The Perverse Incentives Hiding in Your KPIs
Every metric you enforce is a contract with human ingenuity. Most contracts get gamed exactly as written.
A perverse incentive is a reward that produces the opposite of its intent. KPIs create them whenever the measure is easier to move than the outcome it stands for — the fix is to pair every target metric with a guardrail metric that makes gaming visible.
Goodhart's Law is usually quoted and rarely operationalized: when a measure becomes a target, it ceases to be a good measure. In practice this means every KPI you enforce is a standing offer to your smartest employees to find the cheapest path to the number.
They will accept the offer. That is not cynicism; it is competence pointed at the target you supplied.
Four signatures of a KPI going perverse
Perverse incentives announce themselves before they detonate. The tells are consistent across industries.
- —The number improves while customer or employee sentiment declines.
- —Improvement clusters at period boundaries — the last week of the quarter does the work.
- —Nobody can explain the causal path from the metric to the outcome.
- —The people closest to the metric stop volunteering information about it.
The guardrail pattern
The durable fix is structural. For every target metric, name a guardrail metric that would deteriorate if the target were gamed, and report them on the same page, to the same audience, at the same cadence.
Ticket closure rate pairs with reopen rate. Sales velocity pairs with 90-day churn. Time-to-hire pairs with first-year regretted attrition. Cost per unit pairs with defect escape rate. The pairing does more work than any policy memo, because it changes what the room can see.
You cannot police your way out of a bad metric. You can only pair it with the number it would have to damage.
When to retire a metric entirely
Some metrics are unrescuable. If the measure is fully under the control of the person being measured, has no natural guardrail, and carries career consequences, retire it. Replace it with a sampled qualitative review or an outcome measured by someone who does not benefit from the result.
Retiring a metric is cheaper than the twelve months of distorted behavior it is otherwise buying you.
Frequently asked
- What is an example of a perverse incentive at work?
- Paying support teams on tickets closed produces fast closures and repeat contacts. The organization pays twice for the same unsolved problem while the dashboard shows improvement.
- How do you test a new KPI before rolling it out?
- Ask three high performers how they would hit the number if their bonus depended on it and they were unwilling to do the hard version. Their answers are your future incident report.
Most performance problems are payoff problems.
The Incentives Lab reconstructs what your organization actually rewards from evidence people cannot manage — promotions, calendars, budget shifts, attrition, and what happens after a bad quarter. Start with the free diagnostic, or talk to us about an audit.
New essay every Wednesday. Subscribe via RSS.
Aaron Bare
Aaron Bare is a strategist, Wall Street Journal-bestselling author, and the founder of The Incentives Lab. He writes and advises on incentive design inside organizations — why culture is the residue of what a company rewards, how KPIs quietly go perverse, and how AI systems inherit the incentives their designers set.
More from Aaron BareGo deeper
More essays
Anatomy of a Two-Week Incentive Sprint
Ten working days from a behavior nobody can explain to a structural change with a counter-metric attached.
AI Agents Inherit Your Incentives
An agent optimizes the objective you wrote down — which is usually the one your people were quietly ignoring for good reason.
Goodhart's Law in the Real World
When a measure becomes a target, it stops being a good measure. Three field cases and the paired-metric fix.