The dashboard is green. All KPIs are within target. The presentation for the board is ready: We are delivering. Six months later, disillusionment sets in: client satisfaction has plummeted, key employees have resigned, market share is declining. But the numbers were good.
What you measure drives behavior. Those who measure the wrong things optimize the wrong things and are surprised by the results. A dashboard full of green lights can provide deceptive security, while the truly important things erode unmeasured.
Are you familiar with Watermelon KPIs? Green on the outside, red on the inside. The SLA is met, but the client is furious. The project light is green, but the team is exhausted. The numbers are correct, but reality is not.
A CFO I advised was proud of his KPI dashboard: 42 metrics, all automated, all within target. When I asked him which three of these metrics reflected his strategic priorities, there was silence. “Actually, none,” he said after a pause. “Most measure what we’ve always measured.” This is a common pattern: lots of data, little insight.
The British economist Charles Goodhart formulated a principle that every manager should know: When a measure becomes a target, it ceases to be a good measure. As soon as you make a metric a target, people begin to optimize that metric, often at the expense of what the metric was actually intended to measure.
Call center employees, measured by the number of calls handled, conduct short conversations with unresolved issues. Sales representatives, measured by revenue instead of contribution margin, initiate discount battles that destroy margins. The most famous example is the Cobra Effect: British colonial rulers in India offered a bounty for dead cobras. Indians began breeding cobras to collect the bounty. When the program was stopped, the bred cobras were released, and the plague was worse than before.
The pattern is always the same: the metric is optimized, the actual goal is missed. And the side effects are often worse than the original problem. Quarterly targets lead to short-term thinking that harms long-term investments. Individual KPIs create silo thinking instead of collaboration. And where incentives are strong enough, people get creative with gaming: numbers are embellished, definitions stretched, reports doctored.
“Much” is not the same as “good.” More meetings do not mean more results. More projects do not mean more progress.
Most KPI systems suffer from three interconnected problems. The first is confusing activity with impact: activity metrics show what is being done (number of trainings, client contacts, projects), while impact metrics show what is achieved (competence development, client satisfaction, strategic progress). The temptation to measure activity is great; it is visible and quantifiable. But activity without impact is busywork, not progress.
| Measure Activity | Measure Impact |
|---|---|
| Number of trainings | Competence development |
| Number of client contacts | Client satisfaction |
| Number of projects | Strategic progress |
| Number of applications | Quality of hires |
The second problem is the tyranny of easily measurable things. Sales calls, processing times, attendance, and costs are easy to track. The quality of client relationships, creativity, team cohesion, and cultural development are not. If you only measure what’s easy, you optimize for what’s easy and neglect what’s difficult. But what’s difficult is often what makes the long-term difference. A warning sign: If your most important strategic goals do not appear in your KPIs, something is wrong.
The third problem: Those who only observe lagging indicators are driving by looking in the rearview mirror. Employee turnover tells you that people have left. Employee engagement tells you that they will leave before it happens. Combine both: lagging indicators to validate if you are on the right track, and leading indicators to intervene early.
A word on NPS: The Net Promoter Score is the holy grail of client satisfaction in many companies. But NPS alone is a lagging indicator; it shows what was, not why. The true intelligence lies in the free-text comments. Those who only look at the number and do not read the comments measure without understanding.
In addition to quantitative KPIs, there are qualitative signals that are often more important: What do employees say to each other when no manager is present? Not just the turnover rate counts, but who leaves; if your best people leave and the mediocre ones stay, you have a problem that no number shows. What is not said in meetings? Which topics are avoided? The gaps are often more revealing than what is said. And client behavior beyond the survey: What do clients do, not what do they say? Do they buy again? Do they recommend others?
A division head told me: “Our engagement score was 78%, everyone was happy. Then, within three months, four top performers resigned. The survey had measured that employees were not dissatisfied. It had not measured that the best among them were no longer growing. That wasn’t in any dashboard, but it was in the faces of my best people.”
The best managers do not rely solely on dashboards. They keep their eyes and ears open for signals that no number captures.
Less is more. Three levers make the difference.
Start with strategy and limit yourself. The question is not what you can measure, but what you need to know to understand if your strategy is working. Three to five central KPIs are better than thirty. If everything is important, nothing is important. Every additional KPI dilutes attention. For each important area, combine at least one leading indicator with one lagging indicator and balance different perspectives: finance, clients, employees, processes. The Balanced Scorecard had this fundamental idea right.
Check for side effects. Before introducing a KPI, ask: What behavior could this KPI generate that we do not want? How could it be manipulated? What important things might be neglected because they are not measured?
Review regularly. KPIs that were once meaningful can become obsolete. Strategies change, environments change. What you measure should evolve with them. If you find that a KPI has been green for months without anyone looking at it, you probably don’t need it.
Take ten minutes and review your current KPI set using three questions:
First: Which of your KPIs measure activity and which measure impact? If more than half measure activity, you have a problem.
Second: Which strategically important topics do not appear in your KPIs, and what are the consequences of these blind spots?
Third: When was the last time you eliminated a KPI because it was no longer relevant?
If you cannot answer the third question, it is time for an inventory.
Dashboards provide security. Numbers feel objective. Green lights are reassuring. But this security can be deceptive. The biggest wrong decisions do not arise from a lack of data. They arise from incorrect data or from the illusion that the measured data fully reflects reality.
The best managers use KPIs as a tool, not as truth. They know that every metric is a reduction of reality. They supplement numbers with observation, conversations, and judgment. And they have the courage to act even when the dashboard is green, but their gut feeling is red.
Open your dashboard tomorrow morning and ask yourself for each individual metric: If this number were green, would I feel secure? And for which one would that feeling be wrong?
Quantifying Intangibles – Not everything that matters can be captured in numbers. How to make non-monetary value visible.
The strategy is set. But it hasn’t landed. – KPIs only reflect the strategy if the strategy is clear. Otherwise, you’re measuring the wrong things.
All Insights can be found in the overview.
Proven tools and models for self-application are available under Solutions.
If you want to take these thoughts further for your company, a no-obligation initial conversation is worthwhile.
