Strategy & competition
Balanced Scorecard
Translates vision and strategy into linked objectives and measures across four perspectives, financial, customer, internal process, and learning and growth, so an organisation steers by the drivers of future performance rather than by financial results alone.
Also known as BSC, Kaplan-Norton scorecard. First set out by Robert S. Kaplan and David P. Norton in 1992; the primary source is cited in full below.
Where this is contested
Kaplan and Norton's authorship of the framework is settled, but antecedents exist: Art Schneiderman's 1987 corporate scorecard at Analog Devices fed directly into the research behind the 1992 article, and the French tableau de bord tradition had used multi-dimensional performance dashboards for decades before.
- Format
- Structural model
- Level
- Corporate · Business unit
- Best for
- Plan execution · Allocate resources · Evaluate options
- Decision stage
- Plan · Execute · Review
- Difficulty
- Advanced
- Time to apply
- A day to draft in workshop; months to embed measures, targets and the review rhythm properly.
Plate · The model
The components
Vision and strategy
The centre of the scorecard, from which everything else derives. Objectives and measures in all four perspectives exist only to express and test the strategy; the scorecard's job, in Kaplan and Norton's phrase, is to translate strategy into action, never to substitute for it.
Signals of strength
Every measure traceable back to a strategic choice · Perspectives populated by derivation rather than brainstorm · The scorecard changing when the strategy changes
Financial
How the organisation must appear to its shareholders or funders for the strategy to count as a success. The financial perspective keeps the others honest: improvements in the other three that never convert into financial performance are, on the scorecard's own logic, a hypothesis failing its test.
Signals of strength
Revenue growth and mix · Operating margin and cost per unit of output · Return on capital employed · Cash generation and subsidy or funding dependence
Customer
How the organisation must appear to the customers it has chosen to serve: the value proposition, and whether customers are noticing it. Kaplan and Norton place it above the internal perspectives because customer outcomes are what internal excellence is for.
Signals of strength
Satisfaction and advocacy scores · Retention and repeat custom · Market or ridership share in target segments · On-time, first-time service delivery as customers experience it
Internal business processes
The processes the organisation must excel at to deliver the customer value proposition and the financial objectives, including processes it may not yet have. The discipline is selectivity: the scorecard names the few processes where excellence is strategic, rather than auditing all of them.
Signals of strength
Quality, defect and rework rates in the critical processes · Cycle times and asset utilisation · Cost and reliability of the core operating cycle · Innovation measures such as time to launch new services
Learning and growth
The people, systems and organisational capabilities that everything upstream depends on: called innovation and learning in the 1992 article and refined into learning and growth in the 1996 book. It is the perspective that funds the future, and the first to be raided when the financial perspective tightens.
Signals of strength
Staff retention, engagement and absence · Skills coverage against the strategy's demands · Information systems giving frontline staff what the strategy assumes they know · Climate for improvement, measured by suggestions raised and acted on
When it earns its keep
- Financial reporting says the organisation is healthy while customers, operations or staff say otherwise, and you need a measurement system that sees all four.
- A strategy exists on paper but is not moving behaviour, and you need to translate it into objectives, measures and targets people can act on at every level.
- You are managing long-horizon investments, in capability, systems or relationships, that financial measures will punish before they pay off.
- Different functions optimise their own metrics at each other's expense, and you need one linked picture of how the pieces are supposed to produce the strategy.
And when it doesn't
- There is no strategy to translate. The scorecard operationalises strategic choices already made; without them it degenerates into a KPI wallchart.
- The organisation is in crisis, where cash and survival dominate and a multi-perspective measurement build is a distraction until the bleeding stops.
- A small team needs lightweight goal-setting; a full scorecard architecture is heavy machinery for what a page of objectives could do.
- The environment shifts faster than the scorecard can be revised. Cause-and-effect chains built for last year's strategy will steer you precisely towards last year.
How to run it
Before starting, gather the inputs the analysis depends on:
- An articulated vision and strategy, specific enough that measurable objectives can be derived from it rather than invented alongside it.
- Senior leadership time and sponsorship; a scorecard delegated entirely to a measurement team measures what that team can reach, which is rarely the strategy.
- A candid hypothesis of cause and effect: which capabilities drive which processes, which processes drive which customer outcomes, and how those convert to financial results.
- Data, or a plan to get it, for the measures chosen, including the non-financial ones the organisation has probably never collected properly.
- A target-setting and review rhythm with owners for each objective and measure.
