A performance system creates value when it changes decisions and behaviour while there is still time to improve the outcome.
Many organisations treat performance measurement as an act of observation. Data is collected, results are scored, dashboards are updated, and leaders review what has already happened. The system may be technically sophisticated and still arrive too late to influence the result.
That is the strategic problem.
A useful performance system does more than classify success or failure. It clarifies what good looks like, directs attention towards the right outcomes, provides evidence early enough to change course, and makes judgement more consistent. Its purpose is not merely to record performance. Its purpose is to help the organisation perform.
This distinction matters in project portfolios, transformation programs, operational improvement, capability development and executive performance management. Once a target becomes a score, people adapt their behaviour to it. If the measure is poorly designed, they can become better at satisfying the metric while becoming worse at producing the outcome the metric was intended to represent.
The executive question is therefore not simply, "What should we measure?" It is, "What behaviour will this measurement system create before the result is final?"
The Strategic Context
The supplied assessment material offers a useful design analogy because it treats measurement as part of the operating system rather than as an administrative afterthought.
Its logic is clear. Expectations should be explicit. Measures should be aligned with intended outcomes. Different forms of assessment should serve different purposes. Feedback should arrive in time to influence subsequent work. Judgements should be valid, reliable and moderated. Where a single aggregate result may be misleading, the underlying components need to be understood.
Those are not education-specific ideas. They are principles of performance-system design.
In a business context, the equivalent system may include portfolio scorecards, program health reviews, operational KPIs, quality measures, incentive arrangements, executive objectives or transformation benefits. In every case, leaders are creating a structure that tells people what matters.
That structure is never neutral.
If project managers are rewarded mainly for schedule adherence, they may resist necessary scope changes that would improve customer value. If a service team is measured primarily on call duration, shorter conversations can become more important than resolving the customer's problem. If a transformation office reports the number of milestones completed, activity can appear healthy while benefits remain uncertain.
The measurement architecture shapes the management system around it.
What Leaders Commonly Misread
The first common mistake is to assume that more measurement creates more control.
It can create more data without creating better decisions.
A metric is useful only when it improves the quality or timing of a decision. Measures that are easy to collect often survive because they are convenient, not because they are strategically important. This produces a familiar pattern: a large dashboard, repeated reporting cycles and limited change in managerial behaviour.
The second mistake is to confuse a final result with a management signal.
A final result answers a classification question: did the outcome meet the required standard? A management signal answers a different question: what should be changed now?
The distinction is similar to the difference between formative and summative assessment in the supplied material. Formative information exists to improve performance before the final judgement. Summative information records the level ultimately achieved.
Organisations frequently overinvest in the second and underinvest in the first.
The third mistake is to treat an aggregate score as if it were the system itself.
One of the supplied forum examples shows why this can be dangerous. The overall grade could initially appear poor because not all component contributions had yet been assessed. The interim signal was technically correct within the system, but potentially misleading if interpreted as a complete view of performance.
Executive dashboards can behave in the same way.
A single RAG status can conceal conflicting evidence. A portfolio may be "green" because schedule and budget are controlled while benefits are deteriorating. A factory's productivity may be improving while quality losses are rising. A transformation program may report high completion while user adoption remains weak.
When a composite measure is used, leaders need to understand what sits underneath it.
Reframing the Issue
Performance management should be viewed as a behavioural and decision architecture.
Its role is to connect five things:
- the outcome the organisation wants;
- the criteria that define acceptable performance;
- the evidence that shows what is happening;
- the feedback that influences behaviour;
- the final judgement that determines whether the outcome was achieved.
Weak systems jump directly from activity to score.
Strong systems make the logic visible.
This changes the executive conversation. Instead of asking whether the KPI is being achieved, leaders ask whether the KPI is producing the behaviour, trade-offs and decisions the strategy requires.
That is a higher standard.
A strategically sound performance system should therefore satisfy two tests.
The first is an outcome test: does the measure represent something that genuinely matters?
The second is a behaviour test: does the measure encourage the people inside the system to act in ways that improve that outcome?
A metric can pass the first test and fail the second.
For example, on-time delivery is strategically important. But if the organisation penalises any schedule movement regardless of cause, teams may hide emerging problems, defer necessary quality work or resist changes that would increase long-term value.
The measure is valid. The incentive system around it is not.
Strategic Analysis
Clear criteria reduce avoidable ambiguity
The supplied material places strong emphasis on making purpose, requirements, standards and criteria explicit.
This has a direct organisational equivalent.
When performance expectations remain vague, people optimise against local interpretations. Different managers make different judgements. Teams spend time negotiating what "good" means after the work has already been done.
That creates both inefficiency and distrust.
Clear criteria do not require rigid bureaucracy. They require leaders to distinguish between what is genuinely non-negotiable and what still requires judgement.
A capital proposal, for example, might have explicit criteria for strategic alignment, expected benefit, risk exposure, affordability and organisational capacity. The final investment decision still requires executive judgement, but the decision is made against a common frame.
The purpose of criteria is not to eliminate judgement. It is to make judgement more disciplined.
Feedback has value only while action remains possible
The assessment material treats timely feedback as part of the learning process. In enterprise systems, the same principle applies.
Feedback that arrives after the opportunity to act has limited control value.
