The development of the Balanced Scorecard as a strategic management tool
By I.M. Cobbold and G.J.G. Lawrie; 2GC Active Management Ltd., Maidenhead, UK
1. The Balanced Scorecard and its development
The Balanced Scorecard was first introduced in the early 1990s through the work of Robert Kaplan and David Norton of the Harvard Business School. Since then, the concept has become well known and its various forms widely adopted across the world.
By combining financial measures and non-financial measures in a single report, the Balanced Scorecard aims to provide managers with richer and more relevant information about activities they are managing than is provided by financial measures alone. Kaplan and Norton proposed that the number of measures on a Balanced Scorecard should also be constrained in number, and clustered into four groups. Beyond this, the original definition of Balanced Scorecards was sparse.
In essence the Balanced Scorecard has remained unchanged since these early papers, having at its core a limited number of measures clustered into groups, and an underlying strategic focus. But modem Balanced Scorecard designs also have a number of features that clearly differentiate them from earlier examples. This paper describes these changes as an evolution through three distinct 'generations' of Balanced Scorecard design.
EDITORS NOTE: This is an advanced copy of a soon to be published article that was presented at a Balanced Scorecard conference in Washington in July 2002. The authors have developed an adaptation of the Balanced Scorecard for use by staff functions, such as Human resources. The full text (including a detailed reference list) can be downloaded (PDF format) below.
2. First generation Balanced Scorecard
The Balanced Scorecard was initially described as a simple "4 box" approach to performance measurement. In addition to financial measures, managers were encouraged to look at measures drawn from three other "perspectives" of the business: Learning and Growth; Internal Business Process; and Customers, chosen to represent the major stakeholders in a business.
Definition of what comprised a Balanced Scorecard was sparse, and focused on the high level structure of the device. Simple 'causality' between the four perspectives was illustrated, but not used for specific purpose. Kaplan and Norton's original focus was on the selection and reporting of a limited number of measures in each of the four perspectives. The paper suggested use of attitudinal questions relating to the vision and goals of the organisation to he] p in the selection of measures to be used, and encouraged the consideration of 'typical' areas of interest in this process.
Kaplan and Norton's original work makes no specific observations concerning how the Balanced Scorecard might improve the performance of organisations; the implication is that the provision of accessible relevant measurement data itself will trigger improved organisational performance. However, they do imply that the source of these improvements is changes in behaviour: In the light of this, the basis for selecting the goals represented by the Balanced Scorecard is of some importance. But in their first paper Kaplan and Norton say little about how a Balanced Scorecard could be developed in practice beyond a general assertion that design involved "putting vision and strategy at the centre of the measurement system". Later writing includes increasing amounts of proscription about development methods, concluding with a lengthy description of one such process in their first book on the subject published in 1996
First generation Balanced Scorecard
3. Practical Experiences with first generation Balanced Scorecards
The authors' professional experience suggests that First Generation Balanced Scorecards are still being developed, and that they probably still form the large majority of Balanced Scorecard designs introduced into organisations. This is reflected in the literature, where books and articles that use more advanced representations of Balanced Scorecard are only recently appearing. But, despite its huge popularity as a concept, and apparently widespread adoption, relatively few detailed case studies concerning Balanced Scorecard implementation experiences appear to exist in the academic literature.
4. Second generation Balanced Scorecard: strategic linkages
The practical difficulties associated with the design of I at Generation Balanced Scorecards are significant, in part because the definition of a Balanced Scorecard was initially vague, allowing for considerable interpretation. Two significant areas of concern were filtering (the process of choosing specific measures to report), and clustering (deciding how to group measures into 'perspectives '). Discussions relating to clustering continue to be rehearsed in the literature, but discussions relating to filtering are less common, and usually appear as part of descriptions of methods of Balanced Scorecard design.
Perhaps the most significant early change translated the attitudinal approach to measure selection proposed initially be Kaplan and Norton (e.g. "To succeed financially, how should we appear to our shareholders?") into a process that yielded a few appropriate key measures of perfonnance in each perspective. A solution was the introduction of the concept of 'strategic objectives'. Initially these were represented as short sentences attached to the four perspectives, and were used to capture the essence of the organisation's strategy material to each of the areas: measures were then selected that reflected achievement of these strategic objectives. Although subtle, this approach to measure selection quite different from that initially proposed, since strategic objectives were developed directly from strategy statements based on a corporate vision or a strategic plan.
