Chapter 3

Digital leadership
Strategies for AI adoption

STRATEGIZING FOR BUSINESS OPTIMIZATION

Digital business is based on data as a strategic asset. Strategic planning

for business optimization (BO) incorporates data, analytics, and applica-
tion. Providing customer value is the purpose of utilizing data. Artificial
intelligence (AI) impacts organizational culture, business process, systems,
and data. Eventually, AI adoption impacts the customer. Digital business
strategy for BO depends on the context. A market, product, process, and
government policy are examples of the context. Furthermore, automation,
optimization, and humanization play an important role in digital business
strategies. Data usage also leads to new and opportune business models.
Data-driven analytics in business processes enable data-driven decisions,
innovation in products, and efficiency in business functions. BO expands
the use of data beyond IT and into the rest of the organization. BO improves
business performance by enhanced decision-making. Decisions around
operational optimization are based on data-driven analytics. Such digital
strategies lead to the discovery of new means of delivering customer value.
Adopting AI-based decision-making involves both technical and business
risks. Leaders understand and incorporate those risks within digital business
strategies. Costs, returns, risks, security, privacy, and safety associated with
AI-based BO are outlined within the business strategy. Development and
deployment of AI solutions, reengineering of processes, and change manage-
ment form part of business strategy. Digital business strategy also ensures
the security of business data from a regulatory and compliance viewpoint.
Business strategies for optimization treat AI adoption as a holistic, organi-
zational challenge rather than a project-based initiative. AI for business opti-
mization is also not considered a technology-based initiative. Instead, digital
strategies are developed to take a long-term and holistic view of AI adop-
tion in the organization. Apart from collaborating in the digital world, busi-
nesses are continuously merging and acquiring each other. These mergers
and acquisitions lead to the challenge of data and technology silos. Planning
the optimization of a business includes planning for integration and use of
data across multiple organizational boundaries.

59
60 Artificial Intelligence for Business Optimization

Envisioning digital business strategy for AI

Digital business strategy encompasses visions, goals, capabilities, and proj-
ects. The strategies are aimed at developing business capabilities. While
projects play a role in developing the capabilities, projects themselves are
not the mainstay of the BO effort. Digital implementations are accom-
plished through projects but their purpose is to develop capabilities for the
business.
A digital business strategy for AI incorporates the following elements:

• Vision and aspirations of the business and understanding how AI

influences them. These business visions provide the starting point for
the BO effort. Providing customer value is the basis of this vision for
most digital organizations.
• Defining goals that enable business outcomes based on AI capabilities.
The vision provides the guiding principle behind the development of goals.
• Identifying and defining the capabilities of the business – these
include the current and future (required) capabilities for data-driven
decision-making in the business.
• Planning the implementation of capabilities in business processes
through the projects.
• Developing a risk management plan that considered the risks associ-
ated with business optimization, and the risks in the business that can
be handled with the help of AI-based technologies.
• Outlining the relevant metrics and corresponding measurements that
will help in justifying the BO initiative as well as in reporting on its
progress.
• Planning for cybersecurity and privacy considerations to be incorpo-
rated in BO.
• Due consideration to the people’s concerns of job losses due to auto-
mation and optimization.
• Optimization has the potential of customer perception of less person-
alized service which needs to be considered strategically. Optimization
with AI implies change which is a risk. Strategies balance new process
implementation with current operations.
• Seeking regulatory approvals in terms of use of data and processes.

The vision and goals of the organization provide the basis for desired out-
comes in BO. Prioritized outcomes influence the digital business strategy
and refine vision and goals. This iterative approach to vision and priori-
tization continues to refine the strategy. Strategies stabilize after a couple
of iterations. Strategies reduce changes in direction. Processes are defined,
measured, and implemented as part of BO. The implementation of digital
strategies is then assigned to business units. The business strategy generates
actions across the organization.
 Digital leadership 61