- 1
Clarify the vision and strategy
Force the strategy into statements sharp enough to measure. 'Delight customers' translates into nothing; 'win commuters from car travel on our urban corridors' tells you what to count. This step is where most scorecard projects quietly fail, because a vague strategy produces a vague scorecard with great efficiency.
- 2
Set objectives in each of the four perspectives
For financial, customer, internal process, and learning and growth in turn, ask what must be true for the strategy to succeed. Kaplan and Norton's discipline holds: a handful of objectives per perspective, each traceable to the strategy, rather than an inventory of everything the organisation could conceivably want.
- 3
Choose measures, targets and initiatives
Attach to each objective a small number of measures mixing outcome measures (lag indicators) with performance drivers (lead indicators), then set targets and name the initiatives expected to close the gap. Around 20 to 25 measures in total is the conventional ceiling; beyond that the scorecard stops directing attention and starts diluting it.
- 4
Map the cause-and-effect links
Draw the hypothesised chain from learning and growth through internal processes and customer outcomes to financial results, the strategy map Kaplan and Norton developed in their later work. Making the links explicit turns the scorecard into a testable argument about how the organisation creates value, and exposes objectives that connect to nothing.
- 5
Cascade and align
Translate the top-level scorecard into scorecards for divisions and teams that support, rather than photocopy, the objectives above them. Link budgets, initiatives and, carefully, incentives to scorecard targets, remembering that every measure attached to pay will be gamed by someone.
- 6
Review, test and adapt
Run the review rhythm on two loops: are we hitting the targets, and, more importantly, is the causal hypothesis holding? If improved training scores never move process quality, the map is wrong somewhere. Kaplan and Norton call this double-loop learning, and a scorecard that never revises its own links has stopped being one.
Reading the result
A one-page scorecard of objectives, measures, targets and initiatives across the four perspectives, ideally accompanied by a strategy map making the assumed cause-and-effect chain explicit, plus the review rhythm that keeps both honest.
- Read it as an argument rather than a dashboard: learning and growth enables process performance, which drives customer outcomes, which produce financial results. Then look for where the argument is failing.
- Watch the lead indicators for the future and the lag indicators for the past. A scorecard that is all lag is a rear-view mirror with extra rows.
- Persistent green in one perspective alongside persistent red in the next is the interesting finding: either the initiatives are wrong or the causal link is imaginary.
- Count the measures. If the scorecard has sprawled past roughly 25, it has stopped expressing choices.
A worked example
A regional bus operator builds a scorecard to arrest decline
A bus operator running 190 vehicles across a northern English region faces falling patronage, rising congestion, driver shortages and a combined authority weighing franchising. The board adopts a strategy of winning commuters back from car travel on five urban corridors through reliability and simple fares, and builds a balanced scorecard to drive it through the business.
- Vision and strategy
- The strategy is stated tightly: be the obvious choice for urban commuters on the five key corridors by 2028, measured by corridor patronage growth and mode shift, and demonstrate a partnership case strong enough to shape the franchising decision. Every objective below is derived from, and only from, this.
- Financial
- Objectives: grow commercial farebox revenue 4 per cent a year on the five corridors, hold cost per operated mile flat in real terms, and cut the loss-making tail of the network. Measures: corridor revenue, cost per mile, subsidy dependence by route. Lag indicators all, which is precisely why the board resists managing by them alone.
- Customer
- Objectives: turn reliability into the reason to ride, and make fares simple enough to explain at a bus stop. Measures: excess waiting time on frequent corridors, punctuality on timetabled routes, corridor patronage, satisfaction and a mode-shift survey question ('could this journey have been by car?'). Targets are set against the best comparable operator, taking benchmarks from published Transport Focus survey data.
- Internal business processes
- The processes that make reliability: engineering (measure: miles between breakdowns, morning peak vehicle availability above 92 per cent), lost mileage below 0.5 per cent, boarding speed via contactless and mobile ticketing (average dwell time per stop), and schedule quality (timetables rebuilt from actual corridor running times rather than aspiration). Each has a named owner on the executive.
- Learning and growth
- The constraint underneath everything: drivers and fitters. Objectives: cut driver turnover from 24 to 15 per cent, close the fitter vacancy gap, and give control-room staff live corridor data. Measures: turnover, time to fill vacancies, training days per driver, telematics adoption, and an annual climate survey. The scorecard makes explicit that lost mileage, the process measure, is causally downstream of driver turnover, so the retention line is defended when budgets tighten.