Another key development concerned causality. Causality between the perspectives had been introduced in early' 1st Generation' Balanced Scorecard thinking. Second generation' Balanced Scorecard saw the idea of causality developed further. Instead of simply highlighting causal links between perspectives, internal documents from one consulting firm's work in 1993 shows an early attempt to indicate linkages between the measures themselves'. Measure based linkages provided a richer model of causality than before, but presented conceptual problems -for example, the use of measures encouraged attempts to 'prove' the causality between measures using various forms of analysis.
Collectively the changes in design described here represent a materially different definition of what comprises a Balanced Scorecard compared to Kaplan and Norton's original work -we will refer to Balanced Scorecards that incorporate these developments as '2nd Generation Balanced Scorecards'. The impact of these changes were characterised by Kaplan and Norton in 1996 as enabling the Balanced Scorecard to evolve from "an improved measurement system to a core management system". Maintaining the focus that Balanced Scorecard was intended to support the management of strategic implementation, Kaplan and Norton further described the use of this development of the Balanced Scorecard as the central element of ' 'a strategic management system".
One consequence of this change in emphasis was to increase the pressure on the design process to accurately reflect the organisation's strategic goals. Over time the idea of strategic linkage became an increasingly important element of Balanced Scorecard design methodology, and in the mid 1990's Balanced Scorecard documentation began to show graphically linkages between the strategic objectives themselves (rather than the measures) with causality linking across the perspectives toward key objectives relating to financial performance. An example is shown in Figure 2.
Strategic linkage Model
As objectives began to appear in graphical representations of linkages, so they began to require short titles (to fit onto diagrams). To compensate the idea of 'objective descriptions' associated with strategic objectives emerged. These descriptions, which were simply longer paragraphs describing in more detail the 'meaning' of the objective, are symptomatic of a significant increase in the volume of purely design related documentation associated with the design of Balanced Scorecards -objectives began to be assigned to owners, measures to objectives. Early software reporting systems began to enhance these elements of design information by linking it with measurement data, and using email and diary systems to enable speedy diagnosis and interventions in response to data observed: the ability to store and work with these characteristics are now central to leading 'Balanced Scorecard' software systems.
Another consequence was the increased awareness of the need to reflect differences in management agenda within differing parts of organisational structures, and so increasing attention was given to developing 'strategic alignment' between management units by developing Balanced Scorecards as part of a 'cascade' at the Business Unit level.
The representation of causality between strategic objectives -known initially as the 'Strategic Linkage Model' -is now considered to be an important part of any Balanced Scorecard design. The design elements that make up the second Generation Balanced Scorecard now represent 'mainstream' thinking on Balanced Scorecard design -as evidenced by considerable consistency of definition across a range of practitioner and academic texts.
Increasing adoption of the 'explicit' causality present in the strategic linkage model has diminished the value of 'lead' and 'lag' measures -as the predictive nature of 'lead' measures is now more clearly (and less ambiguously) documented in the design of the Balanced Scorecard.
5. Practical Experiences with 2nd Generation Balanced Scorecards
There are still areas that prove difficult to deal with during the development process for both management teams and consultants charged with developing 2nd Generation Balanced Scorecard. The first of these areas concerns the development of the Strategic Linkage Model. Management teams find the necessary selection of priority elements within their collective vision and strategic goals difficult. While there is usually some type of common reference point in the form of visions or plans, often this is either poorly defined, lacking continuity or something that the management team didn't fully agree on.
Working to choose objectives simply flushed these issues to the forefront of management attention, and triggered useful debate, but the activity of actually selecting priority objectives itself is not one that has been found to support open discussion about the collective alignment of strategic goals. Another difficult area is target setting. While measure selection is easier, thanks to Strategic Objectives and the Strategic Linkage Model, for similar reasons to those note above, organisations often lack a common reference point relating from which targets can be extrapolated. Finally, the Strategic Linkage Model documentation, although clear to those familiar with construct, has proven less helpful when used for broadcast communication of strategy -it lacks sufficient supportive information to be usefully stand alone as a communication concerning an organisation's strategic plans.
