For many organisations, the problem isn't a lack of metrics. It's a lack of clarity around which measures truly matter.

The reporting pack arrives on time. The dashboard is full. The scorecard reviewed. Yet when performance shifts, the same questions still surface: What's driving it? What should we do differently? And who owns the response?

That's the uncomfortable reality of performance measurement. An organisation can track hundreds of numbers and still lack a clear view of whether it's moving towards its strategy or just describing what's already happened.

In a separate article, we explored why good strategy fails. One of the central themes was that strategy only creates value when it changes what people do day to day. Ambition agreed in the boardroom has to be translated into choices, priorities and actions throughout the organisation.

That translation doesn't happen by accident.

Planning processes create alignment across different horizons. Well-designed performance measures provide the feedback loop. They show whether the choices being made today are improving the organisation's prospects tomorrow.

When designed properly, KPIs create focus, accountability and clarity. When designed poorly, they create noise, local optimisation and behaviours that can take the business further away from its objectives.

The purpose of a KPI

A KPI is not simply a number on a dashboard. Its purpose is to support a decision, provoke a useful conversation or signal that intervention is required.

Every KPI should therefore answer a simple question: what strategic objective does this help us achieve?

If that connection can't be clearly articulated, the measure may still be useful operational information, but it's not necessarily a key performance indicator.

Too often organisations inherit measures from previous leadership teams, established reporting packs, or whatever data happens to be readily available. The list grows, but the signal doesn't become clearer. Teams spend more time compiling and explaining performance than improving it.

The discipline isn't in measuring more. It's in identifying the few measures that genuinely matter, defining them properly and using them consistently.

Start with strategy, not data

One of the most common mistakes is to begin with the data available rather than the outcomes the business is trying to achieve.

It's easy to measure what's convenient. It's harder, and far more valuable, to measure what matters.

If the strategic objective is to improve customer service, sales alone won't tell you enough. You need on-shelf availability, forecast accuracy, supplier delivery performance and lost-sales data. If the goal is profitable growth, revenue needs to be considered alongside gross margin, price realisation, customer retention and return on capital.

The right measures depend on the choices embedded in the strategy. A standard scorecard can provide a useful starting point, but it can't replace that thinking.

The KPI framework should emerge from the strategy, not the other way around.

Build a line of sight from the long term to the now

Strategy is usually expressed over a multi-year horizon. Execution happens through decisions made this quarter, this month, this week and today.

A useful KPI architecture connects those horizons. Strategic outcomes sit at the top. The drivers that influence those outcomes sit beneath them. Operational measures then show whether the required activities and disciplines are actually happening.

That line of sight matters. It allows a team on the ground to understand how improving forecast accuracy, reducing changeover losses or increasing on-shelf availability, for example, contributes to a broader strategic outcome.

Without that connection, KPIs become isolated targets. With it, they become part of the organisation's execution system.

Balance leading and lagging indicators

Many scorecards are dominated by lagging indicators such as revenue, profit, market share and return on capital. These measures matter, but they describe outcomes that have already occurred.

By the time a lagging indicator deteriorates, some of the opportunity to influence it may already have passed.

Leading indicators give earlier visibility of the conditions likely to shape future performance. Pipeline quality may precede revenue. Forecast accuracy and supplier service may precede availability. Preventive maintenance compliance may precede equipment reliability.

The distinction is not always absolute. A measure can be a lagging indicator for one process and a leading indicator for another. What matters is understanding the cause-and-effect logic and avoiding a scorecard made up entirely of rear-view measures.

A balanced framework monitors both the result and the drivers of that result.

Measure the right thing, in the right way

A metric can look sensible and still be badly designed.

Definitions matter. So does the time window, level of aggregation, exclusions and source data. On-shelf availability calculated as an average of store percentages may produce a different answer from a demand-weighted calculation. Service measured by order lines may tell a different story from service measured by units or value.

Neither answer is automatically wrong, but the organisation must know what question the measure is intended to answer.

Every KPI should have a clear owner, formula, data source, frequency, target, tolerance and response when performance moves outside expectation. If different teams calculate the same KPI in different ways, the review will become a debate about the number rather than a decision about the business.

Beware unintended consequences

People respond to what's measured and rewarded. That makes KPIs powerful but also makes poor KPI design dangerous.

A sales growth target can encourage discounting and margin erosion. An inventory reduction target can damage availability. Purchase price variance can improve while total supply chain cost increases. Production utilisation can rise while excess inventory builds.

Every measure creates an incentive. Leaders should therefore ask: if we optimise this metric aggressively, what might we damage elsewhere?

Balanced measures help expose the trade-offs. They don't eliminate judgement, but they make the consequences of decisions more visible and reduce the risk of one function winning at the expense of the whole business.

Metrics should drive conversations and decisions

The value of a KPI doesn't lie in the reporting itself. It lies in the quality of the conversation and actions that follow.

A performance review should do more than explain movement versus last month, forecast or budget. It should distinguish signal from noise, identify root causes, test whether assumptions have changed and agree what will be done next.

That requires an environment where measures are used to learn and intervene, not simply to allocate blame. Accountability still matters, but accountability without diagnosis often creates defensive reporting rather than better performance.

The strongest reviews are forward-looking. They use the latest evidence to change future outcomes.

Final thoughts

Strategy defines where the organisation wants to go. Planning translates that ambition across time horizons. Performance measures provide the feedback loop that helps keep execution on course.

But measurement is not progress in itself.

A KPI should exist because it supports a strategic objective, informs a decision and influences the right behaviour. It should be clearly defined, balanced against unintended consequences and reviewed at a rhythm that allows action.

The goal is not to measure more. It's to measure what matters.

Because organisations don't execute strategy through dashboards. They execute it through thousands of decisions made every day. Good KPIs help ensure those decisions are pulling in the same direction.

A practical test for every KPI

  • Is it clearly linked to a strategic objective or critical business outcome?
  • Is the definition unambiguous and consistently calculated?
  • Does the team reviewing it have the ability to influence it?
  • Is it leading or lagging?
  • Could optimising it create an unintended consequence elsewhere?
  • Is the review frequency matched to the speed at which action can be taken?
  • Does movement in the measure trigger a decision, investigation or intervention?

If the answer to several of these questions is no, it may be a metric worth reporting, but it is unlikely to be a KPI worth managing.