Commit, best case and pipeline are the usual sales forecast categories, and on most teams they mean whatever each rep wants them to mean. This page gives a definition for each category that rests on evidence in the deal rather than on confidence, the close rate each category should achieve if the definitions are being followed, how to test that on the team's own history, how to roll the categories up into a forecast range, and a copyable forecast sheet.
Three words carry most sales forecasts. This page makes each one mean something that can be checked afterwards.
| Category | Evidence required | Expected share of value closing in period |
|---|---|---|
| Closed | Signed, booked | 100% |
| Commit | Customer has confirmed they will buy and when; decision-maker engaged directly; commercial terms agreed or in final review; paperwork path and signatories known; the rep would be surprised to lose | 85 to 95% |
| Best case | Dated next step agreed with the customer; a realistic path to signature inside the period; decision process known; at least one named risk | 35 to 55% |
| Pipeline | Close date inside the period; neither test above met | 5 to 15% |
| Omitted | Close date outside the period, or deal stalled | 0 |
Two things make these testable. Each rests on evidence in the deal record, not on how the rep feels. And each carries an expected close rate, so it can be wrong in a measurable way.
A deal is commit only if every line is yes.
Five out of six is best case. The list is short enough to apply in a minute and strict enough to keep commit at ninety percent.
Take a fixed point, such as week four of each quarter. For every deal, record its category at that snapshot and whether it closed won inside the quarter.
| Category at week 4 | Value forecast | Value closed in quarter | Close rate | Expected |
|---|---|---|---|---|
| Commit | $3,200,000 | $2,050,000 | 64% | 85 to 95% |
| Best case | $2,600,000 | $780,000 | 30% | 35 to 55% |
| Pipeline | $4,100,000 | $330,000 | 8% | 5 to 15% |
Commit is closing at 64 percent. On this team commit means something closer to likely. Either tighten its use with the checklist, or accept the measured rate and forecast with it; what cannot continue is rolling up commit at face value.
Then the same table per rep:
| Rep | Commit close rate | Best case close rate | Reading |
|---|---|---|---|
| A | 94% | 41% | Using the definitions |
| B | 52% | 18% | Optimistic in both; commit is hope |
| C | 100% | 78% | Sandbagging: commit deals held in best case |
This needs weekly snapshots of the forecast, which most CRMs do not keep; see forecast accuracy and bias in Excel.
With the team's measured rates:
| Component | Value | Rate | Contribution |
|---|---|---|---|
| Closed | $1,400,000 | 100% | $1,400,000 |
| Commit | $3,000,000 | 64% measured | $1,920,000 |
| Best case | $2,400,000 | 30% measured | $720,000 |
| Pipeline | $3,800,000 | 8% measured | $304,000 |
| Expected | $4,344,000 |
Floor = closed + commit at measured rate = $3,320,000 Expected = all four at measured rates = $4,344,000 Ceiling = closed + commit at face value + best case at the top of its measured range = about $5,400,000
The manager's call sits inside the range, adjusted for what they know: a large deal in best case that will either land or not, a rep whose commit is reliable. Report the range and the call. Once the definitions are being followed, the measured rates converge on the expected ones and the range narrows. The forecast accuracy benchmark covers how close the call should be at each point in the quarter.
| Move | Requires |
|---|---|
| Pipeline to best case | A dated next step agreed with the customer |
| Best case to commit | All six checklist lines |
| Commit to best case | Any checklist line turns to no. Move it the same day |
| Any to omitted | Close date leaves the period, or no customer action in 45 days |
Downgrades are the discipline. A commit deal whose signatory goes quiet is best case today, not at quarter end. The snapshot history shows how late downgrades happen: if most commit deals that failed were still in commit in the final week, the category is not being maintained.
Forecast: [rep or team], [period], week [n]
| Deal | Value | Close date | Category | Evidence: what the customer last said or did, date | Risk | Change since last week |
|---|
Totals by category. Measured close rates by category, trailing four quarters. Floor, expected, ceiling. Call: [value]. Change in call since last week, and why.
Categories as confidence. Feels good is commit.
Never tested. Nobody knows that commit closes at 64 percent, so it is summed at 100.
No snapshots. The category at week four is overwritten, and the test is impossible.
Late downgrades. Commit holds until the last day, then collapses.
One number reported. A point forecast with no range invites false precision and then blame.
Define each category by evidence, give it an expected close rate, snapshot the forecast weekly, and measure what each category actually closed at, for the team and per rep. Roll up with measured rates into a floor, an expected value and a ceiling, and make a call inside it. For the deal-level inspection that feeds the categories, see the pipeline review template; for the direction of each rep's miss, forecast bias by rep. Covirage keeps the weekly snapshots from the opportunity export and reports category close rates per rep each quarter.
The forecast is a range with a call inside it. The floor is closed plus commit at its historical rate. The ceiling adds best case at its rate and a little pipeline. The call is where the manager believes it will land. Reporting only commit understates; reporting commit plus best case overstates. The range, with the measured rates, is the honest version.
Stages describe where the deal is in the sales process; categories describe the rep's evidence-based judgement of whether it closes this period. A deal can be late-stage and not commit, because the date is uncertain. Categories add judgement that stages cannot carry, which is why their accuracy has to be measured, per rep.
The measurement shows it: a rep whose commit closes at 100 percent every quarter and whose best case closes at 80 is putting commit deals in best case. The fix is the same as for optimism: show each rep their own category close rates and have the conversation about the definition.