Start with these five checks.
- Confirm what counts as an opportunity, win, loss and no decision.
- Compare mature cohorts by qualification date, not mixed calendar snapshots.
- Control for segment, lead source, product, region and deal size.
- Find the first stage where conversion or elapsed time changed.
- Validate CRM loss reasons against calls, notes and actual activity.
Ask for only what can change the answer.
- CRM historyStage transitions and field changes
- Opportunity mixSegment, source, product and amount
- Sales activityMeetings, email timing and coverage
- Commercial termsPrice, discount and approvals
- Customer evidenceObjections, trials and no decision
- OutcomeWon revenue, margin and cycle time
Test one reversible move.
Take one mature segment with a clear stage break. Apply one change—such as a tighter qualification rule or faster approval path—to a limited cohort, while holding source and pricing stable. Compare profitable won revenue, stage conversion and cycle time against the prior matched cohort; stop if discounting or poor-fit wins increase.
A decision your team can use.
- 01A reconciled win-rate definition
- 02A cohort and mix decomposition
- 03The first supported stage break
- 04A reversible validation plan
Common questions.
What denominator should win rate use?
Use a clearly defined qualified-opportunity cohort and state how open, withdrawn and no-decision deals are treated.
How much history is enough?
Use enough mature cohorts to cover the normal sales cycle and important segments; a short recent window can be misleading.
Should we rank reps first?
No. Territory, source and account mix should be controlled before individual performance is interpreted.