A quarterly review may be useful for governance but insufficient for a fast-moving operational constraint. A monthly transformation status may be too slow if adoption is falling weekly. A project post-implementation review may generate useful learning, but it cannot recover benefits that were lost because warning signals were ignored during delivery.
This suggests a practical distinction between control cadence and reporting cadence.
Reporting cadence answers when information is formally consolidated.
Control cadence answers how frequently evidence must be reviewed to keep the system within an acceptable range.
The two should not automatically be the same.
Quality should not be replaced by volume
The supplied marking guide gives most value to the quality and depth of the initial contribution, with additional value available where participation genuinely adds depth. It does not treat more words or more posts as inherently better.
That is an important design principle for organisations.
Many systems reward activity because activity is easy to count.
Number of meetings held. Reports issued. Ideas submitted. Projects started. Training hours completed. Risks logged. Automation initiatives launched.
These measures may have diagnostic value, but they are rarely outcomes.
A transformation office that celebrates the number of initiatives launched can create portfolio congestion. A sales organisation that rewards call volume may encourage low-quality customer contact. A maintenance function that measures work orders closed may unintentionally favour quick closure over elimination of recurring failure.
Volume is often an input measure masquerading as a performance measure.
The strategic question is whether the activity produces the capability or outcome the organisation actually needs.
Moderation matters wherever judgement is subjective
The assessment policy also emphasises moderation, reliability and consistent judgement.
Businesses often underestimate this problem.
Any system involving human judgement can drift. Different investment committees may apply the same criteria differently. Different plants may classify the same defect differently. Different program sponsors may use different thresholds for escalation.
Without calibration, the organisation appears to have one standard while operating with several.
Moderation can take many forms: peer review, second-level challenge, cross-business calibration, common decision criteria, independent assurance or periodic benchmarking.
The objective is not perfect uniformity. It is to reduce unexplained variation in judgement.
Decision Framework
Leaders can test a performance system using six questions.
| Test | Executive question | Failure mode |
|---|---|---|
| Outcome | Does this measure represent something strategically important? | Teams optimise a proxy rather than the real result |
| Behaviour | What behaviour will people adopt to improve the measure? | Gaming, concealment or local optimisation |
| Timing | Does feedback arrive while corrective action is still possible? | Management learns after value has already been lost |
| Evidence | Can the result be traced to credible underlying information? | Aggregate scores create false confidence |
| Calibration | Would different leaders reach broadly consistent judgements? | Standards vary by person, function or business unit |
| Adaptation | Can the measure be changed when strategy or system conditions change? | The organisation continues managing yesterday's priorities |
A useful design discipline is to classify every important measure as one of four types:
Outcome measures show whether value has been created.
Leading indicators show whether the system is moving towards or away from the desired outcome.
Control measures show whether critical operating conditions remain within tolerance.
Learning measures show whether capability, behaviour or assumptions are changing.
This prevents one metric from being forced to do several jobs badly.
From Strategy to Execution
Immediate action should begin with the measures already used in executive, portfolio and operational reviews.
Select the ten most influential measures and ask what decisions each one is intended to change. If no clear decision exists, the measure may be reporting noise.
Next, examine the incentives and behaviours surrounding each measure. Look particularly for situations where teams can improve the score without improving the outcome.
In the medium term, redesign review forums so they distinguish between final outcomes and early management signals. Leaders should be able to see both the headline result and the evidence beneath it.
Capability also needs to be built around judgement. Managers require common criteria, clearer thresholds and periodic calibration. This is especially important in areas such as investment selection, risk escalation, project health and capability assessment.
Longer term, the organisation should treat the performance architecture as part of strategy execution itself. When strategic priorities change, measures and review structures should change with them.
A business cannot credibly claim that customer retention is now a strategic priority while continuing to manage frontline teams only on transaction speed.
The operating system must move with the strategy.
Signals to Monitor
Leaders should become concerned when:
- teams can improve a KPI without improving the underlying outcome;
- managers spend more time explaining measures than deciding what to do;
- a single aggregate score repeatedly conflicts with operational reality;
- feedback arrives after the next decision point has passed;
- different business units interpret the same performance standard differently;
- people avoid raising bad news because the measurement system punishes transparency;
- activity measures keep increasing while strategic benefits remain flat;
- measures remain unchanged even after strategic priorities have materially shifted.
These are signs that the organisation is measuring performance without effectively managing it.
Questions for the Leadership Team
- Which of our most visible KPIs can be improved without creating additional enterprise value?
- Where are we using final-result measures when we actually need earlier management signals?
- What behaviour are our current scorecards and incentives encouraging people to adopt?
- Which aggregate measures are hiding conflicting evidence underneath?
- Where does managerial judgement vary because our criteria are insufficiently clear or calibrated?
- Which measures would we stop collecting if every metric had to justify a specific decision?
- If our strategy changed tomorrow, which performance measures would need to change with it?
Closing Perspective
The most dangerous performance system is not one with no data. It is one that produces confident numbers while directing attention towards the wrong behaviour.
Leaders should expect measurement to do more than describe the past. It should help the organisation recognise problems earlier, make better trade-offs, improve performance before outcomes become fixed and learn from the difference between intention and result.
A score tells you what happened.
A well-designed performance system helps determine what happens next.
About EraNorth Insights
EraNorth Insights publishes practical analysis on strategy, projects, operations, transformation and decision intelligence for professional and organisational use. About EraNorth.