Key performance indicators are meant to create clarity. They tell people what matters, where to focus and whether the business is making progress.
But many businesses keep adding KPIs until clarity disappears altogether.
One measure becomes five. Five become fifteen. Before long, managers are looking at dashboards filled with revenue, margin, utilisation, conversion rates, customer satisfaction, response times, error rates, employee engagement, overdue debtors and dozens of other numbers.
Each measure may be useful in isolation. Mixed together, however, they make soup.
The problem is not that the numbers are wrong. The problem is that nobody knows which ones matter most.
If everything is a priority, nothing is a priority. Employees cannot make sensible trade-offs because they have not been told what the business values most. Should the salesperson protect margin or close the deal? Should the service team respond quickly or spend more time resolving each problem? Should the operations manager maximise utilisation or preserve capacity for urgent work?
Too many KPIs often give managers the illusion of control. Measuring more feels rigorous, but it can produce the opposite result. People spend more time collecting data, explaining variances and preparing reports, while paying less attention to the few outcomes that actually drive performance.
It also encourages gaming. When employees are assessed against a long list of measures, they learn which numbers are easiest to influence and which ones can be safely ignored. The dashboard may look healthy while the underlying business deteriorates.
Good KPI design begins with strategy. Ask: what are the two or three things this team must achieve for the business to succeed? Those outcomes should become its primary KPIs.
A sales team might focus on gross profit won, new customers acquired and recurring revenue retained. An operations team might focus on work completed on time, gross margin and rework. A finance team might focus on cash collection, reporting accuracy and forecast reliability.
Other measures can still be monitored. Not every useful number needs to be a KPI. Some are diagnostic measures: they help explain why a result changed. Others are guardrails: they ensure that improving one outcome does not cause unacceptable damage elsewhere.
The distinction matters. A KPI directs attention and affects behaviour. A diagnostic measure supports investigation. A guardrail establishes a limit. Calling all three “KPIs” creates confusion.
A useful test is to ask each employee to name the three results that define success in their role. If their answer differs from their manager’s—or if they need to consult a spreadsheet—the performance system is not creating clarity.
Reducing the number of KPIs does require judgement. It forces leaders to make choices, acknowledge trade-offs and state plainly what matters most. That can feel uncomfortable, but it is the essence of management.
Your business does not need more numbers. It needs a small number of meaningful measures, clearly defined and consistently discussed.
Otherwise, you are not managing performance. You are just stirring the soup.
Photo by Jametlene Reskp on Unsplash