Win rate tells you how often you win the deals you actually compete for. It sounds like the simplest metric in sales, yet two people in the same company can calculate it two ways and both defend their number. The fix is to write the definition down and label every win rate you report.
Worked example
During one quarter, a team closed 100 opportunities: 30 won and 70 lost.
| Measure | Calculation | Result |
|---|---|---|
| Win rate by count | 30 ÷ 100 | 30% |
| Value won | 30 deals, average $14,000 | $420,000 |
| Value lost | 70 deals, average $18,000 | $1,260,000 |
| Win rate by value | $420,000 ÷ ($420,000 + $1,260,000) | 25% |
Same deals, same quarter: 30% against 25%. The gap comes from deal size. The team's won deals averaged $14,000 and its lost deals $18,000, so it wins its smaller deals more often than its bigger ones.
Count or value: which one to use
The two answer different questions.
- By count describes sales execution: of the deals we worked to an outcome, how many did we win? Use it in the sales velocity formula, which multiplies a number of deals by the chance of winning each one.
- By value describes revenue: of the dollars we competed for, how many did we win? Use it with dollar-based measures, such as working out the pipeline coverage you need (about 1 ÷ win rate by value).
When win rate by value is lower than by count, you are losing your larger deals. That's a finding worth investigating, not a rounding error. Report both, and always say which one you mean.
What counts as "closed"
The denominator is closed opportunities only. Open deals haven't had an outcome yet, so including them drags the rate down for no real reason. Decide the edge cases once:
- Disqualified opportunities (not a fit, no problem you solve, no access to the buyer) count as closed. Some teams also report a "qualified win rate" that leaves them out; it's a useful coaching number, but it always looks higher, so label it.
- Duplicates and records created in error are excluded completely. They were never real opportunities.
Snapshot vs cohort win rate
There are two ways to choose which deals go into the calculation.
- A snapshot win rate takes every opportunity that closed in the period, whenever it was created. It answers "what did we win this quarter?" and is the usual scoreboard for bookings and forecasts.
- A cohort win rate takes every opportunity created in a period and follows that group over time. For example, 120 opportunities were created in one quarter; as of the end of the next quarter, 36 have been won, 60 lost and 24 are still open. Like any win rate, the cohort win rate divides by closed deals only: 36 ÷ (36 + 60) = 36 ÷ 96 = 37.5% as of that date.
A cohort number keeps moving until the cohort has had time to close, so always state the "as of" date, and compare cohorts at the same age. Cohorts are the honest view when you want to know whether a process change worked, because old deals created under the previous process aren't mixed in. Snapshots can drop sharply after a pipeline clean-up, when many old deals are finally closed out as lost.
Some teams also report the share of the whole cohort won so far, dividing by every deal created, open ones included: 36 ÷ 120 = 30% in the example. That is a different measure, a cohort conversion rate, not a win rate. It rises as open deals close, so label it clearly and don't compare it with a win rate.
Common mistakes
- Including open deals in the denominator. That measures progress, not outcomes.
- Not saying count or value. The two can differ by several points, and both look like "win rate".
- Comparing a snapshot with a cohort. They have different denominators and time windows, so they will rarely match.
- Reading a young cohort as final. Weeks after creation, a cohort win rate rests on the few deals that closed quickly, which are rarely typical, and a cohort conversion rate looks low simply because most deals are still open. Wait until most of the cohort has an outcome.
- Blending segments. Small and large deals often win at different rates. Split by segment once you have enough deals for the numbers to mean something.
Win rate also sits behind weighted pipeline, where each stage's probability is a stage-to-close win rate. The free RevOps Foundations course covers win rate by count and by value in Lesson 3, "The metrics that matter".