European Management Journal Vol. 16, No. 2, pp.

190–203, 1998
Pergamon  1998 Elsevier Science Ltd. All rights reserved
Printed in Great Britain
PII: S0263-2373(97)00087-X 0263-2373/98 $19.00 + 0.00

Implementing Corporate
Strategy:
From Tableaux de Bord to
Balanced Scorecards
MARC EPSTEIN, INSEAD, France
JEAN-FRANÇOIS MANZONI, INSEAD, France

Defining a clear strategy can be a difficult process Academy of Management) recently reprinted Kerr’s
for a large company. Translating this strategy into article, they conducted an informal poll of its Execu-
action can be even more difficult. This paper tive Advisory Panel, a group of over 50 top execu-
presents and compares two related tools that can tives representing companies from all over the world.
help companies articulate and cascade their strat- Ninety percent of the respondents believed that
egy: the Tableau de Bord, a concept popularized in Kerr’s folly is still prevalent in Corporate America
France, and the Balanced Scorecard. We discuss today, and over half assessed that the folly is wide-
why, although the notion of Tableau de Bord spread in their companies (see Table 1).
appeared long before the Balanced Scorecard, the
Balanced Scorecard still represents a welcome When asked what they believed were ‘the most for-
addition to both theory and practice. The paper also midable obstacles in dealing with the folly,’ the panel
illustrates how tools such as Tableau de Bord and identified three themes, all related to the way the
Balanced Scorecard can be cascaded down the firms measure, evaluate and reward performance:
organization to support the development and
implementation of strategy, identifies some ❖ Inability to break out of the old ways of thinking
implementation difficulties that companies may about reward and recognition practices,
encounter, and discusses the role of top manage-
ment and controllers in the definition and use of ❖ Lack of overall system view of performance fac-
such tools.1  1998 Elsevier Science Ltd. All rights tors and results (too much emphasis on sub-unit
reserved performance), and
❖ Continuing focus on short-term results by man-
Among the hundreds of articles dealing with ‘man- agement and shareholders.2
agement’ that get printed every year, a few stand the
test of time and become ‘classics’. One of these ‘clas- This issue is particularly problematic in the context
sics’ was written over twenty years ago by Steven of strategy implementation. While most firms track
Kerr, now vice president and Chief Learning Officer and report multiple performance indicators, they
at GE. Entitled ‘On the folly of rewarding A, while often end up putting most emphasis on profitability
hoping for B,’ the article described how many compa- measures. For example, many companies talk exten-
nies’ performance measurement systems were sively during the year about quality of products and
rewarding different behaviors than the ones they services, and/or customer and employee satisfaction,
hoped to obtain from their employees (Kerr, 1975). but end up evaluating and rewarding managers larg-
ely based on profit. Inspired by evidence supporting
Our recent observations suggest that the ‘folly’ ident- a positive link between stock price and Economic
ified by Kerr is still alive today. Many others agree Value Added, many companies have recently
with this assessment. For example, when Academy of decided to focus more intensely on EVA and share-
Management Executive (the practitioner journal of the holder wealth. EVA was recently discussed in the

190 European Management Journal Vol 16 No 2 April 1998

 IMPLEMENTING CORPORATE STRATEGY: FROM TABLEAUX DE BORD TO BALANCED SCORECARDS

Table 1 Management Reward Follies

Kerr’s Common Management Reward Follies (1995 version)a

We hope for.... But we often reward....

Long term growth; environmental responsibility Quarterly earnings
Teamwork Individual staff
Setting challenging ‘stretch’ objectives Achieving goals; ‘making the numbers’
Downsizing; rightsizing; delayering; restructuring Adding staff; adding budget; adding Hay points
Commitment to total quality Shipping on schedule, even with defects
Candor; surfacing bad news early Reporting good news, whether it’s true or not; agreeing with
the boss, whether or not (s)he’s right

Contemporary Management Reward Follies, as seen by the Academy of Management Executiveb

We hope for.... But we often reward....

Teamwork and collaboration The best team members
Innovative thinking and risk taking Proven methods and not making mistakes
Development of people skills Technical achievements and accomplishments
Employee involvement and empowerment Tight control over operations and resources
High achievement Another year’s effort
a
Kerr, S. (1995) ‘An Academy classic: on the folly of rewarding A, while hoping for B’. Academy of Management Executive February.
b
‘More on the folly’ (1995) Academy of Management Executive February.

European Management Journal by David Young and goes further than what most French and Amer-
(‘Economic Value Added: A Primer for European ican companies were doing. Following is a compari-
Managers’, August 1997). Although healthy in many son of the two tools, highlighting the advantages and
respects, this emergence (or re-emergence?) of the disadvantages of each (see Appendix, Feature 1 on
shareholder value concept is unlikely to solve by ‘The Basic Performance Measurement Problem’.)
itself the ‘folly’ identified by Kerr.

While EVA has the benefit of making managers more

sensitive to the cost of capital, it is based on account- Tableau De Bord in France
ing earnings and thus suffers from some of account-
ing earnings’ shortcomings. In particular, profit cap-
tures performance with a time lag and can thus be A Tableau de Bord is a ‘dashboard’, such as the one
improved by actions that hurt future profitability on which plane pilots and car drivers can observe the
(such as decreasing investment in customers, R & D speed at which they are going, how many miles they
or employees). EVA, in itself, also fails to identify the have covered so far, and how much fuel they are con-
levers that managers must act on to improve future suming. The Tableau de Bord performance manage-
profitability and shareholder wealth. Developing an ment system emerged in France at the turn of this
obsession with EVA is thus not a sufficient answer. century. It was first developed by process engineers
who were looking for ways to improve their pro-
Five years ago, in the first of three Harvard Business duction process by better understanding cause–effect
Review articles on the subject, Robert Kaplan and relationships (the relationships between actions and
David Norton proposed a specific framework to process performance). The same principle was then
facilitate the translation of strategy into action.3 This applied at the top management level, to give senior
framework, called the Balanced Scorecard, is a short managers a set of indicators allowing them to moni-
document summarizing succinctly a set of leading tor the progress of the business, compare it to the
and lagging performance indicators grouped into goals that had been set, and take corrective actions.
four different perspectives: financial, customer,
internal processes, and learning and growth (see Fig- This initial objective, giving managers a succinct
ure 1). overview of key parameters to support decision mak-
ing, has two important implications: First, the Tab-
The idea of having some form of balanced picture leau de Bord cannot be a single document applying
of company performance is not new in itself; many equally well to the whole firm; because each sub-unit,
companies have for years tracked and reported mul- and in fact each manager, has different responsi-
tiple indicators, and many countries have had parti- bilities and objectives, there should be one Tableau
cular traditions. In France, for example, companies de Bord for each sub-unit. These ‘dashboards’ should
have been using a related tool called ‘Tableau de be integrated in a nested structure, like the one illus-
Bord’ for over 50 years. Still, Kaplan and Norton’s trated by a set of Russian dolls (see Figure 2). In this
Balanced Scorecard represents a welcome addition context, the firm’s overall Tableau de Bord would