Digital strategies are holistic

A digital business strategy provides the necessary guidance and control in
optimizing business processes. Such digital business strategy includes mul-
tiple disciplines and functions of the business. A good data-driven strategy
works to integrate data with processes and people with due consideration to
governance, security, storage, and usage. Digital strategies make provision
for the inevitable collaboration and merger of organization that require the
unification of data and modeling of collaborative business processes.
Data unification across multiple processes and businesses enables the cre-
ation of a singular customer view, single point of contact of customer with
the organization, enhanced quality and governance, and greater collabora-
tion between technical and business stakeholders. Technical and business
silos are melted by the holistic, long-term view of the strategies, resulting in
a unified view of the customer. Integrated data also leads to efficient analyt-
ics and accurate insights as it eliminates the need to process the same data
multiple times and focuses on a singular view of complete and accurate data.
The integration of data is thus invaluable in data analytics and BO.
The vision and mission of the organization provide the necessary direc-
tion and impetus in developing BO strategies. The vision, goals, objectives,
and corresponding requirements specific to AI are brought together with BO
strategy and planning. AI, data, and infrastructure of the organization play
a supportive role in the achievement of the goals. Data-based AI technolo-
gies present myriad challenges during optimization, including and especially
security and privacy of data and users. Digital strategies prioritize security
requirements in terms of risks, issues, and concerns across the cloud, Big
Data, and IoT implementations. Evaluation and assessment of the existing
business strategy is a helpful input in the development of the new digital
strategy. Digital business strategy considers an optimal approach that may
work across the enterprise – with a caveat that it may or will change during
the course of transformation – together with a plan to handle the risks.
Business leaders who develop the digital strategies pay close attention to
the culture and values of an organization. The cultural aspect of an organi-
zation is its mindset within think data discussed in the previous chapter. The
mindset is the most challenging aspect of change within BO. Digital strate-
gies give significant importance to the “soft” issues of culture and values.
Leaders take a strategic rather than tactical view of optimization beyond
automation. Tactical use of data is primarily in automation. Strategic use of
data includes removal of redundant activities in a process, creation of new
collaborative processes, and ensuring business agility.

Customer value is the goal

Providing customer value is the key goal of BO. A consistent and succinct
roadmap for BO focuses on the customer experience. Enhanced customer
62 Artificial Intelligence for Business Optimization

experience includes engaging the customers, offering a wide range of busi-

ness services, and providing better quality and timely responses.
BO also changes the organizational structure and behavior as processes
are reengineered and decision-making is decentralized. These changes
impact the competitive position of the organization. Metrics are essential
to monitor the implementation of the digital strategy. For example, sat-
isfaction, loyalty, and cost metrics provide implementation control and
direction. The business architecture keeps customer, infrastructure, and
resources in continuous alignment. Strategies are continuously aligning the
capabilities to achieve business outcomes.1

Addressing the business goal or problem

A business problem is a good starting point for developing BO strate-
gies. Cumbersome or unsecured business processes may result in lost
customers. Consider, for example, a business problem of rise in customer
complaints for billing errors. These operational problems can have an
immediate, short-term fix. However, BO examines the entire business and
re-engineers it not only for the immediate problem but a suite of known
and unknown problems. An upgrade of the software system with embedded
analytics or establishing a call center equipped with AI can be considered a
strategic-technical solution. Prototyping the solution and enabling key busi-
ness stakeholders to provide input make it relevant to the business.
Business problems are investigated and analyzed strategically in BO.
Agility plays a positive role in enabling this investigation in a trustful, hon-
est, and collaborative manner. 2 Investigating business problems provides
strategic options in using AI. A composite Agile approach that balances
formal planning with agility provides an excellent opportunity for develop-
ing and deploying AI in business.

Business agility in decision-making

Digital business strategies aim to use data-driven decision-making across
every business function. Developing strategies to optimize operational
areas of business is not a one-off deliverable. Instead, developing BO strate-
gies is also an iterative and incremental approach that benefits with the use
of agile principles and practices.
Agile as a business characteristic carries immense value.3 The agility goals
of business are included in the strategic planning effort. The balance between
customer focus and operational continuity is achieved only with business
agility. In turn, this strategy depends on an Agile approach to knowledge
management and project management.
The moment agility in processes is introduced, it brings in elements of itera-
tion and increment that aim to handle the uncertainty and change in the pro-
cess.4 Even though ML algorithms are now able to mimic human behavior to
 Digital leadership 63

a large extent, the behavior is itself a function of widely differing values for
the same input. An optimized business makes faster decisions that are more
accurate than before, and that are made over a larger number of touch-points.
Therefore, the more the iterations (agility), the better are the chances of antici-
pating and ameliorating disruptions.
Following are agile-specific considerations in developing a strategy for BO:

