Win rate by deal size

Win rate by deal size is the percentage of opportunities that close as won within each deal value band, calculated separately for small, mid, and large deals rather than as one blended figure. If 40 small deals reach a decision and 24 close as won, the small band win rate is 60 per cent. If 20 large deals reach a decision and 4 close, the large band win rate is 20 per cent. The same pipeline produces two very different numbers.

A single blended win rate averages these together and tells you almost nothing about where to act. A healthy small-deal motion can mask a collapsing enterprise motion, or the reverse. Segmenting by deal size turns one vague number into a map of exactly which part of the funnel needs attention.

Deal size bands are usually set against your own pricing, not absolute thresholds. A common split is below 10,000 pounds, 10,000 to 50,000 pounds, and above 50,000 pounds, but the right cut points are the ones where your sales process genuinely changes: where a solutions engineer joins, where procurement and security review enter, where the buying committee grows.

Calculating win rate by deal size means running the standard win rate formula once per value band, on the same closed-deal population. The discipline is in defining the bands cleanly and in deciding what counts as a closed deal.

1. 1

   Define the value bands

   Set 3 to 4 deal value ranges that map to real changes in your sales motion, not round numbers for their own sake. Use the contract value or first-year value consistently so a deal cannot drift between bands depending on how it is measured.

2. 2

   Assign each closed deal to a band

   Place every opportunity that reached a won or lost outcome into exactly one band based on its value. A deal must not appear in two bands. Open and disqualified opportunities are left out entirely.

3. 3

   Count won and total closed per band

   For each band, count the deals that closed as won and the total deals that closed at all. The total is won plus lost within that band, which becomes the denominator.

4. 4

   Divide and compare across bands

   Divide won by total closed in each band and multiply by 100. The point is the comparison: a 55 per cent small-deal win rate next to a 18 per cent large-deal win rate tells a story that a blended 40 per cent never could.

Sample size matters more here than with a blended rate. If the large band has only 6 closed deals in a quarter, a single win swings the rate by 17 points. Report the deal count alongside each band so readers can judge how much weight to place on the percentage, and widen the time window for thin bands rather than reacting to noise.

A metric tree decomposes win rate by deal size into the stages and conditions that produce each band outcome, then traces those back to the people who can change them. This turns a segmented percentage into a diagnosis.

The first level splits the metric into its value bands. Each band then decomposes into the factors that move its win rate: the quality of fit at qualification, how well the team navigates the buying committee, competitive positioning, and the cycle length that determines how many deals slip rather than close. Large deals fail for different reasons than small deals, so the same leaf nodes carry very different weight across bands.

The value of the tree is precision of ownership. If the large band win rate is dragging, the tree shows whether the cause is weak enterprise qualification, a stalled buying committee, or losses to a specific competitor. Each diagnosis points to a different owner and a different intervention rather than a generic instruction to close more deals.

Win rate by deal size benchmarks shift with motion and segment, so the most useful comparison is the gradient across your own bands rather than any single absolute number. In most B2B teams the rate falls as deal value rises, because larger deals carry longer cycles, more stakeholders, and more competition. A flat or rising gradient is unusual and worth understanding.

Read these ranges as starting points, not targets. A team selling a complex platform into regulated buyers will sit at the low end of every band and still be healthy if its average deal size and sales pipeline velocity are strong. The signal to act on is a band that has drifted well below its own trailing average, especially when the deal count is large enough to trust.

Improving win rate by deal size means treating each band as a separate problem with its own owner. The intervention that lifts small-deal close rates rarely touches large deals, and effort spread evenly across bands usually wastes most of itself.

The metric tree approach starts by finding the band with the widest gap between its current win rate and its own potential, then drilling into which leaf is responsible. If large deals are losing on buying-committee coverage rather than product, no amount of discounting fixes it.

KPI Tree connects each band and each leaf to the people who can move it. Sales leadership owns enterprise qualification, account executives own multi-threading, and revenue operations owns the loss-review discipline that keeps the data honest. With RACI ownership on every node, the accountable owner is pushed the moment their band win rate moves, and the verified impact loop checks whether the change they made actually shifted the number rather than assuming it did.

1. 1

   Reporting a single blended win rate

   A blended figure averages a strong small-deal motion with a weak enterprise one and hides both. The whole point of this metric is the comparison across bands, which a blended number destroys.

2. 2

   Ignoring sample size in thin bands

   When the large band has only a handful of closed deals, one outcome swings the rate by double digits. Reacting to a band with too few deals chases noise. Report the count and widen the window instead.

3. 3

   Letting deals drift between bands

   If one report uses contract value and another uses first-year value, the same deal lands in different bands and the trend becomes unreadable. Fix the value definition once and apply it everywhere.

4. 4

   Counting open deals in the rate

   Win rate is a closed-deal metric. Including opportunities still in the pipeline understates every band, because a deal that has not closed cannot be counted as won.

5. 5

   Treating all bands with the same playbook

   Applying the small-deal motion to enterprise deals, or vice versa, depresses both. Each band needs its own qualification, its own stakeholder coverage, and its own loss analysis.