European Management Journal Vol 16 No 2 April 1998 191

IMPLEMENTING CORPORATE STRATEGY: FROM TABLEAUX DE BORD TO BALANCED SCORECARDS

Figure 1 Translating Vision and Strategy: Four Perspectives. Adapted from Kaplan, R.S. and Norton, D.P. (1996)
‘Linking the Balanced Scorecard to Strategy’. California Management Review July.

Figure 2 Nested Tableaux de Bord. Adapted from de Guerny, J., Guiriec, J.C. and Lavergne, J. (1990) Principes et
Mise en Place du Tableau de Bord de Gestion, 6th edn. Delmas.

translate into a series of documents supporting local tives’ of each unit. (The French use the term ‘strategy’
decision making. less often than Americans do, but the principle
involved here is the same.) The development of the
Secondly, the various Tableaux de Bord used within Tableau de Bord thus involves translating the unit’s
the firm should not be limited to financial indicators. vision and mission into a set of objectives, from
Operational measures often give better information which the unit identifies its Key Success Factors
on the impact of ‘local’ events and decisions, and (KSF), which then get translated into a series of quan-
thus on cause–effect relationships, than overall fin- titative Key Performance Indicators (KPI).

冎
ancial indicators.
Mission
⇒ Objectives ⇒ KSF ⇒ KPI
Many books have been written by French writers to Vision
explain the concept of Tableau de Bord and how to
implement it within a firm.4 One of their common To provide managers with information they can use
messages is that the Tableau de Bord needs to be for decision making, the Tableau de Bord should
developed in the context of the ‘mission’ and ‘objec- primarily contain performance indicators that are lar-

192 European Management Journal Vol 16 No 2 April 1998

 IMPLEMENTING CORPORATE STRATEGY: FROM TABLEAUX DE BORD TO BALANCED SCORECARDS

gely ‘controllable’ by the sub-unit (see Appendix Fea- The Concept of Balanced Scorecard
ture 1 on ‘the Basic Measurement Problem’). At the
same time, sub-units often need to collaborate on
interdependent tasks and projects (e.g., manufactur- The concept of Balanced Scorecard came out of the
ing and process engineering must collaborate to realization that no single performance indicator can
optimize today and tomorrow’s production capture the full complexity of an organization’s per-
processes). Such areas of interdependence should be formance. Financial indicators, for example, are typi-
identified, and then reflected by choosing indicators cally considered to be ‘lagging indicators of perform-
that capture the interdependence and encourage sub- ance,’ because they record the effect of decisions not
units to collaborate effectively. For example, the when the decisions are made, but rather as the finan-
manufacturing and process engineering departments cial impact of these decisions materializes, which can
could both track and report production cycle time be long after the decision was made. As a result they
and quality. tend to be less proactive indicators of potential prob-
lems than operational (non-financial) indicators (see
Concretely, the Tableau de Bord document should Appendix Feature 2 for an illustration).
report actual performance of the (sub-)unit on a small
number of indicators; conciseness is important and Many articles published in the 1980s following the
the danger of overloading managers with infor- TQM movement emphasized the need to comp-
mation is often highlighted by French authors. The lement financial indicators with non-financial ones.
report should include numbers covering the period Several articles went in fact a step further and rec-
since the last report, and may also present cumulat- ommended reducing the focus on financial indi-
ive performance since the beginning of the year. cators, which were said to encourage managers to
Actual performance should be compared to some make decisions that were not in the best interests of
yardstick chosen on the basis of both past perform- the company.
ance and external benchmarking.5 The periodicity of
the report should be a function of the unit’s responsi- We now know that financial and non-financial indi-
bilities and the nature of the data, but a monthly cators should not be viewed as substitutes. While
revision is considered typical at the top manage- financial measures tend to capture the impact of
ment level. decisions with some time lag, they also have three
important benefits: They represent the impact of
The Tableau de Bord is more than a mere document; decisions in a comparable measurement unit —
French writers position it within an overall manage- money, which allows aggregation of results across
ment approach. Aside from the document itself, units; they capture the cost of trade-offs between
which is useful in its own right to support local resources, and they capture the cost of spare capacity
decision making, a firm can derive much value from (see Appendix Feature 2 for an illustration).
the Tableau de Bord development process, which forces
each unit, function or division to identify its objec-
While several organizations were tracking and
tives, Key Success Factors and areas of interdepen-
reporting many non-financial indicators, Kaplan and
dence with other sub-units. In addition, the dis-
Norton went one step further by proposing a frame-
cussions that will be triggered by the publication of
work, called Balanced Scorecard, that has four
the Tableau de Bord will provide opportunities for
important characteristics:
learning throughout the firm and for reinforcing the
firm’s overall mission.
❖ it presents on a single document, a series of indi-
Altogether, the Tableau de Bord document and the cators providing a more complete view of the
processes in which it is embedded could thus yield company’s performance;
four types of benefits. ❖ this document is supposed to be short and connec-
ted to the company’s information system for
1. Provide each manager with a periodic succinct further detail (rather than the monthly ‘book’ that
overview of the performance of its unit to guide many organizations still produce and which
decision making. requires enormous managerial time and skill to
2. Inform the next level up of the sub-unit’s perform- digest);
ance (a complement to decentralization of ❖ instead of listing indicators in an ad hoc manner,
responsibilities). the Balanced Scorecard groups the indicators into
3. Force each sub-unit to (a) position itself within the four ‘boxes,’ each capturing a distinct perspective
context of the firm’s overall strategy and the on the company’s performance (see Figure 1
responsibilities of other sub-units, and (b) identify above);
corresponding KSF and KPI. ❖ last, but maybe most important, performance indi-
4. The company’s overall Tableau de Bord and its cators presented in the Balanced Scorecard must
sub-unit applications contribute to structuring be chosen on the basis of their link with the vision
management’s agenda and directing managerial and strategy of the firm/unit. Rather than starting
focus and discussions. from the set of performance measures already