• Identifying the opportunities for products and services innovation in

terms of their processes, creating their prototype, and experimenting
with the prototype in order to provide input in the strategies
• Identifying, categorizing, and prioritizing the risks in optimizing busi-
ness processes with AI and incorporating those risks in the strategy
• Using Agile iterations to understanding and scope the overall require-
ments of the BO initiative and directing investments in AI accordingly
• Modeling the parameters around providing customer value and using
those parameters as input for the development of strategy
• Digitizing the external (marketing and sales) functions of an organi-
zation and planning to incorporate them in social media in an itera-
tive and incremental manner
• Aiming to optimize the internal (operational) business processes with
analytics and AI in an iterative and incremental manner
• Ensuring governance, risk, and compliance of all business functions
digital reporting of compliance requirements using the principles of
Composite Agile Method and Strategy (CAMS)5
• Applying the principles and practices of cybersecurity to the digitized
business processes in a holistic manner
• Keeping the metrics and measurements to the bare minimum at the
start of the initiative and slowly enhancing and enriching the mea-
sures with the initial iterations of implementing AI

STRATEGIC PLANNING FOR BO

Strategic planning for digital business is the creation of a plan for the digi-
tal business to achieve its vision and goal. Such plan also focuses on the
alignment of business functions with all service and support functions,
especially IT. The strategic plans, goals, objectives, and expectations of the
digital business are periodically synchronized with capabilities and project
initiatives of the organization. Strategic planning identifies goals, builds a
roadmap, and develops a blueprint for implementing change. Modifications
and upgrades to IT capabilities result from strategic planning which aids in
the transition to a digital business.
Strategic planning is also associated with qualitative and quantitative
measures. The metrics and measurements enable buy-in from the business
64 Artificial Intelligence for Business Optimization

for the digital transformation. The interactive, innovative, and collabora-

tive coalition of IT and business has a significant implication towards deliv-
ering the right business results.
Strategic planning for BO considers the following impact of the changes
in the decision-making process:

• Current Decision-Making Process – this is important to understand

as it will have gaps in delivering value to the customer. Any changes
to the decision-making process are based on the current process and
the risks associated with changing it based on data-driven decisions.
• Impact of Data on Decision-Making – including the sources, stor-
age, and manipulation of data. Data sources from collaborative busi-
ness partners, third-party and government sources, and the in-house
generated data stored on the Cloud impact decision-making and are
included in the strategic planning process.
• Impact of Analytics on Decision-Making – especially as this impact will
change (improve) the accuracy and time taking for making decisions.
• Impact of People and Biases on Decision-Making – strategies need to
factor in the changes that digital business brings about on people and
the way they make decisions. Changes associated with human bias
and sensitivity require careful strategic planning as these, perhaps, are
the most important and difficult parts to change in a DT.
• Business Process Reengineering – when undertaken while the orga-
nization is still in operation requires substantial and holistic strategic
planning. This is so because the impact of digitizing a process on
another process and functions of the organization can be substantial.
• Collaborative Impact – especially because DT will render the business
much more collaborative on the electronic platforms than ever before.
Strategic planning needs to include due considerations to the services
offered and consumed from collaborative business partners, policies
governing those services, and the impact on the value being delivered
to the customer.
• Cybersecurity – is a vital strategic consideration in DT as changes on
the electronic platform open up the organization to the possibility
of cybercrime and related risks. Data, analytics, and digital business
assets need additional protection and those protections need to be
factored in, in strategic planning.

“ Think data” in strategies

Strategies ensure AI plays an important yet supportive role in business
decision-making. AI is an enabler and an agent for change. Data is a busi-
ness asset that provides value. Think data is a holistic approach to discovery
of that value. Extracting knowledge from data requires a strategic approach.
 Digital leadership 65

