Why is our sales forecast always wrong?
Sales and finance bring different pipeline numbers to the same meeting, the forecast misses in both directions, and each miss is explained after the fact by a different one-off reason.
Almost always because stage definitions are subjective. If 'proposal' means one thing to one rep and something else to another, stage-weighted forecasting is arithmetic on inconsistent inputs and no amount of dashboarding fixes it. The other three common causes, in order of frequency: close dates that reps update to avoid pipeline reviews rather than to reflect reality, deals sitting in stages nobody audits, and forecast calls that reward optimism over accuracy. Every one is a definition or incentive problem, not a tooling problem.
The fix that costs nothing: rewrite each stage so that entry depends on something the buyer did, not something the rep feels. 'Proposal sent' is observable. 'Verbal commitment' is not. Then check historical conversion by stage — if two stages convert at the same rate, they are one stage wearing two names, and merging them will improve forecast accuracy more than any model will.
Why forecasts drift from reality
- Stages describe activity, not evidence. 'Had a demo' is an activity. 'Buyer has confirmed budget and named a decision date' is evidence. Only the second is forecastable.
- Close dates are negotiated, not observed. When a slipping date triggers an uncomfortable conversation, dates stop reflecting reality.
- Nobody audits stale deals. Pipeline that has not moved in 90 days is usually dead, and it inflates every rollup until someone removes it.
- The forecast call rewards the wrong thing. If optimism is praised and caution is questioned, you have trained the number you receive.
How to make the forecast trustworthy
-
01
Write entry and exit criteria for every stage, in evidence terms. This is the single highest-leverage fix and costs nothing.
-
02
Have two leaders independently forecast the same quarter. The size of the gap tells you how bad the definition problem is.
-
03
Age your pipeline. Anything untouched for 90 days goes to a separate bucket and out of the forecast.
-
04
Separate the commit number from the pipeline number, and track commit accuracy per rep over time. Accuracy is a coachable skill; optimism is not.
-
05
Only then look at tooling. A CRM cannot forecast a process you have not defined.
Fixing definitions is a week of work. Making them stick is the hard part, because it changes what people can say in a pipeline review — and the first quarter after you tighten stages, the forecast usually gets worse before it gets better, as previously hidden slippage becomes visible. Leaders who are not warned about that dip frequently abandon the change one quarter before it pays off.
More on forecast accuracy
One to two full sales cycles after definitions are fixed. Trust comes from being right several times consecutively, which cannot be compressed below the length of your cycle.
Not to start. The definition work is a leadership exercise. A RevOps function matters for maintaining discipline once it exists — hiring one before the definitions are agreed just adds a person to an unresolved argument.
No. It will render the same undefined process more attractively. Every failed forecast we have diagnosed traced back to definitions and incentives, not to the platform.
More on CRM Services & Revenue Systems
Still not sure this is your problem?
A 20-minute fit check. We will tell you if it is something you can fix without us — that happens often enough that we lead with it.
Book a Strategy Session →Browse every answer.
The full Solution Desk — specific questions, straight answers, no gate.
Open the Solution Desk →