm your work environment. But with a clear plan,
you can build a culture that matches your goals and ensures
success.
To enable you to design the culture you need to drive success
and stay out of trouble, follow the six phases outlined as
follows.
Phase 1: Review your organizational strategy to
derive critical organizational outcomes
Connecting culture to strategy is an obvious first step for most,
but it is rarely done with the rigour and precision that is
required. Start by reviewing your organizational strategy and
mission, highlighting key objectives and goals. Ask questions
like, ‘What tangible results are we aiming for in the next year?’,
‘What employee or customer experiences do we want to drive?’
and ‘What do we want to control and minimize around risk,
disruption and change?’
From this, outline and populate the specific outcomes you
wish to achieve, such as increased sales, improved
psychological safety, enhanced customer satisfaction or high
compliance with safety processes.

Phase 2: Assess desired outcomes and identify key

behaviours to achieve them
As you progress, identify the specific behaviours – individual
behaviour (capability and willingness) and organizational
behaviour (the systems, processes and ways of working) – that
may be needed to achieve these outcomes. Keep these
behaviours specific and actionable. For every individual
behaviour, there should be a corresponding organizational
behaviour to drive and reinforce it.
For example, to drive the outcome of psychological safety,
instead of ‘employees speaking up’ as the behaviour needed to
drive this outcome, more specific and actionable behaviours
would be ‘capability and willingness to work with others to
achieve goals’ (individual behaviours) and ‘rewarding those
who speak up and voice concerns’ (organizational behaviour).
This direct approach ensures that every strategic element has
corresponding outcomes and behaviours, setting the stage for a
culture that is both purposeful and aligned with your core
objectives.

Phase 3: Measure the outcomes and behaviours to

connect the dots
Achieving great things is always the result of various
behaviours and outcomes working together. Almost like a web.
If you want to achieve excellent risk management, you will
need to drive multiple behaviours and outcomes to get there.
Every organization is unique and so the behavioural
roadmap to desired outcomes will be unique to you. Therefore,
it is critical to understand your own behavioural roadmap. This
requires robust measurement of the outcomes and behaviours.
From this data, you can start to understand how specific
behaviours predict outcomes, and you can strategically focus
your interventions to maximum impact.

Phase 4: Plan and implement impactful

interventions and organizational changes
To achieve meaningful change within an organization,
interventions need to be meticulously planned and executed.
Below we outline the essential ingredients for effective
intervention planning.

OWNERSHIP FROM SENIOR LEADERS

Any intervention, regardless of how well-crafted, will only be as
effective as the support it receives. It’s essential that senior
leaders not only endorse but actively participate in the
intervention design process. Their involvement underscores the
intervention’s importance, ensuring broader organizational
buy-in.

ACCOUNTABILITY FROM KEY STAKEHOLDERS

Ownership ensures endorsement, but for an intervention to
truly take off, key stakeholders must be accountable for driving
it forward. These stakeholders act as torchbearers, ensuring
that the interventions are implemented and maintained as
intended, and that feedback loops are in place to measure
progress and make necessary adjustments. These could be from
relevant departments or identified change champions.