Business optimization is the application of think data in effective utiliza-

tion of limited resources and capabilities. Data thinking is the extraction of
useful knowledge and business value from the data. A strategic approach
brings close collaboration between business stakeholders and data scientists.
Business architectures6 help business organizations achieve effective uti-
lization of resources, as well as generating customer value. Data and archi-
tectures based on 4+1 V (Volume, Velocity, Variety, Veracity + Value) of Big
Data help align AI solutions with BA based on Big Data strategies for Agile
business.7 Big Data and AI are strategically important in optimizing busi-
ness processes.
Machine learning (ML) systems learn from previous decisions and pro-
vide insights into those decisions for future decisions. Big Data provides
vast repositories of data (decisions and the data on which those decisions
are based) in order to improve the speed, accuracy, and reach.
Collaborations between businesses need to be strategically thought through.
Organizations are brought together by digital platforms. Viewing the entire
organization as a whole is essential for developing business strategies.
Effective use of data creates a fine granular decision-making engine for
multiple decision points. A well-constructed and well-maintained strategy
leads to a lean and agile, digital business. Lean processes enable a business
to update its strategies to correspond to the changing business environment.

Strategic AI considerations
Strategies describe the actions needed to achieve business goals. AI is a part
of the enablers to provide customer value. Risks associated with the imple-
mentation of BO are described and managed through the strategic plan.
Projects are aligned to the desired business outcomes.8
Strategies based on the business outcomes make the businesses customer-
centric. A business strategy to improve customer interaction by optimizing
supply chains used AI to optimize the entire process. An online business
fulfills fast-moving consumer goods through AI. Products are offered based
on optimized supply chains. Strategies also protect innovations from being
mimicked by the competition. Sound business strategies incorporate the AI
technologies in business processes, pay attention to customer needs, and
provide continually balancing implementation of AI.
Composite Agile Method and Strategy (CAMS)­9 professes a comprehen-
sive and strategic approach to developing organization-wide agile capabili-
ties that include people–process–technology–money aspects of the entire
business. This strategic approach is also applicable to AI adoption. While AI
adoption across an entire organization is more complex than implementing a
single AI solution, the rewards are correspondingly greater. Organizational
complexity (e.g., incumbency, communication, dynamicity, sustaining busi-
ness, social, psychological, and cultural) combined with the complexity of AI
66 Artificial Intelligence for Business Optimization

People Process
• Activities & Tasks in a
• Acceptance by leaders Process
• Regular Communication • Estimation & Metrics of a
with multiple channels; Process
• Risks (privacy) concerns • Method Friction in
• Training & Coaching – Organization
systematically • Usability / User
Experience Analysis for
Customer

Technology Money
• Data – Usage • New Investment
• Tools for Analytics, Strategies
• Networks & • Returns on Investment
Communications (ROI) on AI
• Devices (IoT) • Business Transformation
• Data Quality; Security Costs

­Figure 3.1 Strategic AI considerations.

technologies provides challenges to their adoption. Despite these complexities

and associated challenges, AI enhances the ability of the business to respond
to changing external and internal circumstances.
Customers, suppliers, and strategic partners are all able to derive benefit
from an organization-wide use of AI.
Figure 3.1 summarizes the four aspects of developing a business strategy
for data-driven BO. These are as follows:

People
This is the social dimension of BO. Sociocultural factors influencing the
business are given high importance. Leadership focuses attention on how
BO will affect clients, employees, and other users. For example, changes
to work formats (e.g., telecommuting, telemarketing) and their resultant
impact on the organizational and social structures are all part of this social
dimension. Due consideration needs to be provided to areas of individ-
ual and team strengths and the corresponding weaknesses. For example,
customer-facing individuals can change the perception of the organization
with basic training and a positive attitude. Senior managers and leaders of
the organization, working as individuals, can also have a substantial effect
in changing to an agile business.
These organizational changes, however, cannot be sudden when people are
involved. Training, motivation, and individual aspirations need to be consid-
ered, and both performance and functionality need to be kept in balance.
 Digital leadership 67

Broader social issues such as the effects of promoting agility, balancing risks
with advantages, and ethical and legal business practices (including relevant
documentation) are also part of the social dimension. Usability/user experi-
ence analysis is also a people issue because of its subjective nature.
Strategic planning for BO starts with a list of key stakeholders (e.g., custom-
ers, staff) and the RACI matrix.10 The level of acceptance by the stakeholders
of new ways of doing business and the mechanisms for communicating with
them on a regular and consistent basis are crucial to the strategic plan; risks
associated with privacy of data and the perception by users is people issue.
Training and coaching of users and staff is an important people consideration
in BO plans.

