When Metrics Start Driving the Wrong Behavior

If only metrics came with a wrong way sign.

Metrics don't lie. But they don't always tell the truth either.

The dashboard is green. The team is hitting all their targets. Somewhere underneath all of it though, the behavior the metrics were supposed to encourage has quietly been replaced by the behavior required to move the number.

The Metric That Outlived Its Purpose

Every metric made sense when it was chosen. Velocity tracked how much the team could ship in a sprint. NPS captured how customers felt about the product. DAU showed whether people were coming back. At the right stage, with the right context, each of those numbers told a meaningful story.

The problem is that metrics have a shelf life. What matters in the first year of a product is not always what matters in the third. A team optimizing for sprint velocity at a stage where strategic focus matters more than output is not solving the right problem. A product measuring NPS across a user base that has fundamentally changed since the survey was designed is not getting signal. It is getting noise that looks like signal.

Nobody sets out to measure the wrong thing. It happens when the business evolves and the measurement system does not.

When the Number Becomes the Goal

There is a predictable pattern that follows when a metric stays in place longer than it should. The team stops asking what the number is supposed to represent and starts asking how to move it. That distinction matters more than it seems.

A team focused on what a metric represents will change their behavior when the product is not delivering real value. A team focused on moving the number will find ways to move it regardless of whether real value is being delivered. That’s not a character failure, it’s a rational response to an environment where the metric is what gets measured, reported, and rewarded.

It starts small. Sprint velocity climbs because the team has learned how to estimate in their favor. The work isn’t getting done faster, the points are just getting bigger. NPS improves because the survey is only sent to customers who recently had a positive interaction, quietly filtering out the signal that would have been most useful. DAU holds steady because a notification strategy is pulling people back into an app they are not genuinely finding valuable. Each of those adjustments feels reasonable in isolation. Taken together they build a measurement system that is technically accurate and practically useless.

The number moves. The business does not. And because the dashboard still looks healthy, nobody asks why.

The Leadership Responsibility

Metric distortion is rarely a team problem. It is a leadership problem. The team optimizes for what leadership measures. If leadership is measuring the wrong things, or measuring the right things past their useful life, the team will follow.

The most effective product leaders treat their measurement systems as living decisions, not permanent infrastructure. They ask regularly whether the metrics they are tracking still reflect the outcomes they actually care about. They create space for the team to raise concerns when a number starts feeling disconnected from reality. And they are willing to retire a metric that has stopped being honest, even when it means losing a number that was easy to report.

Signal integrity is not a data problem. It is a discipline problem. And discipline starts at the top.

Closing Thought

The goal was never supposed to be the number. It was what the number was supposed to represent. When those two things drift apart and nobody says anything, the metric stops being a tool. It becomes a story the team tells to avoid a harder conversation.