Why Your Sales Forecast Is Just Your Pipeline Total

Most small-business sales forecasts are the total of every open deal, adjusted by a gut feeling. That number misses because it treats a deal that had one call last month the same as a deal waiting on a signature. A forecast that holds up separates deals by evidence, not by hope.

What is wrong with forecasting from the pipeline total?

An unweighted pipeline total counts every open deal at full value, including deals that have gone quiet, deals with no decision date, and deals the buyer has already mentally rejected. Adding them up tells you what happens if every buyer says yes this month. A $5,000 deal nobody has touched in six weeks still adds $5,000 to that total, so the forecast overshoots by the full value of every stalled deal.

On a small team the problem compounds because the same person owns the deals and the forecast. Nobody wants to mark their own deal as dead, so stalled deals stay open with their original close date, then get pushed a month, then pushed again. The pipeline total stays high while the closed number stays flat.

How does a weighted pipeline forecast work?

A weighted forecast multiplies each deal's value by a probability tied to its stage, then adds the results. A $10,000 deal at an early stage might count as $2,000, while the same deal at contract review counts as $8,000. The total is smaller than the pipeline and, once the stage probabilities match your own history, often closer to what actually closes.

The weakness is that stage probabilities are often guesses copied from a template. If your own history shows that deals at the proposal stage close half as often as the template says, the weighted forecast inherits that error. The fix is to set each stage's probability from your own closed and lost deals over the last year, not from a default.

What evidence should a deal need before it counts?

A deal should only count toward this period's forecast when the buyer has done something, not when the seller has. A confirmed decision date, a named decision maker who has seen the price, and a next meeting on the calendar are buyer actions. A proposal sent and a follow-up email are seller actions and say little about when the buyer will decide.

A simple rule set for a small team:

• Commit: the buyer has confirmed a decision date inside this period and has seen final pricing.
• Possible: the buyer is engaged and has a next meeting booked, but no decision date.
• Not this period: no buyer response in two weeks (or whatever counts as silence in your sales cycle), or no decision maker identified.

Forecast the commit column as your number. Report the possible column separately as upside. Deals in the last group come out of the forecast entirely until the buyer does something new.

Why do close dates keep slipping?

Close dates slip because they are usually set by the seller when the deal is created, often as the end of the current month or quarter. Nothing about the buyer's process went into that date. When the date passes, the deal gets pushed forward without anyone asking what changed, so the pipeline fills with deals that have been "closing next month" for a season.

Track how many times each deal's close date has moved. A deal pushed three times is a different kind of deal from one that is on its first date, and it should not carry the same weight. This is the same pattern behind late-stage deals that push past quarter end: the date reflects when the seller hoped to close, not when the buyer planned to decide.

How do you know if your forecast is getting better?

Comparing each period's committed forecast to actual closed revenue over three or four periods reveals the accuracy of sales forecast rules. (source) If you consistently close well under the commit, your rules are too loose. If you close well over, deals are being held back from commit.

Keep the comparison in the same sheet as the forecast so it cannot be skipped. If last quarter's commit said $40,000 and $26,000 closed, that $14,000 gap is the amount you should not have hired or spent against, and it tells you exactly how far to tighten the rules.

When a stalled deal keeps its open status, a default pipeline report keeps including it until someone marks it lost, so a deal that died in June can still be counted in a September forecast. Rebuilding stage probabilities from your own won and lost history is part of what the Boost Audit does, since that history is usually sitting in the CRM unused.