How do I design a comp plan that doesn't backfire?
You want a plan that drives the right behavior. But you are really asking for a plan that needs no judgment afterward — one that runs itself. No such plan exists. Every incentive is a bet on what people will do when the number is all that matters to them. The mistake is not choosing the wrong metric. The mistake is believing a metric can replace your attention. You design the plan, hand it off, and hope. Then you are shocked when someone optimizes exactly what you measured instead of what you meant. You cannot eliminate the risk of gaming. You can only decide, in advance, what you will do when someone finds the loophole — and build that answer into the plan now, while you are calm, instead of later, when you are angry.
You will not out-think a smart person motivated to game your plan. Stop trying to predict every trick. Instead, write down the single ugliest behavior the plan could reward, and remove the path to it. Choose the cap, the clawback, or the condition. Do this before you choose the metric.
What changes unlock by starting
- A written list of the plan's weak points, made before launch instead of discovered after
- At least one cap or clawback built in against the worst-case gaming scenario
- A second person's honest read on where the plan can be gamed
- A fixed date to check what behavior the plan actually produced