Process
This process dimension deals with how the business conducts its transac-
tions both internally and externally. Business process is an integral part
of this dimension. AI is considered strategically to change the business
processes of the organization to L ean-Agile. This change to the processes
impacts the way the business interacts with customers, the way in which it
manages its employees, and the way it sets up and conducts collaborations
with other business partners. Process consideration in BO enhances BO
quality and value provided to the interacting parties without sacrificing the
current offerings of the business.
Strategic planning in BO includes which processes to optimize and the
extent to which they have to be optimized. For example, in an insurance
process, the strategic approach of business leadership can limit the AI-based
analytics to only predicting the possibility of a claim without going into
the actual decision on whether to accept the insurance proposal or not.
“Certain AI tools are likely to transform the boundaries of your business.
Prediction machines will change how businesses think about everything,
from their capital equipment to their data and people.”11
Modeling and examination of the activities and tasks of the business pro-
cesses is the starting point for BO. Embedding analytics in the processes is
undertaken only after they are modeled. Estimates of activities and tasks
based on metrics indicate how well they are optimized. For example, redun-
dant activities and tasks are eliminated. Due consideration is required in the
use of multiple methods within BO because these methods create friction.
BO strategies give due consideration to method friction.

Technology
The technical considerations include AI technologies, tools, networks,
and devices. AI technologies are the key enablers of business optimiza-
tion. Technologies include computing machines and IoT devices, software,
68 Artificial Intelligence for Business Optimization

algorithms, network, security, and systems. The network infrastructure

also handles the security protocols. Data-driven decisions in organization’s
system are made possible through the AI technologies. Internet-based com-
munications protocols, IoT devices, Semantic Web, and mobile and cloud
computing are all brought together by enterprise and business architectures.
Data sources, storages, and analytics are supported by the Cloud. Security,
privacy, and quality of data are incorporated in the strategic plans for BO.

Money
The money (or economic) aspect of AI adoption deals with costs, prof-
its, and investments. This money aspect of the strategic plan outlines the
costs, budgets, and ROI. Competitor risks, the financial impact, the lack of
analytics, and potential customer loss are factored into the strategic plan.
Costs and benefits of AI adoption are discussed keeping the ROI in mind.
Managing the investments, its customer relationships, and its partners all
have financial impact on the AI initiative. The success criteria for AI include
enhanced customer experience which may not be easy to measure. Strategic
plans include costs associated with change management.

STRATEGIC PLANNING FOR BO

­
­

AI is a disruptive technology that demands new capabilities. Big Data

along with its volume, variety, and velocity needs data analytics as a capabil-
ity. Capabilities also provide the means ingest, store, and analyze IoT data.
Capability allows business optimization that enhances decision-making
during operations. Analytics improve estimates for revenues and associated
costs and, thereby, provide internal business value. BO includes resource
optimization, financial optimization, critical asset optimization, informa-
tion optimization, and knowledge optimization. Investment in people skills,
processes, and technologies is required to undertake the strategies to opera-
tions use of AI.
 Digital leadership 69

Vision & Mission

Strategies
Application in Business

Consequence; Metrics;
Capabilities

& the Feedback Loop

Building the Capabilities

Building AI Projects
Algorithms

Deploy across Organization; Tactics

Train Users;

Using AI in Practice Operations

Customer Value

­Figure 3.2 From strategies to operations.

Digital business capabilities address the key business outcomes. Business

capabilities are directly connected to the values, objectives, and goals of the
business.
The capabilities of the business are continuously augmented through a
rapid feedback loop that is based on the results generated by the initial ana-
lytics, corresponding decisions, and their consequences (Figure 3.2).
While data and its corresponding analytics are getting accepted in busi-
ness decision-making, what is still not fully exploited is storing and ana-
lyzing the previous decisions. With self-learning systems, businesses are
benefitted not only by the use of data but also by previously made decisions.
Decisions based on a large number of previous decisions can be further cat-
egorized and used in even faster, more accurate, and automated decisions.
Self-learning systems also need to be self-correcting. This self-correcting
nature of our systems requires a feedback loop that is also accurate and
quick. Feedback for corrections needs to be provided well within the time
required for the next decision.
Capabilities are developed through projects. These AI implementation
projects smooth the information flow and optimize decision-making. The
people resources are further supported through training and coaching. BO
tactical implementation thrives on flexibility in the digital strategy, align-
ment of capabilities with strategies, and agility in projects.
Digital capabilities also enhance management practices, resourcing,
intelligence, and processes to extract value from data. Multiple dimensions
of the organization are brought together via digital strategies in order to
70 Artificial Intelligence for Business Optimization