THE THREE PILLARS OF FOCUS – CAPABILITY, OPPORTUNITY

AND MOTIVATION
For individuals to adopt new behaviours or ways of working,
they require three essential elements:
1. Capability: This refers to an individual’s ability to enact
the desired behaviour. Does the person possess the
necessary skills, knowledge and physical ability? When
planning interventions, it’s vital to ensure that people
have the requisite capabilities or that training is provided
to equip them with those capabilities. For example, if
you’re looking to shift your team from tactical thinking to
a more innovative mindset, it’s not enough to simply tell
them to ‘think outside the box’. Practical steps might
 involve offering workshops in creative problem-solving or
bringing in guest speakers from industries known for
innovation. By doing this, you are building their capability
to think innovatively.
2. Opportunity: This pertains to the workplace factors – the
environment, resources and circumstances – that make a
behaviour possible. Organizational systems, processes and
ways of working can either facilitate or hinder these
opportunities. Interventions must address these
organizational aspects to pave the way for desired
behaviours. For instance, if an organization wants to
promote collaborative work instead of teams working in
silos, the environment should be conducive to that. This
could mean reconfiguring office spaces to have more
open-plan areas or shared workstations, introducing
software that facilitates teamwork, or establishing regular
inter-departmental brainstorming sessions. These
interventions create opportunities for collaboration.
3. Motivation: This is the driving force behind behaviour. It
encompasses the intrinsic and extrinsic incentives that
drive people to act in a certain way. When planning
interventions, consider both the personal motivations of
employees and the broader organizational incentives that
can be leveraged. Consider a scenario where you want to
shift the reward system from one that values status to one
that celebrates achievement. Instead of providing perks
based on seniority or titles, you could introduce an
‘Innovator of the Month’ award. This recognizes and
rewards individuals who have come up with
groundbreaking ideas, irrespective of their position in the
 company hierarchy. By aligning rewards with
achievement, you’re bolstering motivation to act in the
desired way.
It’s paramount to direct interventions where they’re most
needed. This means ensuring that efforts targeting individuals
build capability, while those addressing the organizational
framework enhance opportunity and motivation. The precision
of these targeted interventions maximizes their effectiveness
and ensures that resources are utilized optimally.
By integrating leadership ownership, stakeholder
accountability, a deep understanding of capability, opportunity
and motivation and the insights from detailed measurements,
you’ll be well-positioned to drive impactful, lasting change
within your organization.

Phase 5: Seeing your interventions in action

Designing an intervention is only half the journey. The real test
lies in implementing these strategies effectively. Here’s how to
move forward with this crucial phase:
Strategic roll-out: Before introducing any interventions,
ensure there’s a comprehensive roll-out strategy in place.
This includes setting clear timelines, designating
responsible teams or individuals and defining milestones.
Communication is key during this phase: inform all
relevant stakeholders about the impending changes, the
reasons behind them and the benefits they’re expected to
bring.
Offer support: Especially if your interventions involve
new ways of working or changes in behaviour, it’s
 essential to offer training sessions, workshops or
resources. These provide individuals with the tools and
understanding they need to adapt and thrive.
Foster open communication: Encourage feedback loops
by encouraging teams and individuals to voice their
experiences, challenges and suggestions regarding the
intervention. This not only fosters a sense of involvement
and ownership but also provides valuable insights that
can help refine the intervention.

Phase 6: Re-measure behaviours and outcomes to

refine your culture transformation efforts
After a set period post-implementation (this could range from a
few months to a year, depending on the intervention and
expected results), it’s critically important to measure again.
Assess both behaviours and outcomes using the same metrics
and methodologies you employed initially. With the new data in
hand, compare the post-intervention results to the baseline
measurements. This will offer a clear picture of the shifts that
have occurred, whether positive or negative.
Are the changes in line with your goals? If so, celebrate the
success and consider how to further amplify or sustain these
results. If the results aren’t as expected, delve deeper to
understand why. Perhaps there are external factors at play, or
maybe certain aspects of the intervention need tweaking.
Based on the insights from the re-measurement, it’s time to
refine your interventions if necessary and make adjustments
based on feedback and the data you’ve collected. Once
refinements are in place, continue the cycle of implementation
and measurement to ensure continuous improvement.
 In essence, the process of implementing interventions and
subsequently measuring their impact is iterative. It’s about
creating a feedback loop where interventions are continuously
improved upon based on real-world results. This cyclical
approach ensures that your strategies are always in alignment
with the ever-evolving needs and dynamics of your
organization.
The next time you’re tempted to simply copy a value from a
high-performing competitor, or when someone insists that
engagement surveys are the true reflection of your
organizational culture, or when you’re about to green-light a
wellbeing initiative without deeper insights, remember this
book! This book has equipped you with the unique lessons that
I have humbly learnt and shared with you. I truly hope you are
now better positioned to cultivate a resilient, purposeful and
thriving organizational culture.
It’s your choice now – make your workplace culture your
organizational hero.
INDEX

Page numbers in italic denote information within a figure or table.