6. Third generation Balanced Scorecard: destination statements
The third generation Balanced Scorecard model is based on a refinement of the second Generation design characteristics and mechanisms to give better functionality and more strategic relevance. The origin of the developments stem from the issues relating to target setting and the validation of strategic objective selection outlined above. These triggered the development in the late 1990's of a further design element- the 'Destination Statement' -initially at the end of the design process to 'check' the objectives, measures and targets chosen.
The first Destination Statements were created as a formal consensus estimate of the consequences at a particular future date (e.g. 'in three years time') of implementing the strategic objectives previously selected for the strategic linkage model. By agreeing in this statement 'how much' of key things would have been achieved by this time (e.g. headcount, revenues, customer satisfaction, quality levels etc.) the hope was it would subsequently be easier (for example) to check for (or set) a consistent set of annual targets. Figure 3 shows an example extract from an early 'Destination Statement'.
It was quickly found that management teams were able to discuss, create, and relate to the 'Destination Statement' much easily and without reference to the selected objectives. Consequently the design process was 'reversed', with the creation of the 'Destination Statement' being the first design activity, rather than a final one. Further it was found that by working from Destination Statements, the selection of strategic objectives, and articulation of hypotheses of causality was also much easier, and consensus could be achieved within a management team more quickly. We will refer to Balanced Scorecards that incorporate Destination Statements as third generation Balanced Scorecards'.
7. Key components of a third generation Balanced Scorecard
A. Destination statement:
In order to make rational decisions about organisational activity and not least set targets for those activities, an enterprise should develop a clear idea about what the organisation is trying to achieve. A destination statement describes, ideally in some detail, what the organisation is likely to look like at an agreed future date. In many cases this exercise builds on existing plans and documents -but it is rare in practice to find a pre-existing document that offers the necessary clarity and certainty to fully serve this purpose within an enterprise.
B. Strategic Objectives:
The destination statement offers a clear and shared picture of an organisation at some point in the future, but it does not provide a suitable focus for management attention between-now and then. What needs to be done and achieved in the medium term for the organisation to "reach" its destination on time is agreed upon in the form of objectives or priorities. By representing the selected objectives on a "strategic linkage model", the design team is encouraged to apply "systems thinking" to identify cause-and-effect relationships between the selected objectives i.e. what do we need to do to achieve the results we expect. This approach also helps ensure the objectives chosen are mutually supportive and represent the combined thinking of the team's high-level perception of the business model.
C. Strategic Linkage Model and perspectives:
The chosen strategic objectives are spread across four zones or 'perspectives'. The lower two perspectives contain objectives relating to the most important activities in terms of business processes, cycle time, productivity etc. (Internal Processes) and what needs to happen for these processes to be sustained and further developed in terms of people, product and process development (Learning & Growth). The two top perspectives house objectives relating to the desired results of the activities undertaken i.e. how we wish external stakeholders (e.g. the general public, partner agencies and organisations to perceive us (External Relations) and how this will ultimately translate into financial results and economic value (Financial).
D. Measures and Initiatives: Once objectives have been agreed measures can be identified and constructed with the intention to support management's ability to monitor the organisation's progress towards achievement of its goals. Initiatives are special projects with a finite start and end date and are mapped to strategic objectives to give an indication of the projects or actions needed in order to realise the objectives.
8. Practical Experiences with 3rd Generation Balanced Scorecards
The first Balanced Scorecards to have included Destination Statements were designed during 1998/9. Examples of applications of this new approach are emerging. These experiences show that the third generation approach to Balanced Scorecard design and development does appear to have material benefits to organisations resulting from improved functionality as a strategic management tool, and as a result of its ability to support a more flexible and engaging approach to design and development within complex organisations.
During the 10 years since the advent of Balanced Scorecard many changes have been made to the physical design, utility and the design processes used to create the tool within organisations. This evolution of Balanced Scorecard, at least in terms of these three parameters, can be largely attributed to empirical evidence driven primarily by observed weaknesses in the design process rather than in the architecture of the original idea.
The alignment between developments in Balanced Scorecard principles and the theoretical aspects of control and management process are a positive indication that the more modern ideas about Balanced Scorecard design processes and structure are indeed 'better' than the original concept described by Kaplan and Norton, in so far as they are more likely to have a beneficial consequence for the organisation adopting the tool. However while more recent Balanced Scorecard designs are substantial improvements on original ideas, there is room for improvement.