­Table 3.1 BO strategy characteristics to optimize business processes

Effective Improve accessibility, timeliness, agility, and effectiveness of processes
used by staff and customers to help achieve their goals.
Efficient Prioritize business functions and AI solutions to maximize process
value. Efficiency is enabled through the personalization of business
processes through AI. Efficiency may complement effectiveness but it
is not always possible to do so. This is because effectiveness deals
with the achievement of results, whereas efficiency deals with the
most optimum way to those results. Strategies aim to harmonize and
personalize functions to enhance efficiency.
Relevant Business processes should meet the user requirements and business
goals. AI can help automate and optimize processes but AI is not able
to understand the relevance of the process to the user. This is the
contextualization of the processes to the needs of the customer.
­Customer-driven Customer values drive the AI-based optimization of processes.
Priorities of AI are based on value to the customer.
Cyber secure The business goals and AI solutions should be secured. This security
includes perceptions by the customers.
Sustainable Sustainability of business process-based analytics.
Agile AI helps to respond promptly to customers and partners. The
processes and systems to be sufficiently agile to enable change.
Practical AI solutions should be realistic and achievable. AI is carefully
embedded in organizations’ systems and information.

enable quality response to the organization’s core functions. Capabilities

align data activities with the goals of providing customer value.
The operational model is transformed from an “inside out” to an “outside
in” model. This means that the organization changes its business processes
to make sure that the operational model aligns itself to meeting customer
needs – rather than creating or offering business values in such a way that
the target customers have to change their requirements to receive what the
business offers.
Table 3.1 lists the principles to achieve strategies that enable alignment of
outcomes and business capabilities for BO.

ML types in BO strategies
ML in AI is performed at various levels of abstraction (Table 3.2). The
rules, patterns, heuristics, and meta-heuristics of ML are summarized in
Table 3.2. At the start, domain experts define the rules for the learning
algorithm to work on the target data. Definitive rules like “Customers
buying product X also buy product Y, Z% of the times,” for example, led
to data-mined discoveries like “young fathers buying beer cans also buy
diapers.”
 Digital leadership 71

­Table 3.2 Machine learning types corresponding to abstraction and task complexity
levels
Abstraction level Task complexity level ML type
Rules Find data matching the rules Data mining
Pattern recognition Find consistent patterns in a Supervised, unsupervised ML
homogeneous dataset
Heuristics Find related patterns in a Transfer learning
heterogeneous dataset
Meta-heuristics Learn how to learn Self-learning
­

At the basic level of abstraction, the machine is instructed where to

begin its learning, and what kind of patterns to look for. In other words,
it is provided with the “what and how” of learning. At the next second
level of abstraction, the machine is given a large quantity of data, and
through a specific algorithm instructed to look for patterns in the data.
The machine is given the how of learning and it finds the what (contents)
in the form of patterns. Supervised and unsupervised learning typically
work along the algorithmic “hows.” The third level of abstraction is
applied in transfer learning, where the machine is told what to look for,
but has to figure out for itself how to look for what it is looking. In other
words, a machine learns the heuristics, just as experts learn the heuristics
in their field of expertise after amassing a great depth of expert knowl-
edge. This is the deepest (or the highest) level of ML that is currently
available. Arguably, these three levels of abstraction in learning are not
adequate to respond to all business needs. A future level of abstraction
where machines will learn how to learn and what to learn is needed.
They will be provided neither with the what nor the how of learning.
Rather, based on the lower levels of abstractions, they will arrive at the
level of meta-heuristics, where they will engage in learning the very art
of learning, given a context.

