Sales Head committed ₹6Cr MRR for the quarter. Closed ₹3.4Cr.
Who decides when a deal moves to commit?
The executive owns his number. My Sales Head reviews it.
Reviews it against what?
[long pause]
"His judgement."
The myth vs the pattern
The myth: the team is over-promising. Tighten the discipline and the forecast will hold.
The pattern: when stages are defined by judgement, two sales executives looking at identical deals will stage them differently. Neither is wrong. There is nothing to be wrong against.
So the review can only test confidence. Not evidence.
A forecast built that way is not a forecast. It is an average of opinions, presented as a number — and it may miss every quarter, in a different direction, for reasons nobody can trace afterwards
The reason it misses in a different direction each time is important. If you were consistently over by 40%, you could model for it. The unpredictable miss is the signature of a judgement-based system — there is no consistent bias to correct, because the inputs are different people's confidence on different days.
The 20-minute CRM test
Run this before the next forecast call. It takes twenty minutes and will tell you more than the last three reviews did.
The 4-Step Test
Split last quarter's commit deals into two lists
Deals that closed. Deals that didn't. This is your control group — same stage definition applied to both, real outcomes already known.
For every deal in both lists, write down the last verifiable event
Not a call. Not "good conversation." Something that demonstrably happened:
- → A document sent and acknowledged
- → A price approved in writing
- → A security or legal review started
- → A specific date named by the buyer for a decision
- → A purchase order or contract initiated
Count how many deals in each list have nothing
That number is your answer. If most closed deals had a verifiable event and most missed ones had nothing — your team is not over-promising. They are staging on confidence, because nobody gave them anything else to stage on. The problem is the definition, not the discipline.
For the deals with nothing — check who put them in commit
This one separates two completely different problems:
Spread across the team
Definition problem. Nobody was told what commit actually requires.
One person
Management problem. That individual needs coaching or is working around a system.
Different fix. Same symptom.
What the test usually shows
In most teams that run this, the closed-deal list has a clear pattern: something verifiable happened in the two to four weeks before close. A document exchange. A named date. A procurement process started. Something external to the salesperson's assessment of the situation.
The missed-deal list tends to look different. Activity — calls, meetings, follow-ups — but nothing you could point to as evidence that the buyer had moved. The salesperson's read on the deal was positive. The buyer's behaviour had not confirmed it.
That gap is the forecast miss. It is not a people problem. It is a definition problem that has been reviewed as a people problem, which is why tightening discipline alone has never fixed it.
Define what commit requires in verifiable terms. Run the same test again in two quarters. The forecast will not be perfect — deals still fall through for reasons outside any system's control. But the misses will cluster closer to the number, and when they happen, you will be able to say why.