The on-call rotation paid a bonus for fastest mean time to resolution. The metric was clean. The intent was right: reward the team that closes incidents fast.
After three months, the team started closing tickets instead of fixing them. Not all of them. Not blatantly. Just the ones ambiguous enough to call resolved. Just enough to protect the number without triggering a review.
Nobody told them to do this. Nobody told them not to. The incentive taught them. That’s what incentives do.
Incentives Don’t Have Beliefs
An incentive doesn’t know what you meant. It doesn’t read your culture deck or attend your all-hands or factor in your good intentions. It just produces behavior by making one outcome more rational than another, and then it waits to see what people do.
That’s the whole mechanism. Nothing more complicated is happening.
You tell the team that quality matters. The incentive rewards speed. The team produces speed. Not because they don’t care about quality. Because the incentive is the real message, and they know how to read it.
This is where moral injury comes from. Not burnout from overwork, though that’s real too. Moral injury is what happens when the behavior the incentive requires conflicts with what you believe is right, and you do it anyway because the math makes sense. Because everyone else is doing it. Because the alternative is to be the one person who doesn’t, and watch that choice penalize you while others advance.
The incentive didn’t corrupt you. It just revealed what the system actually wanted. You were paying attention. You responded rationally to a system that was rewarding the wrong thing. That’s not a character failure. It’s a design problem.
Two Scenes
The first one.
The quarterly business review tracked three metrics. Everyone knew two of them mattered. The third had been on the slide for three years because it was the kind of number that made the organization look serious about customers, and being serious about customers was a thing the organization believed it valued.
Nobody said so. But it got smaller each quarter. By year two, it lived in the appendix.
The third metric was customer complaint resolution time. Not complaint volume. Resolution time: how long it actually took to fix a problem once a customer raised it.
It moved to the appendix because it kept going up. Not because leadership didn’t care. Because there was no incentive attached to it. No bonus, no performance review line, no OKR. It was a number with no consequence. So it did what numbers always do when nobody is watching: it drifted in the direction of least resistance, quarter after quarter, until it was too uncomfortable to feature and too inconvenient to fix.
The second one.
The company ran perspective reviews. Employees scored their managers on five dimensions. Results went to HR. HR shared them with the manager.
Employees knew this from the first year. By year two, the scores for every manager had converged upward. Not because managers had gotten better at their jobs. Because employees had learned what the feedback actually cost, and they had made the rational calculation that honesty in a 360 that goes directly to your manager is not honesty at all. It’s documentation.
The perspective wasn’t measuring trust. It was measuring risk tolerance.
The incentive wasn’t “tell us the truth.” It was “tell us what you’re willing to sign your name to.”
What the Incentive Is Actually Measuring
Here is the pattern underneath both of those:
When you measure something without attaching a consequence, you are not measuring performance. You are measuring your own sincerity. You are recording a number that shows you care, while the system demonstrates that you don’t, and everyone who works for you can see the gap between the two.
When you measure something and attach a consequence, you get that thing. Not because people are cynical. Because they are rational. Because they understood the assignment.
The danger isn’t that people optimize for the incentive. That’s expected. That’s the point. The danger is when the incentive optimizes for the wrong thing: when speed beats accuracy, when ticket closure beats root cause, when the 360 score becomes a measure of political intelligence rather than managerial quality.
You built a feedback loop. It’s working exactly as designed. The design is the problem.
The One Lever
You cannot change everything at once. Restructuring incentive systems in bulk produces confusion: the organization doesn’t know what to optimize for, and behavior becomes erratic while people wait to see which metric you’re actually serious about this time.
But you can change one thing.
Pick the behavior that matters most and that your current incentive structure is actively working against. Not the behavior you wish you had, or the one that looks best in a strategy doc. The specific thing the system is punishing right now, today, in this team.
Then change the incentive. Not the value statement. Not the message at the all-hands. The actual measurement and the actual consequence attached to it.
If you measure speed, you get speed. If you measure closure, you get closure. If you measure nothing, you get whatever is easiest. And if you measure something and attach nothing to it, you are performing seriousness, not building accountability. You are adding a number to a slide so you don’t have to say you aren’t tracking it.
Change one incentive.
Or stop pretending you value the opposite.
This is part of a series on leadership, silence, and human systems. Previously: Permission to Tell the Truth. Next: Accountability Without Humiliation.
Resources & Further Reading
Foundational
Drive: The Surprising Truth About What Motivates Us by Daniel H. Pink (Riverhead Books, 2009) Why it’s relevant: Pink’s research on intrinsic vs. extrinsic motivation establishes why incentive design backfires when it attaches external rewards to work that requires judgment and creativity.
The Principal-Agent Problem — Economic concept, widely documented Why it’s relevant: The formal framework behind every misaligned incentive: when the person doing the work (agent) has different interests than the person who set the incentive (principal), the agent optimizes for their own outcome. Ticket closure beats root cause every time.
Practical
Physicians Aren’t ‘Burning Out.’ They’re Suffering from Moral Injury. by Simon G. Talbot and Wendy Dean, STAT, 2018 Why it’s relevant: Talbot and Dean named moral injury as a distinct condition: the damage done by repeatedly acting against your own values under institutional pressure. The concept originated in clinical settings but the mechanism is identical in corporate ones.