LEADERSHIP IN BUSINESS OPTIMIZATION

Leaders hold a strategic outlook that goes beyond the immediate opera-
tional and tactical needs of the organization. Leaders envision the future
and contemplate its challenges for the business. Therefore, leaders in BO
do not limit the vision to AI technologies but go beyond into the customer
value, user experience, user perception, and related subjective spaces.
Anticipating the future includes the creation of multiple “what if” sce-
narios to handle the business response as the reality changes. Good leaders
72 Artificial Intelligence for Business Optimization

carry the many “what if” scenarios in their heads all the time. “Technology,
however, is dramatically rewriting the rules of business and life.”12 For
example, ML algorithms self-propagate themselves – requiring a deeper
understanding of the need to “explain” the insights generated. The absence
of “explainability” of AI can create sociocultural challenges that are way
beyond technologies. Another example is the decision on the level of granu-
larity of analytics. The extent to which fine-granular analytics are required
is also a strategic question based on the need of the business.
Given the explosion of “data,” AI tools and technologies are required to
create multiple futuristic scenarios. Data, combined with experience, is the
fuel for the decision-making engine of leadership. Leaders also look beyond
the traditional sources of data and delve into using alternative data. Leaders
also ensure that automation with AI is complimented with optimization
and humanization of business processes.

Automation strategies
Automation repeats the same business function and process – albeit using
technologies. As a result, automation makes the same process faster. Since
the process is coded and tested, the chances of error in an automated pro-
cess are greatly reduced. Automation strategies may not incorporate the
creativity and human ingenuity. Therefore, Natural Intelligence (NI) is not
a part of automation strategies.
Automation strategies are important in the business optimization because
of the following reasons:

• Mundane tasks that people will not enjoy can be offloaded to machines.
Robotics is an example of automation wherein A I-based machines con-
trol operational processes.
• Routine and well-defined tasks are performed much better (faster and
with greater accuracy) by machines. For example, inventory manage-
ment and SCM.
• Machines can automate large processes in parallel thereby further
reducing the time required to perform the tasks.
• Automation provides ability to provide detailed data-driven audit
trails that can be used for analytics including forensics.
• Machines can measure, report, and improve their performance.
• Artificial intelligence fast tracks actions as time is not lost in deter-
mining the cause – it is already known.
• Automation also supports business continuity.
• Machines learn from each episode and thereby improve their
predictability.
• AI systems learn from previous disruptions, predict the use of alternate
resources, and recommend automated response and recovery options.
 Digital leadership 73

Optimization strategies
AI is strategically used to optimize business processes. This optimization
starts within the BO strategy by assessment and prioritization of existing
processes. Strategies enable AI systems to be as self-organized and opti-
mized as possible. AI takes BO to the next level. Optimized processes are
able to understand the urgency of a customer need and the severity of an
issue, and identify approach to the solutions. The A I-based optimization of
business functions makes use of people, processes, money, and technology.
Following are some key aspects of optimization strategies:

• Business are made up of multiple attributes. Each attribute has multiple

perspectives depending on the context. These attributes, or parameters,
are identified, documented, and customized based on the business goals.
• Optimization evaluates the organization’s technical capabilities.
Assessing current performance levels (including gaps, risks, inefficien-
cies, and opportunities) against required performance provides oppor-
tunities and direction for the application of AI to business processes.
• Optimization aligns strategic intents and outcomes with capability-
building initiatives in the business. Continuous alignment of strategic
intents with capabilities and projects is essential for BO.
• Optimization uses capability management to provide resources and
their allocation to meet various competitive dimension.13
• Optimization of business needs governance of organizational
resources to enable focus on the desired business outcome.
• Optimization strategies focus on enabling the users to achieve their
objectives in an effective and efficient manner. Time and accuracy are
treated as enabling factors in optimization strategy.
• Optimization strategies require a change of the mindset of the organi-
zation to a customer-focused one.
• Optimization strategies develop models for data assets and corre-
sponding processes keeping the end-user in mind.
• Optimized processes implement policies and rules electronically to
enable efficiency.
• Cybersecurity ensures the safety and integrity of customer data.
• Optimization strategies assimilate disparate and disjointed processes
in order to collaborate and eliminate redundancies.
• Optimization models existing processes and policy frameworks to define
gaps. Processes help understand the impact of adoption of AI better.
• Optimization blends the physical, electronic, mobile, and collabora-
tive processes to provide a unified view of the business to the customer.
• Creating a unified view for the customer (through MDM) also enables
a single point of contact for the customer.
• Optimization aims to provide insights for decision-making at the
point of action.
74 Artificial Intelligence for Business Optimization

