Anatomy of a Two-Week Incentive Sprint

Most incentive work fails for a structural reason rather than an analytical one. A transformation program takes nine months, during which the sponsor changes roles, the strategy shifts, and the recommendation is absorbed into a slide nobody executes.

A sprint is designed against that failure mode. Ten working days, one behavior, one edit that survives the sponsor.

The constraint is not what we can analyze in two weeks. It is what an organization can actually absorb in one change.

Days 1-2: name the behavior precisely

The engagement starts with a behavior statement that a skeptic could verify: not 'sales is short-termist' but 'in six of the last eight quarters, deals closing in the final week of the quarter had materially worse twelve-month retention.' If the behavior cannot be stated that way, the first two days are spent making it so.

Half of all disagreement inside leadership teams dissolves here, because the argument was about the description rather than the cause.

Days 3-6: reconstruct the payoff structure

We pull the records that were created for other purposes — promotions, quota rules, budget movement, escalation history, calendars in aggregate — and reconstruct what the system rewards across the four layers: economic, status, safety, and temporal.

Interviews come after the records, not before, and they exist to test the hypotheses the records produce. Interviews conducted first collect opinion; interviews conducted second collect confirmation or contradiction.

Days 7-8: find the single highest-leverage edit

There are always five or six things worth changing. Shipping five or six is how nothing changes. We rank candidate edits on how much of the behavior each explains, how quickly it can be made, and whether it survives a leadership transition.

The winner is usually a removal rather than an addition — a ratchet taken out of the quota, a retroactive penalty retired, a definition moved to a downstream owner. Removals are cheaper politically and they do not decay when attention moves on.

• —How much of the observed behavior does this edit explain?
• —Can it be implemented inside one planning cycle?
• —Does it survive a change of sponsor?
• —What counter-metric proves it worked or failed?

Days 9-10: ship it with the measurement attached

The final two days produce the change itself — the revised rule, the reassigned definition ownership, the paired metric — plus the review date and the person accountable for reading it. A change without a scheduled read is a change that will be quietly reversed by the next planning cycle.

Then we leave. The engagement is designed so that nothing depends on us being in the room in month three.

If the fix requires our continued presence, it was not a fix. It was a dependency.

What happens in the quarter after

Behavior moves slower than structure. The rule changes in week two; the behavior shifts over the following six to twelve weeks as people test whether the new payoff is real. The counter-metric is read at the end of that window, and it either confirms the mechanism or tells you the payoff you identified was not the binding one.

Both outcomes are useful. A failed edit with a clean measurement narrows the search far more than a successful program with no measurement at all.