Incentives Drive Behavior
What is the fastest way to get exactly what you asked for and none of what you wanted?
The idea
People respond to measured and rewarded outcomes, including unintended ones.
In the real world
Paying only for speed can reduce quality.
Going deeper
People optimise what is measured and rewarded, including along dimensions you did not intend. This is not cynicism; it is what a clear target does.
Tickets closed per hour is a complete specification of what is wanted, and the fastest route to it is closing quickly rather than resolving. Quality was not in the specification, so it was not protected, and repeat contacts rise. The incentive worked perfectly. The design error is upstream: any single metric will be met by whatever route is cheapest, so the question is always what the cheapest route to this number would look like.
Where it stops applying
Adding metrics to cover every dimension creates its own problems — conflicting targets, gaming across measures, and effort spent on reporting. Fewer measures with human judgement often beats a comprehensive scorecard.
Why it matters
People optimise for what is measured and rewarded, including in directions you did not intend.
Try this today
Check what behavior a metric actually rewards.
Test yourself
A support team is paid on tickets closed per hour. Volume rises 40% and repeat contacts rise with it. What did the incentive buy?
Show the answer
Closures, precisely as specified. Quality was not measured, so it was not protected, and the fastest route to the reward is to close quickly rather than resolve. The incentive worked; the target was wrong.
Learn this in the feed Answering from memory, then again days later, is what makes it stick.