European Management Journal Vol 16 No 2 April 1998 193

IMPLEMENTING CORPORATE STRATEGY: FROM TABLEAUX DE BORD TO BALANCED SCORECARDS

available within the firm, the selection process indicators, and (2) alignment of the sub-units within
should be a conscious, deductive effort starting the company’s overall vision and strategy. Figure 3
from the objectives the firm is trying to achieve and Figure 4 provide examples of these processes.
and the critical means that will get it there. This
process often results in the selection of perform-
Figure 3 shows the Balanced Scorecard for an
ance indicators that are not currently available,
Insurance company affiliated with a Bank, itself part
and for which a data collection process must be
of a large Banking group. The insurance company
developed.
developed a Scorecard for the company as a whole,
featuring the four ‘traditional’ perspectives. It also
Why four perspectives? The financial perspective
asked its various sub-units to develop scorecards at
focuses on the shareholders’ interests: is the company
their own level. The complete scorecard for the ‘cus-
generating satisfactory return on investment and cre-
tomer services’ sub-unit is also shown in Figure 3,
ating shareholder value? The three other perspectives
while Figure 4 describes in greater detail how one
can be explained through the following reasoning:
perspective (in this case, the ‘customer perspective’),
how does a company succeed financially? Through a
can be ‘cascaded down’ the organization.
combination of two elements: One is creating value
for customers; we thus need to know how customers
perceive our performance. But a firm can delight cus- As shown in Figure 4, all the measures selected by
tomers all the way into bankruptcy, so it also needs the company as a whole can be tracked at the
to make sure that it performs well on key internal regional level. The regions’ ‘customer box’ is thus
dimensions. For example, we can improve customer identical to that used by the overall company. The
service by having massive numbers of employees customer service unit, on the other hand, needed to
servicing customers, or by having fewer employees adapt some of the indicators to its specific conditions
whose time we utilize more efficiently and whom we (e.g., salesforce perception); its ‘customer box’ thus
support with excellent information technology. Cre- includes a mix of ‘corporate’ and ‘local’ indicators
ating value for customers only translates into share- (see Figure 3). Note also how indicators can some-
holder value if it is based on effective and efficient times be placed in different perspectives depending
key internal processes. on the unit. Teleservicing is part of the customer per-
spective for the ‘customer service unit,’ but it is con-
The next step is to make this value creation sus- sidered part of ‘business efficiency’ for the company
tainable over time. The company may create value as a whole.
for customers and make excellent use of its resources
today, but the world does not stand still and perform-
More generally, cascading the Balanced Scorecard
ance requirements keep ratcheting up over time. To
involves two inter-related processes: taking the part
make sure that the company will still be appreciated
of the overall strategy and indicators that are appli-
by tomorrow’s customers and will keep making
cable to the sub-unit, and designing other indicators
excellent use of its resources, the organization and its
that reflect local needs. While the Balanced Scorecard
employees must keep learning and developing. This
concept is meant to translate a unit’s strategy and is
perspective should thus group indicators capturing
thus best applied to ‘units’ that have a strategy, it can
the company’s performance with respect to inno-
be cascaded all the way down to individual man-
vation, learning and growth.
agers, who could use the scorecard’s four perspec-
tives to organize their personal goals and anchor
From this reasoning comes a set of four perspectives,
them in the larger unit’s strategic framework.6
each characterized by a small set of performance
measures. As in the Tableau de Bord, the specific con-
tent of these four ‘boxes’ must be adapted to the cir- A Balanced Scorecard contains a set of performance
cumstances of each organization. In particular, the metrics, some considered ‘lagging indicators’, others
four sets of indicators should reflect and oper- considered ‘leading indicators’. In practice the notion
ationalize the organization’s mission and strategy. A of leading vs. lagging indicator should really be
company following a low cost strategy will have dif- thought of as a continuum. Consider the following
ferent Key Success Factors than one creating value example, this time taken from a manufacturing
through very innovative products targeted at a sub- environment: Customer satisfaction is a leading indi-
set of the overall market. These two organizations cator of financial performance, but — assuming on-
should track different indicators to assess how well time-delivery is an important factor for the firm’s
they are doing and guide performance improve- customers — it is also a lagging indicator of on-time-
ment programs. delivery. On-time-delivery is a leading indicator of
customer satisfaction, but it is determined in part by,
Similar to the Tableau de Bord, the concept of Bal- and thus is a lagging indicator of, production cycle
anced Scorecard can be cascaded down through the time and quality of both product and process. It is
organization to achieve a dual purpose: (1) customiz- possible to reflect such relationships between means
ation of the Balanced Scorecard to the (sub-)unit by and ends, or between causes and effects, through the
identifying its own set of actionable performance Balanced Scorecard (see Figure 5).

194 European Management Journal Vol 16 No 2 April 1998

 IMPLEMENTING CORPORATE STRATEGY: FROM TABLEAUX DE BORD TO BALANCED SCORECARDS

Figure 3 Translating Corporate Objectives into Functional Objectives.

Figure 4 Cascading the Strategy through Nested Scorecards.

Main Benefits of Implementing a Tool summarizes in a single, succinct document, four dif-
ferent perspectives on the company’s performance.
Such as Balanced Scorecard or Tableau Most organizations collect some performance meas-
De Bord ures addressing some or all of these perspectives, but
these measures are typically reported in several dif-
Developing and using a Balanced Scorecard or Tab- ferent documents, often bulky, and typically contain-
leau de Bord can bring several benefits to a firm, ing too much information to allow good analysis. Bal-
some quite obvious, others more subtle. The most anced Scorecards and Tableaux de Bord group a
obvious advantage is that the Balanced Scorecard small set of selected indicators on a single, succinct
document.