The read. The scorecard's value showed within two quarters, in a causal link failing its test. Boarding-speed investment moved dwell times but corridor punctuality barely shifted; the binding constraint was roadworks and congestion outside the operator's control. That finding redirected effort towards bus-priority lobbying with the combined authority, an objective no financial report would ever have surfaced, and the retention measures gave early warning that one depot's turnover was undoing the engineering gains. The scorecard did what Kaplan and Norton claim for it: it made the strategy's assumptions visible enough to be wrong in public.
Pitfalls
- Building the scorecard without a strategy and hoping measurement will secrete one. It will not; you get a balanced list of things that seemed important in the workshop.
- Measure sprawl. Every stakeholder adds a metric, the scorecard passes 40 measures, and attention returns to the two numbers the board always looked at anyway.
- Choosing measures because data exists rather than because the strategy needs them, which biases the scorecard towards the already-measured past.
- Cascading by photocopy, so every team carries corporate objectives it cannot influence, instead of objectives that feed the level above.
- Wiring incentives to scorecard targets too early, before measures are trusted, which converts a learning tool into a gaming arena.
- Treating the causal links as facts rather than hypotheses, and never revisiting them. The map is the scorecard's boldest claim and should be its most-audited part.
What the critics say
Norreklit's analysis argued that the scorecard's central claim, a causal chain from learning and growth through processes and customers to financial results, is not causality at all: the relationships are logical and finance-derived rather than empirically established, the model has no time dimension in which cause could precede effect, and the top-down control loop makes it doubtful as a strategic management system.
Norreklit, H. (2000) 'The balance on the balanced scorecard: a critical analysis of some of its assumptions', Management Accounting Research, 11(1), pp. 65-88.
Ittner and Larcker found that most companies adopting non-financial measurement failed to validate the assumed links between non-financial measures and financial outcomes; fewer than a quarter built and tested causal models, so scorecards routinely track drivers that drive nothing.
Ittner, C. D. and Larcker, D. F. (2003) 'Coming Up Short on Nonfinancial Performance Measurement', Harvard Business Review, 81(11), November 2003.
In a follow-up study, Norreklit argued that the scorecard's diffusion owed more to persuasive rhetoric than demonstrated results, analysing Kaplan and Norton's texts as promotional argument, and later survey evidence on scorecard success rates has remained mixed, with implementation failures commonly attributed to the tool's demands on strategy clarity and data.
Norreklit, H. (2003) 'The Balanced Scorecard: what is the score? A rhetorical analysis of the Balanced Scorecard', Accounting, Organizations and Society, 28(6), pp. 591-619.
Jensen's objection is that the scorecard is the managerial equivalent of stakeholder theory: it hands managers a couple of dozen measures with no weights and no way to trade one against another, a scorecard without a score. Purposeful behaviour requires a single-valued objective, and a system that cannot say whether overall performance improved leaves managers effectively unaccountable.
Jensen, M. C. (2001) 'Value Maximization, Stakeholder Theory, and the Corporate Objective Function', Journal of Applied Corporate Finance, 14(3), pp. 8-21.
Voelpel and colleagues argued the four-perspective architecture becomes a measurement straitjacket in innovation-driven contexts: it is static, focused inside the firm, and blind to the ecosystem and knowledge flows that increasingly generate value, so faithful scorecard discipline can actively hinder the innovation it is meant to steer. Kaplan and Norton published a rebuttal in the same journal, but the exchange marked out real limits of the design.
Voelpel, S. C., Leibold, M. and Eckhoff, R. A. (2006) 'The tyranny of the Balanced Scorecard in the innovation economy', Journal of Intellectual Capital, 7(1), pp. 43-60.
Sources and further reading
- Kaplan, R. S. and Norton, D. P. (1992) 'The Balanced Scorecard: Measures That Drive Performance', Harvard Business Review, 70(1), January-February 1992. ↗
- Kaplan, R. S. and Norton, D. P. (1996) The Balanced Scorecard: Translating Strategy into Action. Boston: Harvard Business School Press.
- Kaplan, R. S. and Norton, D. P. (1996) 'Using the Balanced Scorecard as a Strategic Management System', Harvard Business Review, 74(1), January-February 1996. ↗
- Kaplan, R. S. and Norton, D. P. (2004) Strategy Maps: Converting Intangible Assets into Tangible Outcomes. Boston: Harvard Business School Press.