A company wants better customer support.
Leadership chooses a clean metric: average handle time. Shorter calls mean faster service, lower cost, and happier customers. The dashboard turns green. Managers coach to the number. Agents learn which calls threaten the target and how to move them along.
Then the same customers call back.
The first call got shorter because the problem was not fully solved. The metric improved while the customer experience got worse. The incentive did exactly what it was asked to do. It narrowed attention to the measured behavior and pushed everything else into the background.
This is what incentive quotes usually skip. They make incentives sound like levers. Pull the right one, get the right behavior. Inside real organizations, incentives become maps of what can be ignored.
”Show Me the Incentive and I’ll Show You the Outcome”
The quote works best when the outcome is simple and the measurement is hard to fake.
Pay a bounty for rats and people bring rats. Pay per unit and workers produce more units. Reward sales volume and sales teams push harder for volume.
The trouble begins when the measured target is only a proxy for the real goal.
A rat bounty can produce rat farming. Lines of code can produce bloated code. Sales volume can produce bad-fit customers, discounting games, and revenue that support teams spend the next year cleaning up.
The incentive does predict behavior. It predicts the behavior that maximizes the reward, not the behavior the designer had in mind.
The gap between proxy and goal is where the system gets gamed. People do not need to be cynical for this to happen. They only need to be evaluated on the proxy often enough to learn that the proxy is what counts.
”What Gets Measured Gets Managed”
Measurement creates attention. It also creates target fixation.
A hospital tracks discharge timing to improve throughput. Staff learn how to move patients through the measured gate. If the measure is too crude, patients may leave earlier than they should and return later in worse condition.
A software team tracks velocity to understand delivery capacity. The number becomes part of performance review. Story points inflate. Hardening work gets postponed. Complex problems are split into pieces that look productive without reducing system risk.
A recruiting team tracks time-to-fill. Roles close faster. Candidate quality, onboarding fit, and manager calibration become secondary unless they are measured with equal force.
The metric becomes the work’s public face. People manage the face.
That does not make measurement useless. It means measurement needs suspicion built into it. Every metric asks people to simplify a complicated reality. Once consequences attach, the simplification becomes attractive to optimize.
”You Get What You Incentivize”
You also lose what you stop protecting.
Before a formal incentive arrives, people often balance several pressures at once: quality, reputation, craft, customer need, future maintenance, professional norms, team trust. The judgment is messy, but it contains more of the work.
A strong incentive narrows the field.
If teachers are judged by test scores, teaching shifts toward test performance. If researchers are judged by citation count, topic choice shifts toward publishable and citable work. If engineers are promoted for feature launches, maintenance becomes charity.
The new behavior may look more focused. It may even improve the chosen number. But the system has displaced unmeasured obligations.
The quote is true in a colder way than intended. You get what you incentivize, and you train people to treat everything outside the incentive as optional.
”Incentives Are Superpowers”
Incentives can move behavior quickly because they attach consequence to repetition.
That speed is why they are dangerous.
A value statement asks people to care. An incentive system tells them what will happen if they care about the wrong thing. The incentive wins during tradeoffs because it affects bonus, promotion, status, workload, and safety.
A company says collaboration matters. Then it ranks teams against one another for budget. A manager says quality matters. Then promotion goes to the person who shipped the most visible work. A leadership team says long-term trust matters. Then quarterly targets decide compensation.
People learn from the consequence, not the slogan.
The superpower is not motivation. It is priority enforcement.
”People Do What They’re Rewarded For”
People also do what they are protected for.
Rewards get attention because they are visible. Protection is quieter. A leader who misses a number while doing the right long-term work may be protected, punished, or quietly sidelined. The next layer watches.
If someone is rewarded for raising quality concerns, the behavior spreads. If someone is thanked publicly and penalized privately, the private lesson spreads faster.
The same is true for bad behavior. A rainmaker who breaks process and keeps getting promoted has been protected. A senior manager who hoards information and keeps authority has been protected. A team that burns people out and still receives headcount has been protected.
Incentives include what the organization tolerates when results are convenient.
Gaming Is a Design Outcome
Gaming is often described as employee misconduct. Sometimes it is. More often, it is feedback from the system.
People are showing the designer what the metric actually rewards.
If support agents close tickets before solving problems, the system rewards closure over resolution. If salespeople sell to customers who churn immediately, the system rewards booking over fit. If managers hire quickly and churn rises, the system rewards speed over durability.
The organization can blame people for exploiting loopholes. It still has to face the fact that the loophole was part of the incentive surface.
A good incentive design assumes people will optimize. It asks what damage optimization could do before the behavior becomes normal.
Power Decides Which Incentives Count
Not every incentive is written in a comp plan.
Some are embedded in access, attention, and fear. Who gets invited to strategy meetings. Whose mistakes are forgiven. Which departments get budget after missing targets. Which leaders are allowed to be difficult because they produce revenue.
Employees build a practical model of the organization from these signals.
They know when a stated value carries no consequence. They know when a metric can be missed by someone powerful and cannot be missed by someone disposable. They know when the rule is real and when it is theater.
Incentive quotes often talk as if incentives are neutral design tools. In organizations, incentives are also expressions of power. They reveal whose goals become metrics and whose costs stay invisible.
The Useful Question
The useful question is not, what behavior are we trying to incentivize?
It is: what will people stop doing once this incentive becomes real?
What will get gamed. What will become invisible. Who will absorb the externality. Which judgment calls will be replaced by target chasing. Which values will survive only when they do not conflict with the metric.
Incentives change behavior by changing the price of attention. The behavior you wanted is only part of what moves.