European Management Journal Vol 16 No 2 April 1998 195

IMPLEMENTING CORPORATE STRATEGY: FROM TABLEAUX DE BORD TO BALANCED SCORECARDS

Figure 5 Organizing Performance Indicators in a Causal Chain. Adapted from Kaplan, R.S. and Norton, D.P. (1996)
‘Using the Balanced Scorecard as a strategic management system’. Harvard Business Review January–February.

Because the four perspectives provide a more ‘balan- ence of a Balanced Scorecard does not eliminate the
ced’ view of the company, they also allow managers need for top managers to explain what they are try-
to keep an eye on the way performance is achieved ing to achieve and why, but it reinforces traditional
(on the means used by managers). In particular, the means of communication by translating the strategy
Balanced Scorecard highlights trade-offs between into quantifiable indicators. The firm’s Balanced Sco-
measures. A liquor company we worked with was in recard, in turn, can be translated into ‘local’ score-
the habit of pushing sales to distributors during the cards for lower level units, thus ‘cascading’ the strat-
last week of each period, thus overloading the distri- egy and creating a set of ‘nested’ performance
bution pipe line and ‘borrowing’ sales from the next management systems.
period. Everybody knew that the practice was com-
monly used, but no one (but the people close to the Of course, tools like Balanced Scorecard and Tableau
operations) knew exactly the amounts involved. In a de Bord will have little impact if they are never dis-
Balanced Scorecard context, such practices could eas- cussed. Rather than being used as a part of manage-
ily be spotted through various measures focused on ment-by-exception, where indicators are only dis-
customer satisfaction and number of weeks of dis- cussed if they fail to reach some pre-set standard, the
tributor inventory. Balanced Scorecard should be part of a form of ‘inter-
active control’ and involve frequent and regular
Beyond these obvious benefits, the Balanced Score- attention from operating managers through face-to-
card is also a way for the company to communicate face meetings of superiors, subordinates and peers.
and reinforce its strategy through its ranks. The pres- Rather than discussing numbers for numbers’ sake,

196 European Management Journal Vol 16 No 2 April 1998

 IMPLEMENTING CORPORATE STRATEGY: FROM TABLEAUX DE BORD TO BALANCED SCORECARDS

discussions should be used to challenge and debate and Norton’s Balanced Scorecard from a conceptual
the underlying data, assumptions and action plans. point of view. In practice, however, French Tableaux
Unlike management-by-exception, interactive control de Bord tend to fall significantly short of the Bal-
is not limited to unfavorable variances, it is system- anced Scorecard.
atic; we always discuss where we stand on this met-
ric.7
❖ First, the French Tableaux de Bord we have
In this context, quantitative data are used not as an observed tend to over-emphasize financial meas-
end in themselves but rather as a means to understand ures and to contain much less non-financial meas-
and improve the underlying activities. The Balanced ures than books on Tableaux de Bord recommend.
Scorecard then contributes to learning by structuring A comparative study of French and American
the agenda for meetings and discussions. Such dis- companies published shortly after Kaplan and
cussions can occur when reviewing the unit’s per- Norton’s first Balanced Scorecard article reported
formance for the month (or the quarter), when evalu- similar findings.8
ating new investments (how does this investment ❖ This under-emphasis on non-financial measures is
impact the four perspectives presented on our not due to ‘lack of space on the report’; French
scorecard), and/or during the preparation of the Tableaux de Bord tend to be significantly longer
unit’s strategic plan and budget. As mentioned than the ideal Balanced Scorecard and than rec-
above, the process can even be cascaded down to ommended in textbooks. They also often tend to
individual managers and thus serve as a framework collect and disseminate existing performance indi-
for management by objectives. cators, rather than starting from the unit’s vision
and strategy to deduct which indicators should be
Finally, many companies have derived benefits from collected and reported.
the very process of developing a Balanced Scorecard.
Just as many companies learn a lot about their activi- ❖ Also contrary to written advice, many companies
ties and processes from implementing Activity- chose goals and targets that were mostly internal,
Based-Costing, many firms realized while developing with comparisons to last year’s performance or
a Balanced Scorecard that they did not really have this year’s budget, as opposed to systematic
clear views on the strategy they were pursuing, nor benchmarking of best-in-class performers.
on the key success factors of this strategy! In some ❖ Writings on the French Tableau de Bord go back
cases, senior management realized that they could more than forty years and often fail to highlight
not articulate a clear strategy. In other cases, the important notions that we have learned over the
problem was the opposite: too many people were art- last few years. In particular, measures described
iculating different views on what the strategy of the in books on Tableau de Bord tend to be gathered
firm really was! In both types of situation, develop- internally inside the firm, rather than externally
ment of a Balanced Scorecard can force the top man- from customers. French writers also tend to refer
agement team to sit down and develop a clear and to the company’s or the sub-units’ ‘mission and
shared view on what they are trying to achieve objectives’, rather than refer explicitly to their
(‘where do we want to go?’), and what are the critical ‘strategy’ as we would be more likely to do today.
levers and means to reach these objectives (‘how do ❖ Finally, with respect to the way the Tableau de
we get there?’). Bord was used, French managers seem to have
often fallen into the trap of using the Tableau de
This process can also act as a trigger for development Bord as a device supporting management-from-a-
of information technology capabilities allowing the distance and management-by-exception, rather
firm to track and disseminate critical performance than using it interactively to create an agenda for
indicators. As Michael Hegarty, Vice Chairman of discussions and meetings. As a result the Tableau
Chemical Bank, explained: ‘The scorecard also has de Bord lost much of its power and usefulness.
served as a catalyst for MIS improvements. We found
we could not track all Balanced Scorecard measures
even though we thought they were important and There are indications that French Tableaux de Bord
even though we are in the information business. So are improving rapidly; several recent articles in
we are now devoting substantial energy and French business magazines report that an increasing
resources to enhancing our ability to manage stra- number of companies are collecting and reporting
tegic information.’ (Hegarty, 1996) non-financial indicators, particularly external data on
customer perception. The increasing use of integrated
computer systems such as SAP and Oracle is also
helping firms to collect, process and integrate data
Did the French Tableau de Bord Have that goes beyond ‘traditional’ accounting measures.
It All? Still, it is fair to say that, at least at the top manage-
ment level, the content and style of use of French
As the previous sections make clear, the French companies’ Tableaux de Bord often fell short of the
notion of Tableau de Bord is quite close to Kaplan theory on the subject.