accountability 262, 264, 269–72, 330

achievement 26, 52, 132, 219, 240
acquisitions 150–53
act first culture types 124, 127
actionable insights 15, 35–42, 287–91, 299–301
active learning 26, 105–20, 131, 173, 218
active listening 318
adaptability 150, 279–82
Adaptive Computer Testing 30
adjusted R square 45, 46, 103
agility 164, 279–96
airline industry 8, 24, 154–55
APEASE intervention design 135
see also equity
Apple 245
apply type culture 122
automatic motivation (system 1 thinking) 11, 14, 140, 141
average mean correlation 79, 83, 92, 97

Bandura, Albert 12
banking sector 165, 166–77, 247–59
challenger banks 171
behaviour 18–19, 19–22, 24, 326–27, 328–29
individual 27
organizational 13, 27
behaviour change 13–17, 140–42
behaviour interventions (culture by design) 15–17, 135–48, 260
behaviour mapping 30–34, 58–61
behavioural clusters 42, 121–34, 174
behavioural economics 13
behavioural factors descriptors 25–26, 31, 38–39
behavioural measurement (diagnostics) 26–30, 69–72, 151–53, 156, 157
behavioural nudges 12, 144, 147–48
behavioural research studies 24–26
behavioural roadmaps 183–84, 194–202, 329
behavioural science 10–17
behavioural selection 135, 136–40
beta coefficient 47, 50, 52, 103
between-cluster differences 190
bias 12, 13, 159, 216
big data 156–57, 245
Black employees 226–27, 231
Boeing 737 Max 8
briefings 264

capability (people focus) 14, 26, 35–42, 54–61, 132–34, 136–40, 166–67
change management 186, 195–202
risk management 275, 276, 277
wellbeing 302, 307, 309, 311, 315, 320, 330
challenger banks 171, 247, 254–59
change
behavioural 13–17, 140–42
commitment to 184–88
transformational 180–204
change champions 54, 57, 181–88, 197, 202, 306–07, 328
change management 180–204
change programme success/ failure rate 181, 188, 202
Chernobyl disaster 278
classical conditioning 11
cluster analysis 75, 190–93
behavioural 42, 121–34, 174
cluster maps 33, 199
coefficients 45, 47, 48, 75
beta 50, 52, 103
see also correlation coefficient tables
cognitive behavioural therapy 12
collaboration 251, 263, 280, 281–82, 284–86
collectivity (partnerships) 26, 122, 132, 175, 212–13, 218, 276, 280
collegial relationships 256–57
COM-B model 13–17, 140–42
see also capability (people focus); motivation; opportunity
communication 7, 147, 150, 237, 239, 262–64, 267–68, 320, 331
see also listening skills
competition (competitive edge) 251, 282
compliance 70, 124, 125, 278, 328
compound annual growth rate 245
conformity (non-conformity) 26, 130, 175, 212–13
diversity 222–27, 229–36, 239–40
innovation 249, 259–60
safety culture 266–71, 275
wellbeing 299–301, 302–05, 307, 309, 312, 316
consent 158
consider first culture types 124, 126, 127
consistency (moderation) 26, 76, 130, 248, 252, 280, 281
consolidation (risk avoidance) 26, 52, 59, 116–17, 131, 212–13, 248
consultants 220–21
context 135, 136–40, 189
continuous improvement (radical behaviour) 26, 52, 60, 130, 248, 280, 281, 317
continuous learning 262, 264, 268–69
correlation coefficient tables 77–84, 89
see also multiple R
creativity 282
criterion (dependent) variables 25, 44, 50, 103, 177
cross-functional teams 280, 281–82
culture 216
see also social norms
culture analytics function 157
culture by design 15–17, 135–48, 260
customer-centricity 281–83
culture champions 54, 57, 181–88, 197, 202, 306–07, 328
customer experience 254–56
customer satisfaction 68, 70, 71, 282
customer service 171, 200, 254
cybersecurity 251