Humanization strategies
Humanization importance is understood by investigating the nature of AI
models – which, by its very nature, is performance driven. The users and
the business leaders aim to maximize the performance. ML code embedded
in AI models is agnostic to the specific situation of the user. AI systems are
based on data correlation, whereas human systems are based on cognitive
knowledge and understanding causation.
AI systems are not cognizant of human values on their own. As a result,
business decisions can result in a loss of value. Due consideration to this
loss of value is a crucial element of “humanization.” Providing customer
value is a subjective phenomenon. AI cannot understand or produce value
on its own. The humanization of AI-based decision-making is a strategic
imperative. AI-based predictions and user experience is the right judicious
combination needed for decision-making.
The application of AI together with NI for prediction, handling of the
crises, and recovery of business is part of the humanization aspect of the
BO strategy. Humanization ensures that AI-based decision-making does
not replace people entirely. Instead, the human factor is brought in the auto-
mation and optimization of business processes in a balanced manner.
Intellectual property, people skills, processes, and finances are crucial
organizational resources that are put together in humanizing strategy.
Customer engagement and operational management work together. For
example, for a given customer engagement, if the tasking and coordina-
tion systems are not integrated with human resource management systems,
resource allocation is manual. Such resource allocation has duplication and
results in multiple handling of the customer queries.
Sourcing people (recruiting) and maintaining skill levels is part of human
capability building. Empowering team members to take quick and innova-
tive decisions and provide high-quality services requires a change of mind-
set. This change of mindset also requires a corresponding change in the
organizational structure that will facilitate and nurture the decision-making
capabilities of staff at the customer-facing level of the organization.
Mindset changes in style, structure, culture, and skills of people. A flexible
and agile organization structure reflects changing context in which the busi-
ness operates. The need to empower staff members and encourage them to
use AI-based decision-making is a humanization strategy. Upskilling people
is made up of sourcing new people and training existing people that result in
a workable mix of business, technology, and functional capabilities.

Users and culture changes

The AI-based culture shift of the organization involves a number of struc-
tural, social, and individual changes. Users need the opportunity to try
new approaches and learn lessons. Developing soft skill contributions of the
 Digital leadership 75

team members and managing team spirit are agile leadership characteris-
tics. Agile also relieves stress, facilitates communication, and enables team
spirit. The hierarchy reporting structure changes to a flat one. Individual
personalities and their biases come into play in humanization.
Visibility, collaboration, discussions, and objective reality of the environ-
ment in which the business operates is beneficial in developing a digital
business strategy. Cross-functional teams at the business, corporate, and
departmental levels work to iteratively define and manage these strategies
without the defined timeframe on an ongoing basis. Each strategy outcome
is mapped to the business capabilities and the desired business outcomes.

BUSINESS OPTIMIZATION INITIATIVES

Business optimization is based on optimization of business processes. AI/

ML helps in identifying redundancies, gaps, and lags in processes. The
starting point for a strategic approach to BO is the desired business out-
come. Understanding the desired outcomes provides the basis for the BO
initiative. AI-enabled BO is based on data. Therefore, understanding of
the outcome is followed by a detailed understanding of the data, including
its sources, quality, and costs. Approach to analytics and the mechanisms
to embed the results in business processes are outlined next. Eventually,
the analytics are presented – mainly through visualizations. Visualizations
include a wide range of techniques, including bar and pie charts, bubble
charts and heat maps, and so on. They show trends and patterns in data.
Figure 3.3 shows the organization factors impacting the BO initiatives
and their role in developing the BO strategies. Business context provides
the business objectives and key business requirements for the AI and Big
Data initiative. The two subfactors arising from the business context are the
financial impact and ROI of the entire initiative and the GRC. The financial
impact and ROI of the BO initiative leads to an understanding of how the AI
capabilities can be applied to understand the markets. The opportunity costs
associated with the initiative and the possible loss of opportunities without
AI are established. Understanding the technical and business capabilities of
cybersecurity and disruptions and recovering from them are outlined next.
Data, analytics, networks, devices, processes, and people are each explored
in terms of how they contribute to the BO initiative. Each aforementioned
element has a strength and a risk – which are explored and documented in
the context of the particular business.
Strategic approach to BO initiatives covers the entire data lifecycle – start-
ing with data input, then understanding the context of that data, establish-
ing the controls for the inputs and their variations, deciding on the level of
granularity, and creating a feedback loop based on the results. This data
lifecycle stands to benefit by an agile approach to developing analytics. For