European Management Journal Vol 16 No 2 April 1998 197

IMPLEMENTING CORPORATE STRATEGY: FROM TABLEAUX DE BORD TO BALANCED SCORECARDS

Grouping Indicators Into ‘perspectives’ default (as has happened in many French
companies), and the danger of missing one of the
four dimensions proposed in the Balanced Scorecard.
Some French authors point out that the notion of Tab-
Because they have to be collected for external pur-
leau de Bord is wider and more general than Kaplan
poses and will always receive much attention from
and Norton’s Balanced Scorecard, which they regard
shareholders, financial measures have in-bred advan-
as a ‘special case’ of Tableau de Bord. These authors
tages with respect to top management attention and
base their argument on the fact that texts on the Tab-
are bound to ‘creep up’ on top managers unless their
leau de Bord strongly emphasize the need to tailor-
attention is specifically focused in other directions
make the Tableau de Bord to each company and to
as well.
each manager within the company. As a result, writ-
ings on the French Tableau de Bord generally do not
specify structured sets of indicators, while Kaplan
and Norton’s Balanced Scorecard proposes four gen- Implementation Issues
eric sets of indicators (Chiapello and Lebas, 1996).

The Balanced Scorecard approach can thus look more Introducing a Balanced Scorecard means introducing
rigid than the notion of Tableau de Bord and be change in a company, which is never an easy process.
faulted by some as disregarding potentially Changes affecting the availability of performance-
important dimensions of firm performance. Some related information within the firm can be parti-
companies could indeed want to create other ‘boxes’ cularly threatening, as they have the potential of
of indicators beyond the initial four perspectives pro- modifying the balance of power within the organiza-
posed by Kaplan and Norton. Two companies we tion. Information can indeed be power, and in more
studied, for example, both added a fifth perspective ways than one.
(‘employee perspective’ in one case, ‘impact on
society’ in the other). Table 2 presents examples of Companies trying to implement a Balanced Score-
possible additional perspectives. card-type of mechanism can expect to encounter four
types of difficulties:
In all fairness, however, this is not really a limitation
of the Balanced Scorecard model. First, some of the 1. The first problem that many firms encounter is the
indicators addressing the ‘other’ dimensions can (and realization that the top management team cannot
often should) be included in the four boxes proposed articulate a clear and shared view of the firm’s
in the framework. In particular, employee-related strategy; in some cases, the strategy is not clear,
indicators often feature prominently in the three in other cases members of the top management
‘non-financial’ perspectives. Secondly, Kaplan and team hold different views on what the strategy of
Norton’s four perspectives are presented as an the firm is or ought to be. The first step of the
organizing framework rather than a constraining process is thus to get to a consensus on what the
straightjacket. Nothing prevents companies from firm should try to achieve.
adding one or two additional boxes, although part of 2. Developing and maintaining a Balanced Scorecard
the power of the Balanced Scorecard comes from its can create a workload for many people. In parti-
conciseness and the clarity of its presentation; it is cular, some of the data required may not currently
thus probably better to try to keep the number of exist within the firm and thus needs to be collected
boxes rather small. specifically for the Scorecard. Managers that are
often already stretched by their normal workload
Finally, having a framework helps protect potential may not be enthusiastic about this additional
users against two dangers: the danger of one parti- demand on their time. Furthermore, many compa-
cular perspective, for example the financial perspec- nies have a track record of starting and later aban-
tive, coming to dominate the other perspectives by doning initiatives like the Balanced Scorecard. As

Table 2 Potential Additional Perspectives for the Balanced Scorecard

Impact of the firm on society

Firm’s impact on employment, local communities and environment; firm’s record on health and safety
Consumers vs customers
For firms selling through distributors and which have two levels of constituencies: their immediate customers and the final
consumers
The firm’s economic, political and social environment
Environmental contingencies that can have a major impact on the firm’s performance (e.g., price of raw materials, overall
industry performance, etc.)
Major projects currently underway
Key milestones for the firm’s major projects (to ensure that progress on these projects — an important leading indicator of
future performance — receives as much attention as necessary)