data privacy 158

data sampling 159, 176, 276, 283, 287, 299
data security 251, 252
DE&I 215–44
see also diversity; inclusion
decision making 5, 11, 250
see also empowerment (delegation); system 1 thinking (automatic motivation);
system 2 thinking
defaults 147
define the answer culture types 125, 128
delegation (empowerment) 26, 36–40, 42, 52, 130, 137–38, 139–41, 148, 218, 327
 agility 289–91
safety culture 262, 275
wellbeing 302
delivery centric (focus) 26, 132, 186, 212–13
dependent variables 25, 44, 50, 103, 177
descriptive statistics 43–44
df 46
‘dieselgate’ 8
digital transformation 154–59
disenfranchisement 26, 39, 122–30, 133, 139–41, 212–13, 224–27, 285, 293
diversity 150, 159, 215, 217
double negative 170–71, 172, 207, 209, 210, 211, 212

effect size 49, 51, 52, 157, 177, 222, 229, 291, 299
electronics sector 245
‘emissionsgate’ 8
employee silence 164–65
employee voice 163–79
employees 5, 6–9, 328–32
commitment to change 184–88
engagement 5, 8, 64, 219, 221, 283
retention 5, 68, 71–72, 208, 209–13, 219–20
sentiment analysis 157, 163–79, 184
surveys 8, 64, 167–70, 176, 219, 221–22
empowerment (delegation) 26, 36–40, 42, 52, 130, 137–38, 139–41, 148, 218, 327
agility 289–91
safety culture 262, 275
wellbeing 302
endogenous variables 75
engagement 5, 8, 64, 219, 221, 283
ensure safety culture 125, 126, 127, 129
Environmental Protection Agency 8
envision type culture 122, 123, 124–29, 131, 134
equity 135, 217
ethics 158–59, 321
ethnicity 226–37, 240–41
ethnography 155
exam performance analysis 44–45
Excel tracking sheets 142, 144, 146
executive committees 142–44, 145, 237, 239
exogenous variables 75
experimentation 173, 196, 250, 251, 252, 258, 282
explanatory variables 44
exploratory multivariate analysis 75
explore type culture 122, 123–29, 130
expressiveness 26, 36, 130, 173, 186, 197, 302, 306–09, 313
DE&I 218, 223, 224–27, 229–37, 239–40
external (outer) focus 26, 131, 248–49, 275, 280, 281
external perceptions 257–58

F-test 46
factor analysis 23, 25, 73, 74, 75, 176, 228
factors 25
fairness 159, 165, 222
false positives 63, 64
fatigue effect 28
fear 164–65, 278
feedback 30, 32, 147, 254, 256, 280, 281, 282, 284
feedback loops 17, 263, 330, 331, 332
finance sector 166–77, 184–88, 189–94, 219–42
Financial Conduct Authority (FCA) 216, 219–20, 241
financial performance indicators 68–69, 71
first level modelling 49
fit 45, 167
five participant rule 49–50, 158
flexibility 26, 130, 186, 197, 248, 258, 279–80
front office functions 222, 223, 225–26, 231, 236, 237, 241

gas industry 283–96

gender 209, 219–20, 222–26, 230–33, 239–41
geographical context 135, 155, 167
‘get-me-there-it is’ 66
goal setting 182, 183, 218
good practice 21, 183, 221
goodness of fit 45–46
growth 149–53, 205
growth mindset 288–89

health sector 13
heatmaps 35–36, 37–38, 40, 43, 54–55, 62–63
hiring (recruitment) 5, 149–50, 208–13
HR processes 205–14
humility 30, 32–33, 35, 36, 37, 39, 59

IBM 246–47
identification strategy 76
implementation plans 144, 146
impression management 28
inclusion 155, 159, 217–19, 222–27, 230–35, 263, 266, 327
incremental change 280–81
independence tension 130, 132, 134
independent variables 44
individual behaviour 27, 326
individualism 26, 132, 280, 284–85
information exclusivity 250
information technology sector 245
innovation 26, 36, 37, 59–60, 99, 131, 245–49, 282, 302
innovation centres 248, 249
innovation indexes 247, 248–49, 252–55, 258–59
innovation showcases 251
internal (inner) focus 26, 131, 249, 280, 281
Internet of Things 245
intervention design 135, 140–42
intervention plans 135, 142–48, 306–24, 329–31
interviews 248, 249–52
iPhone 245
ipsative diagnostics 27, 28–29, 30
Item Response Theory model 27, 29–30
iteration 280–81

