Define the segment before you measure it
The definition has to be stable and set before the outcome is known. Employee count band, revenue band, or an assigned account tier stamped at opportunity creation all work. Segmenting by closed deal size does not, because deal size is partly a result of the sale you are trying to evaluate.
Use the same definition everywhere. If quota planning uses one segmentation and win rate reporting uses another, the two reports will disagree and nobody will trust either.
Reading the spread
Three questions make the spread useful.
Is the gap structural or new? Compare segment win rates across the last several quarters. A gap that has been stable for two years is a property of the motion. A gap that opened last quarter is an event worth investigating. Is pipeline mix shifting toward the weaker segment? Total win rate can fall while every segment holds steady, purely because more of the pipeline moved into the lower converting band. That is a routing and targeting change, not a selling problem. Does deal size compensate? A lower win rate in enterprise can still produce more revenue per opportunity worked. Rank segments on expected value per opportunity, not on the percentage alone.Segment win rates change your coverage math
Coverage means nothing without the conversion rate behind it. A segment that wins one in five needs far more pipeline to land its number than a segment that wins one in three, so a single company wide pipeline coverage target either starves one segment or overbuilds the other.
ORM sees coverage across its customer base run from 1.4x to 5x, with most customers near 3.5x. The useful move is to set the target per segment from that segment's own win rate and average deal size, then check whether the pipeline sitting in each band is capable of producing the number. The argument for abandoning a blanket multiple is laid out in the 3x pipeline coverage rule is wrong.
Segment level rates also improve the forecast directly. See win rate for the base calculation and the other cuts worth building.
Frequently Asked Questions
Why does win rate differ so much between SMB and enterprise?
The two motions have different buying committees, different evaluation lengths, and different qualification bars. SMB deals often close or die inside a few weeks with one or two people involved. Enterprise deals run longer, involve more stakeholders, and carry more chances to stall. A single blended win rate hides both patterns and describes neither motion accurately.
How should you define segments for win rate reporting?
Use a definition that is stamped on the account and does not move, such as employee count band or revenue band at the time the opportunity was created. Segmenting by deal size after the fact creates circular logic, because deal size is partly an outcome of the sale. Whatever definition you pick, apply it identically to pipeline, quota, and win rate reporting.
What does a widening gap between segment win rates tell you?
It usually means one motion has changed while reporting still treats them as one business. Common causes are a shift in lead mix into a segment you are weaker in, a new competitor concentrated in one band, or reps being asked to sell upmarket without the supporting process. Look at the segment mix of created pipeline before concluding that execution got worse.
Should each segment have its own pipeline coverage target?
Yes. Coverage is only meaningful relative to the conversion rate behind it. A segment that wins less often needs more coverage to land the same number. ORM sees coverage ratios across its customer base ranging from 1.4x to 5x, with most sitting near 3.5x, which is a reminder that a single company wide multiple rarely fits every motion inside the company.
Put these metrics to work
ORM builds custom revenue forecast models that turn concepts like win rate by segment into prescriptive action for your team.
Schedule a Demo