198 European Management Journal Vol 16 No 2 April 1998

 IMPLEMENTING CORPORATE STRATEGY: FROM TABLEAUX DE BORD TO BALANCED SCORECARDS

a result some employees may have grown weary While the need for consistency is common sense and
of such change efforts and may have developed easy to accept in principle, displaying such consist-
an attitude along the lines of ‘if I wait long ency is not easy for top managers who have reached
enough, this will go away!’ Top managers inter- their position in part because of their ability to reach
ested in the Scorecard concept may thus encounter financial targets. This need for consistency is not spe-
some cynicism among their troops when they cific to scorecards or Tableaux de Bord; the same
bring up the idea. problem exists with any project that goes beyond
3. Aside from resistance to increased workload for short-term financial results, e.g., Quality or Just-in-
an initiative that may or may not have clear bene- Time programs. Just as managers should not talk
fits for the managers involved, companies may about quality for all but the last two days of the
encounter resistance motivated by a desire to pro- month, they cannot expect subordinates to develop a
tect one’s turf or power base. Senior managers broader view of performance encompassing a more
should not assume that the absence of specific balanced Scorecard if they do not consistently
quantitative indicators in a firm is always due to reinforce this broader view themselves.
ignorance or excessive workload. In many cases,
such absence of information reflects what we call
opaqueness by design. Local managers have learned
to develop secondary sources of information that To Link, or Not to Link the Balanced
are not accessible to top management and/or to Scorecard to Compensation?
their subordinates. Maintaining opaqueness can be
a way for these managers to centralize authority
(thus taking power from lower levels of the One way for top management to show focus and con-
organization), and/or to protect themselves from sistency in outlook is to link employees’ performance
scrutiny and questioning by their boss (thus taking evaluation and reward to performance on the Bal-
power from upper levels). anced Scorecard. Some companies have started doing
so; the insurance company Cigna Property and Casu-
In particular, the Balanced Scorecard highlights alty, for example, has developed a series of Score-
trade-offs and thus brings increased transparency, cards at various levels of the organization (corporate,
which may be threatening for some managers. In divisions and business units), and awards all of its
one company we worked with, for example, local employees incentive compensation based on ‘bal-
managers were under intense pressure to meet anced scorecard performance’ of the unit most rel-
quarterly profit targets. For years, they had dealt evant for each employee (Cigna Property and Casu-
with this pressure by using a variety of debatable alty, 1996).
mechanisms to reach these targets, including
active ‘management’ of advertising and promotion Similarly, the US Marketing and Refining Division of
expenses, of temporary price rebates and of the Mobil Corporation offers a number of incentive plans
distribution pipeline. Every one knew that these associated with Balanced Scorecard performance. The
practices existed, but top management could not Division’s Balanced Scorecard has been cascaded one
tell exactly how much was going on at any point level down to ‘business units’ and the plan rewards
in time. These practices, however debatable, were employees based on Balanced Scorecard performance
an important ‘safety valve’ for local managers, of both the Division and the business unit the
who would find themselves under even more employee belongs to. Each business unit was allowed
pressure should top management develop a clear to determine how much weight each measure on its
idea of some of the actions they were taking in Scorecard should carry; most business units attri-
order to meet their targets. Predictably, local man- buted some weight to all the measures on their Score-
agers were not strong supporters of increased card. One of Mobil USM & R’s regions also created
transparency! an incentive plan for all its employees, featuring five
4. Once the Balanced Scorecard is developed, it must of the region’s key performance indicators. The plan
then survive and prosper among competing triggered a group reward (all or nothing) for achieve-
reporting mechanisms. It should also evolve over ment of stretch targets going beyond the region’s
time, as the company’s (or the unit’s) environ- ‘official’ targets.9
ment, capabilities and/or strategy change. For this
to happen, top managers must be consistent in Beyond these examples, surveys by consulting firms
their decision to widen their perspective from a suggest that an increasing number of large firms are
narrow emphasis on financial measures. In one tying executive compensation of senior executives to
company we studied, for example, a manager said: Balanced Scorecard-types of indicators. In one such
‘we developed a Balanced Scorecard and used it survey, more than 60% of the 100 large organizations
in our monthly discussions. This lasted until the polled said that they linked the Balanced Scorecard
first time we failed to meet our financial targets, to incentive pay for their senior executives.10 Other
at which point we didn’t talk about the Balanced companies prefer to gain more experience with their
Scorecard anymore; we reverted back to talking Scorecard before creating a tight link with reward
only about financial performance.’ systems. Top managers of NatWest Bancorp, for

European Management Journal Vol 16 No 2 April 1998 199

IMPLEMENTING CORPORATE STRATEGY: FROM TABLEAUX DE BORD TO BALANCED SCORECARDS

example, explained that while they have started to develop a more focused and performance oriented
broaden the basis of performance evaluation and approach to their business. (The project was similar
reward, they were not yet linking the whole Score- to the (re)-introduction of a management-by-objec-
card to incentive compensation. They added that ‘the tives system). Our first contribution was to help these
pace of linkage will increase as confidence in the managers be more rigorous in articulating their mis-
quality of measures grows.’ (Tugwell et al., 1996) sion and their strategy, then in translating them into
a manageable subset of objectives. Executives often
This ‘wait while we learn’ approach is quite reason- came up with lists of 15 to 20 goals, which could in
able given the enhanced resistance that changes to fact fit into four or five major blocks of related initiat-
compensation systems tend to create, but it is clear ives.
that companies that tie significant rewards with fin-
ancial performance only are using a dangerous ‘dou-
ble talk’ when they also try to emphasize a broader
Selecting the Right Performance Indicators
outlook during Balanced Scorecard progress meet-
ings. The next step was for the managers to translate the
objectives into measurable targets, which many
For example, one of the companies we studied experienced as a very challenging process. Aside
recently was pressing managers to pay more atten- from sales and manufacturing functions, many man-
tion to customer satisfaction. While managers did not agers remain moderately skilled at identifying quan-
disagree with the need to be sensitive to customer tifiable performance indicators. This process is not as
concerns, they expressed frustration over the fact that easy as it seems; the indicators must be controllable
they were still being rewarded, financially and other- (i.e., target achievement should not be overly influ-
wise, only based on historical financial indicators. enced by events or decisions that are out of man-
Top managers told us they were not yet ready to tie agers’ control), but they should also be reasonably
incentive compensation to customer satisfaction complete (i.e., they should not fail to capture
because (a) they did not yet have enough reliable and important dimensions of performance such as
satisfactory measures of customer satisfaction, and cooperation with other business units or adaptation
(b) they did not know enough about the link between to significant external events). Because of their
customer satisfaction and future company profitabil- experience and training, controllers tend to be better
ity. skilled than line managers at identifying the right
performance indicators.
While we understand the company’s reasoning, we
have two pieces of advice to offer. First, it is urgent
for this company to clarify the set of causal factors
of its financial performance. What are the company’s
key success factors and key performance indicators?
Using the Tools Correctly
In particular, how much emphasis should this com-
pany put on customer satisfaction? What are the most Over the last few years, many companies have real-
important facets and drivers of customer satisfaction ized that financial indicators of performance are not
for their business, and how can they be measured? sufficient. Though these indicators are important,
Secondly, as Kerr pointed out over twenty years ago, companies are finding that they need additional mea-
it is unreasonable to consistently reward behavior A sures of current performance and indicators of future
and hope that managers will invest valuable time and performance to properly monitor corporate progress.
energy into behavior B, particularly when pursuing These measures need to focus on indicators of future
B can entail some short-term trade-off with perform- financial performance and relate to items such as cus-
ance on dimension A. Ultimately, companies have to tomer satisfaction, product and process quality, and
put their money where their mouth is! innovation and growth. The Balanced Scorecard is a
powerful tool that can help managers translate strat-
egy into action. It has many potential benefits which
we have tried to describe in this article. The French
Role of ‘Financial Specialists’ Tableau de Bord had many of the Balanced Score-
card’s features at the conceptual level but often fell
‘Financial Specialists’ can support the Scorecard short in practice, particularly in terms of emphasis
development process in two very significant ways given to non-financial indicators.
because of the skills they have with respect to
measurement and structured reasoning. Like all management tools, however, the Balanced
Scorecard is not a sufficient condition for success; it
cannot do everything! For example, it should not be
Structured Reasoning a tool supporting attempts at management-by-excep-
tion and management-from-a-distance. Neither is it a
On a recent consulting assignment we were asked to substitute for sound strategy, clear focus and strong
help the executive teams of several business units to alignment of energies within the firm.

