Designing Variable Pay That Survives Contact with Reality

What this topic is

Variable pay is any portion of someone's compensation that depends on a measured result rather than on showing up: commission, quarterly bonus, objective-based payout, profit share, gainsharing. Designing it means making five decisions: the measure, the mix between fixed and variable, the shape of the payout curve, the period it settles over, and who adjudicates the number. The phrase "survives contact with reality" is the whole test. A plan is not judged by the model that produced it. It is judged by what a rational person does in the eleventh month of a twelve-month plan when they can see exactly where they stand.

Why it matters

Variable pay is usually the largest deliberately designed behavioural lever a manager controls, and it is routinely designed by copying whatever the last company did. The costs of getting it wrong are not subtle: effort moves to the measured thing and away from everything else, deals get pushed across period boundaries, high performers leave when a target is raised on them, and the finance line for "incentive compensation" grows without any observable change in what people do. Every one of those failures is predictable from the plan document before it ships.

Who should learn it

Founders writing their first commission plan, sales and operations leaders inheriting one, general managers who have watched three consecutive bonus schemes change nothing, and anyone who has to defend a compensation structure to both a board and the people paid under it.

What you will understand

• The five design decisions, and which one people spend the least time on relative to its impact
• How to build a plan from on-target earnings backwards, and what the resulting cost-of-sales ratio tells you
• What thresholds, caps and accelerators do to behaviour at the exact moments they bind
• The clauses that quietly decide whether the promise is worth its face value

Prerequisites

• The Founder-to-manager Transition and the First Fifty People
• Escalation of Commitment and the Project Nobody Will Kill

Common misconception

"If we get the numbers right, the plan will work." The numbers are the easiest part and the least load-bearing. What determines whether a plan works is whether the measure is attributable to the person, whether the promise is credible enough that they plan their year around it, and whether the payout curve is flat in the places where you need continuous effort. A second and more expensive misconception is that more variable pay always means more motivation. Past a point it means more risk transferred to someone who charges you for carrying it, more effort reallocated onto the measured dimension, and more negotiation about the target. That is why the highest-variable plans in a business are frequently the ones the finance team spends the most time arguing about.