K-means clustering 190

key behaviour identification 328–29
known knowns 63, 216
known unknowns 63, 65–66, 216

lagging indicators 20–22, 67

latent constructs 23, 74, 75
leadership 5, 6–9, 150, 155, 173, 262, 263, 318, 320–21, 328–32
team interventions 143–44
training reviews 321
see also executive committees
leading indicators 20–22, 25, 67–72, 327
learning 117–18
active 26, 105–20, 131, 173, 218
continuous 262, 264, 268–69
see also training
least squares regression models 187–88
life sciences sector 245
Likert scale 27, 157, 287
line graphs 57–59
linear regression 49, 50, 103, 187–88
listening skills 318
logistic regression 49, 187–88

mean 43
see also average mean correlation
measurement error 23, 74
measurement (metrics) 6, 15, 75, 184–88, 329
behavioural 26–30, 69–72, 151–53, 156, 157
financial 68–69, 71
HR 206, 208
outcomes 19
retention 68, 71–72
safety 20–21, 22, 68, 69–70
see also lagging indicators; leading indicators; re-measurement; validation
measurement model 75
median 43–44
mediation analysis 76
mergers 150–53
Microsoft Teams 239
mindset 288–89
Mixed race employees 226–27
model coefficient 103
model fit assessment 46, 76, 103, 187–88
moderation 26, 130, 248, 252, 280, 281
moderation analysis 76
motivation 11, 14, 139, 140, 141, 195, 208, 211, 330–31
MS 46
multi-traits multi-methods analysis 25
multiple R 45, 49, 50, 102–03, 177, 188, 299
multiple regression analysis 25, 44–53, 71, 102–03
multivariate analysis 74–75
multivariate analysis of variance 25

navigate possibility culture 125, 126, 127, 129

negative reinforcement 12
nested behavioural model 49, 52, 187–88
neutrality 26, 130, 186, 224–26, 227, 232–35
non-conformity (conformity) 26, 130, 175, 212–13
diversity 222–27, 229–36, 239–40
innovation 249, 259–60
safety culture 266–71, 275
wellbeing 299–301, 302–05, 307, 309, 312, 316
normative diagnostics 27–28
norms 147, 166
Nuclear Leadership Effectiveness Attributes 263
nuclear power industry 262–78
nudges 12, 144, 147–48

objectivity 159
observations 46, 50
odds ratio 103–20, 228–30, 265–70, 274, 287–91, 295, 300–01
off-the-shelf frameworks 21, 65, 327–28
oil industry 283–96
onboarding 208, 211
one-size-fits-all 183, 191–92
open communication 237, 262–63, 331
operant conditioning 12
operational efficiency 44, 68, 71
operational systems (processes) 151, 165–66, 282
opportunistic modelling 38
opportunity 14, 139–41, 195–96, 280, 330
opt-out mechanisms 147
organizational behaviour 13, 27, 326
organizational change 180–204
organizational culture 4–5, 6, 65–66, 149–53
and outcomes 43–53
organizational development 220, 221, 296
organizational growth 149–53
organizational strategy 15, 328
organizational values 4–5, 7–8, 14–15, 107–08, 128, 150
behaviour mapping 30–34, 58–61
orthogonal rotation 25
outcome (dependent) variables 25, 44, 50, 103, 177
outcomes 18, 19–22, 24, 25, 43–53, 328–29
outer (external) focus 26, 131, 248–49, 275, 280, 281
own type culture 122
ownership 257, 329