200 European Management Journal Vol 16 No 2 April 1998

 IMPLEMENTING CORPORATE STRATEGY: FROM TABLEAUX DE BORD TO BALANCED SCORECARDS

On the other hand, developing and using a Balanced presented at the 19th Annual Meeting of the European
Scorecard-type of system can help develop these con- Accounting Association.
Cigna Property and Casualty (1996) A Balanced Scorecard.
ditions by forcing top management to articulate a CFO Magazine October.
strategy and Key Success Factors, and focusing man- Cooper, R. and Kaplan, R.S. (1992) Activity-based systems:
agers’ attention on the firm’s progress on these measuring the cost of resource usage. Accounting Hor-
elements. To achieve these benefits, top management izons September, 1–12.
Epstein, M.J. and Manzoni, J.-F. (1997) ‘The balanced score-
needs to show focus and consistency: focus during card and tableau de bord: translating strategy into
the design of the system, and consistency when using action,’ published by the same authors in Management
it. ‘Financial Specialists’ also have an important role Accounting August 1997, 28–36.
to play, through their understanding of measurement de Guerny, J., Guiriec, J.C. and Lavergne, J. (1990) Tableau de
issues and their capability to add structure and disci- Bord de Gestion, (6th edition, 1990).
Hegarty, M. (1996) Letter to the editor. Harvard Business
pline to discussions of the company’s strategy and Review March–April, 172.
the identification of the firm’s key success factors. Kaplan and Norton (1996) recently expanded on their ideas
in a book called The Balanced Scorecard: Translating Strat-
egy into Action. Harvard Business School Press.
Notes Kerr, S. (1975) On the folly of rewarding A, while hoping for
B. Academy of Management Journal, 769–783.
1. Some of the material presented in this article also Lauzel, P. and Cibert, A. (1962) Ratios au Tableau de Bord,
appeared in ‘The balanced scorecard and tableau de bord: 2nd edn.
translating strategy into action,’ published by the same Lebas, M. (1994) Managerial accounting in France: overview
authors (see Epstein and Manzoni, 1997). of past tradition and current practice. The European
2. See, More on the folly. Academy of Management Executive Accounting Review 3(3).
February 1995. Moisson, M. (1968) Pratique du Tableau de Bord de l’Entreprise.
3. See, The balanced scorecard — measures that drive per- Simons, R. (1995) Control in an age of empowerment. Harvard
formance. Harvard Business Review January–February Business Review March–April.
1992; Putting the balanced scorecard to work. Harvard Tugwell, J. (CEO), Bye, B. and Hatrick, K. (both vice-president
Business Review September–October 1993; and Using the Strategic Development) (1996) Letter to the editor. Harv-
balanced scorecard as a strategic management system. ard Business Review March–April, 173.
Harvard Business Review January–February 1996. Kaplan
and Norton (1996) recently expanded on their ideas in a
book called The Balanced Scorecard: Translating Strategy
into Action.
4. See, for example, Lauzel and Cibert (1962); Moisson Appendix
(1968); or de Guerny et al. (1990), which is updated
periodically. For two articles discussing the concept in
English, see Lebas (1994) and Chiapello and Lebas (1996).
5. This emphasis on benchmarking with other firms already Feature 1
appeared in Lauzel and Cibert’s 1962 textbook on Tab-
leaux de Bord (op.cit.).
6. This process of cascading the balanced scorecard to indi- The Basic Performance Measurement Problem
vidual levels is illustrated in Robert S. Kaplan’s case,
Mobil USM and R (C): Lubricants Business Unit (HBS # Tableaux de Bord and Balanced Scorecards are per-
9-197-027). formance measurement instruments. Measuring a
7. For more on the notion of interactive vs. diagnostic con-
trol, see Simons (1995).
unit’s performance can be complex task. In particular,
8. The study was conducted by two academics, one Amer- performance measures should be as complete as
ican and the other French. Their findings were reported possible, but also as controllable as possible for man-
in the journal of the French Association of Accountants agers.
(Recent evolution of the Tableau de Bord systems: com-
parison of practices in a few American and French multi-
nationals, Jack Gray and Yvon Pesqueux, Revue Française
Completeness
de Comptabilité, February 1993). A measure is complete when it captures ‘the whole
9. See the Harvard Business School cases ‘Mobil USM and truth’ about the unit’s performance. Perfect complete-
R (A): Linking the Balanced Scorecard’ (HBS # 9-197-025) ness of measurement is hard to achieve because
and ‘Mobil USM and R (B): New England Sales and Distri- actions and decisions taken today in a given sub-unit
bution’ (HBS #9-197-026), both by Kaplan, R.S.
10. See, for example, a 1996 study by Towers Perrin, New may have consequences today and/or in the future,
York, quoted in HR Focus June. in this and/or in other sub-units (see A1)
11. Some clients may do all three, although British Airways
estimates that they receive only one official complaint for
every twenty dissatisfied customers (Financial Times,
24/2/94, p.9).
12. This point was first made by Robin Cooper and Robert
Kaplan, who highlighted that reducing resource con-
sumption did not necessarily reduce spending. See
Cooper and Kaplan (1992).

Figure A.1
References
Chiapello, E. and Lebas, M. (1996) The Tableau de Bord, a An incomplete measure captures part of the impact
French approach to management information. Paper of the unit’s activities, but fails to reflect other dimen-

European Management Journal Vol 16 No 2 April 1998 201

IMPLEMENTING CORPORATE STRATEGY: FROM TABLEAUX DE BORD TO BALANCED SCORECARDS

sions. For example, it is typically easier to capture the measure so many dimensions that capture so many
immediate impact of a decision on the deciding unit trade-offs, that people reach a state of ‘information
(top left quadrant), than future impact of that decision overload’ and learn to disregard most of the data
on other administrative units (bottom right quadrant). they receive.

Controllability
A performance indicator is controllable to the extent
that it is only influenced by elements under the unit’s Feature 2
control; the indicator tells ‘nothing but the truth’
about the unit’s performance. Most of the time, this Financial Vs Non-Financial Measures
ideal is impossible to achieve. For example, the
amount of profit generated by a commodity trader is Tableaux de Bord and Balanced Scorecards typically
influenced by the trader’s decisions and actions, but contain financial as well as non-financial measures of
also by fluctuations of the market over which the performance. Both types of indicators have strengths
trader has no control. and weaknesses.