parsimonious model 45, 49, 70, 77, 190, 265, 287, 299
partnerships (collectivity) 26, 122, 132, 175, 212–13, 218, 276, 280
passive learning 26, 131
path analysis 22–23, 72–101, 168–69, 177, 271–78, 291–96, 301–06
path coefficients 75
Pavlov, Ivan 11
people focus (capability) 14, 26, 35–42, 54–61, 132–34, 136–40, 166–77
change management 186, 195–202
risk management 275, 276, 277
wellbeing 302, 307, 309, 311, 315, 320, 330
perceptions 257–58
performance 5, 55–59, 68–69, 71, 223
personality questionnaires 27–30, 150
persuasive design 195–96
pharmaceutical sector 245
physical capability 14
physical opportunity 14
‘place’ (workplace environment) 5–6, 133–34, 166–68, 182–83, 188–94, 197–202, 297–
324
positivity 30, 33, 35, 36, 39, 60–61
predictive analytics 25, 44–61
see also multiple regression analysis
predictor variables 44
primary behavioural clusters 121–23
principal component analysis 25, 75
proactive safety management 272–73
process management 151, 165–66, 282
professional services sector 208–13
progressive tension 130, 131, 134
prohibitive cultures 169, 306
promotive cultures 169, 306
psychological capability 14
psychological safety 91–96, 101, 111–12, 237, 263, 265, 278, 294, 329
wellbeing 297, 302–06, 318–19
psychometric testing 27–30
public health 13
purpose 109–10

quadrant distribution method 40, 41, 54–57, 62–63, 64, 136, 171–76, 236, 238, 308–12
qualitative data 157, 249–52, 254–58
quality indicators 68, 70
quantitative behavioural research studies 24–26
quantitative behavioural science 10–11
quantitative data 157, 252–54, 258–59
question and listen culture 125, 126, 128
questioning 266–67

R 45–46, 272, 273

R&D investment 245–61
see also innovation
R square 45–46
see also adjusted R square
racial discrimination 226–37
radar graphs 222, 223–27, 232–35, 252, 253, 255, 259
radical behaviour (continuous improvement) 26, 52, 60, 130, 248, 280, 281, 317
re-measurement 17, 332
recognition 91–96, 101, 108–09
recruitment 5, 149–50, 208–13
reflective motivation (system 2 thinking) 11, 14
regression analysis 104, 229, 230
linear 40, 50, 103, 187–88
logistic 49, 187–88
multiple regression 25, 44–53, 71, 102–03
regressor variables 44
regulation 251, 263
see also Financial Conduct Authority (FCA)
reinforcement theory 12
relationship building 26, 132, 276
collegial 256–57
residual plots (model fit) 46, 76, 103, 187–88
resilience 94, 110–11
resource implementation plans 144, 146
respect 30, 32, 33, 35, 36, 37, 39
retention 5, 68, 71–72, 208, 209–13, 219–20
return on investment 102–20, 245, 274–75, 295
reward systems 319
risk 117–18
risk anticipation 115–16
risk aversion 257, 258
risk avoidance 116–17
risk avoidance (consolidation) 26, 52, 59, 116–17, 131, 212–13, 248
risk management 5, 114–15, 251, 262–78
risk ownership 113–14
risk path analysis 84–91
risk tension 129, 130, 134
roadmaps 183–84, 194–202, 329
role modelling 6, 7
roll-out strategy 331
‘roll up your sleeves’ attitude 256
Romerty, Ginni 246
rules and responsibilities campaigns 322
Rumsfeld, Donald 216

safety culture 125, 126, 127, 129, 262–78

safety indicators 20–21, 22, 68, 69–70
sample size 43, 46, 55, 64, 176, 208, 276, 277, 283, 299
sampling 159, 176, 276, 283, 287, 299
Schein’s culture model 4
second level modelling 49
secondary behavioural clusters 124–29
secondary regression output 46–47
security 251, 252
self focus 70, 132, 133, 186, 224–27, 232–35, 280, 285
semi-structured interviews 248, 249–52
senior leaders 329
sentiment analysis 157, 163–79, 184
sequential ordering 26, 36, 37, 71, 131, 248, 281
short-termism 26, 131, 249, 281
significance F 46, 50, 52, 103
simultaneous analysis 76
Skinner, B F 12
social desirability 28
social learning theory 12
see also role modelling
social norms 147, 166
social opportunity 14
spurious model 45
SS 46
stakeholders 330, 331
see also employees
standard deviation 44
standard error 45–46, 48, 50, 52, 103
standardization 26, 130, 186, 248, 279
statistical validity 49
status 26, 132
strategic alignment 205–14
strategic thinking 26, 131, 249, 281
strategy 15, 328
roll-out 331
strong foundation quadrant 170–72, 173, 207, 209, 210
structural equation modelling 22–23, 73–101, 271–78
structural model 75
suppressed talent quadrant 173, 207, 209, 210
surveys 8, 64, 167–70, 176, 219, 221–22
synchronous working 26, 131, 248, 258, 281
system 1 thinking (automatic motivation) 11, 14, 140, 141
system 2 thinking 11, 14

tactical focus 26, 131, 249, 281

target (dependent) variables 25, 44, 50, 103, 177
team focus 26, 52, 71–72, 132, 186, 197, 275, 280, 319
DE&I 218, 222–27, 229–36, 239, 240
wellbeing 299–301, 302, 307, 310, 314
see also cross-functional teams
tech fests 251
tech stacks 258
technology 154–59, 245, 256, 257, 258
telecommunications sector 298–324
test-retests 25
thematic maps 249–50, 254–58
time and resource implementation plans 144, 145, 146
time management 319
‘tone from the top’ (role modelling) 6, 7
top-down decision making 250
toxic cultures 8
tradition 250
traditional banks 249–54
train industry 154–55
training 165, 215, 318, 319, 321, 331
training analysis 105–20
training reviews 321
transaction focus 26, 132
transformation consultants 220–21
transformational change 180–204
transparency 257, 262, 284, 287–88
trust 7, 158, 167, 168–75

unconscious bias 216

univariate analysis 74
universal truths 21–22
unknown unknowns 216

validation 135–36, 148

see also feedback loops
validity 25, 49
values 4–5, 7–8, 14–15, 107–08, 128, 150
variables 25, 44, 50, 75, 103, 177
varimax rotation 25
verify type culture 122, 130
virtuous alignment quadrant 170–73, 207, 209, 210
Volkswagen emissions scandal 8

WebMD Health Services 215–16, 242

wellbeing 167, 168–70, 171–74, 297–324
Wells Fargo 8
White employees 226–27
within-cluster differences 190
workplace culture 154–55
workplace environment 5–6, 133–34, 166–68, 182–83, 188–94, 197–202, 297–324
workshops 189, 215, 313–16
World Association of Nuclear Operators 263, 276–78
Figure details
 Publisher’s note
Every possible effort has been made to ensure that the information contained in this
book is accurate at the time of going to press, and the publishers and authors cannot
accept responsibility for any errors or omissions, however caused. No responsibility for
loss or damage occasioned to any person acting, or refraining from action, as a result of
the material in this publication can be accepted by the editor, the publisher or the
author.

First published in Great Britain and the United States in 2025 by Kogan Page Limited

Apart from any fair dealing for the purposes of research or private study, or criticism or
review, as permitted under the Copyright, Designs and Patents Act 1988, this publication may
only be reproduced, stored or transmitted, in any form or by any means, with the prior
permission in writing of the publishers, or in the case of reprographic reproduction in
accordance with the terms and licences issued by the CLA. Enquiries concerning reproduction
outside these terms should be sent to the publishers at the undermentioned addresses:

2nd Floor, 45 Gee Street

London
EC1V 3RS
United Kingdom
www.koganpage.com

8 W 38th Street, Suite 902

New York, NY 10018
USA

Kogan Page books are printed on paper from sustainable forests.

© Hani Nabeel 2025

The right of Hani Nabeel to be identified as the author of this work has been asserted by her in
accordance with the Copyright, Designs and Patents Act 1988.
All trademarks, service marks, and company names are the property of their respective
owners.
ISBNs

Hardback 978 1 3986 1747 6

Paperback 978 1 3986 1744 5
Ebook 978 1 3986 1748 3
British Library Cataloguing-in-Publication Data

A CIP record for this book is available from the British Library.

Library of Congress Control Number

2024035275