Controllability of performance indicators is The Downside of Financial Measures

important because it is often said that in order to Take an insurance company’s claim-processing unit
maximize the motivation and minimize perceptions that is organized in a variety of cross-functional cells,
of unfairness, ‘people should only be evaluated based each supervised by a first level manager. Assume
on what they control’ (the so-called ‘controllability that one of these managers decides to change the way
principle’). Controllability is also important from a work is allocated within the unit s/he manages.
learning point of view; uncontrollable dimensions Partly because of the content of the innovation and
pollute the information contained in the indicator by partly because of the way the manager presented it,
making it difficult to untangle the impact of one’s the unit’s staff reacts very negatively and their work
efforts from the effect of the ‘external’ factors (e.g., is disrupted. Some customers’ claims will be pro-
did the measure go up because of my actions or cessed as well as usual, others will experience prob-
because the overall industry is doing well?). lems (e.g., with accuracy and/or speed of processing,
discourtesy over the phone, etc.).
Completeness and Controllability: Often a Trade-Off
Performance is harder to measure whenever two or Among the customers who experience problems,
more sub-units are inter-dependent, because inter- some proportion will complain to the company, some
dependence makes it hard to trace the costs and proportion will complain to their friends, and some
benefits of each decision to the source of the decision. will decide to take their business elsewhere.11 The
As a result costs and benefits of activities could end customers who take their policy to the competition
up being recorded in different administrative units will not do so immediately; it will take a few weeks,
(e.g., the cost of rework undertaken by department a few months, in some cases. Similarly, some of the
B to correct mistakes introduced at department A; or friends who came to know about these customers’
lost sales due to quality problems created by problem will move to the competition. Some will
manufacturing). One way to deal with this problem move now, others will move after one more negative
is by measuring performance at a higher level in the incident, etc.. The impact of the unit’s manager inno-
organization (e.g., we can measure performance at vation on the firm’s revenue will thus take some time
the division level rather than for each function). But to materialize, and it will probably never be traced
enlarging the administrative unit also dilutes the to him/her. In contrast, it would have been possible
impact of individual decisions, thus decreasing the to capture some of the impact of the unit manager’s
extent to which the indicators capture ‘nothing but’ actions by measuring the performance of the unit, its
the activities of a specific group of employees (i.e., ‘well being,’ and customers’ reactions. Frequent
decreasing indicator controllability for each member measurement of speed and accuracy of processing,
of the unit). courtesy over the phone, of the unit’s
morale/attitude and of customer
Another way of obtaining more completeness in satisfaction/complaints, would have allowed the
measurement, this time without enlarging too much identification that ‘something was wrong,’ which
the size of the unit being measured, is to use several could then become the basis for an analysis and a
performance indicators to try to capture more facets of discussion. In this (as well as most cases), non-finan-
performance. Increasing the number of performance cial indicators capture the manager’s performance
indicators probably involves decreasing marginal more quickly than financial measures do.
returns, however. First, because measurement typi-
cally does not come for free; there are data collection The Upside of Financial Measures
and processing costs. Secondly, because people can An airline’s baggage handling department redesigns
only digest and act upon a limited amount of infor- some of its processes and, with the help of more
mation; they can take very seriously only a limited modern equipment and better work routines, suc-
number of performance indicators. Some companies ceeds at becoming more effective and more efficient.

202 European Management Journal Vol 16 No 2 April 1998

 IMPLEMENTING CORPORATE STRATEGY: FROM TABLEAUX DE BORD TO BALANCED SCORECARDS

It is now making fewer mistakes and processes lug- ate more spare capacity than real profit increases.
gage more quickly, which lowers the burden on the Spare capacity offers an opportunity to improve fin-
staff. The department is now performing better, but ancial performance, but management must take
is the airline making more money? In time, improved actions to capitalize on this opportunity.12 Actions
baggage handling may lead to increased sales can go along either, or both of the following lines:
(because of increased customer satisfaction), which increasing sales (through increased net price and/or
would improve profit. In the short run, however, the increased volume), and/or managing the spare capacity
airline had to spend money on equipment, and its out of the system (which means laying off people, sell-
savings will be minimal as long as the baggage hand- ing equipment, etc.). It could be one, it could be the
ling department still employs the same number of other, it could be a little bit of both, but it’s got to
people. Some of the employees may work fewer be something!
hours, which might save some overtime pay
depending on the union agreement, but that does not This reality is easy to overlook when one looks at
take away their basic salary. The same reasoning non-financial measures only. Non-financial measures
applies to a bank restructuring its back office oper- are very helpful in guiding operational improve-
ation to centralize it into service centers. The centers ments but they do not capture how much it cost to
will be more efficient than the branches’ back offices, improve operations, nor whether the spare capacity
but removing the work from back offices does not, liberated by the operational improvements has been
by itself, make the branch space and employees go used and/or re-deployed. Financial measures can
away. help answer both questions because they capture
trade-offs between resources and they capture the
In the short run, process improvements often gener- cost of capacity.

MARC EPSTEIN, JEAN-FRANÇOIS MAN-

INSEAD, Boulevard de ZONI, INSEAD, Boul-
Constance, 77305 Fontai- evard de Constance,
nebleau, France. 77305 Fontainebleau,
France.
Marc J. Epstein is
presently Price Water- Jean-François Manzoni is
house Visiting Professor presently Assistant Pro-
of Accounting and Con- fessor of Accounting and
trol at INSEAD, France. Control at INSEAD, and
Formerly a professor at Affiliate Member of the
both Harvard Business Corporate Renewal
School and Stanford Business School, he has pub- Initiative (CORE) and Member of the Centre for
lished numerous articles and books, including the Advanced Learning Technologies (CALT).
recent award winning book Measuring Corporate Research interests centre on the management of
Environmental Performance: Best Practices for Costing change, especially behavioral issues associated
and Managing an Effective Environmental Strategy with the development and use of performance
(Irwin, 1996). Research interests include manage- measurement, evaluation and reward systems.
ment control and the development of performance
metrics to link to corporate strategies.

European Management Journal Vol 16 No 2 April 1998 203