Weighted pipeline is a mechanical way to estimate how much of today's open pipeline will close. Instead of relying on each rep's judgment, it gives every stage a probability and multiplies. It works best as a second opinion next to the forecast call, not as a replacement for it.
Worked example
A team's open pipeline for the quarter, with every deal at $10,000 to keep the arithmetic simple:
| Stage | Deals | Amount | Stage probability | Weighted |
|---|---|---|---|---|
| Discovery | 40 | $400,000 | 10% | $40,000 |
| Evaluation | 20 | $200,000 | 30% | $60,000 |
| Proposal | 10 | $100,000 | 50% | $50,000 |
| Negotiation | 5 | $50,000 | 80% | $40,000 |
| Total | 75 | $750,000 | $190,000 |
$750,000 of open pipeline becomes $190,000 of weighted pipeline. If the team has already won $120,000 this quarter, the weighted forecast is $120,000 + $190,000 = $310,000.
Where the probabilities come from
A stage probability is a claim: "this share of deals in this stage will be won." The only way to test it is to follow deals that reached the stage through to their outcome. That is a cohort calculation, a stage-to-close win rate:
Stage probability ≈ deals that reached the stage and were won ÷ all deals that reached the stage
Use cohorts old enough for nearly all their deals to have an outcome. For example, if 90 deals reached Proposal over the last four quarters and 45 of them were won, Proposal's probability is about 50%. Use your own history, not the CRM's installed defaults or a guess, and review the numbers once or twice a year at a scheduled point. When you change a stage's definition, mark the date and don't mix old and new history.
Probabilities only mean something if stages do. If deals move forward because the seller did something ("proposal sent") rather than because the buyer verified something, each stage mixes real and hopeful deals and its probability becomes an average of both. That's why buyer-verified stage exit criteria come before weighting.
Where weighting breaks
- Small numbers. Weighting works on averages. It's reasonably reliable across a hundred deals and unreliable across five.
- Large single deals. No deal closes at 80%. A $200,000 deal in Negotiation adds $160,000 to the weighted number, but the real outcome will be $0 or $200,000.
- Dirty pipeline. Weighting trusts the stage and close date on every record. Stale deals (no buyer activity for more than 30 days) and past-due close dates inflate the result.
- Wrong probabilities. Probabilities that are too low understate early pipeline; too high, and they overstate it. Either way, the error repeats every quarter until someone recalibrates.
When the reps' call and the weighted number disagree by a lot, don't average them. Ask why. A weighted forecast well below the call is often a hint that some committed deals deserve a second look.
Common mistakes
- Using default probabilities. CRM defaults are placeholders, not evidence about your buyers.
- Weighting stale deals. Clean the pipeline first, exactly as for pipeline coverage.
- Forgetting what's already closed. A weighted forecast adds closed-won revenue to the weighted open pipeline.
- Treating it as the forecast. Use it to challenge the call, alongside forecast categories, not to replace judgment.
- Calibrating from one quarter. Change probabilities only once three or four quarterly cohorts are old enough to have outcomes, and the pattern holds across them.
To check how well a weighted forecast did, score saved snapshots of it with forecast accuracy. The free RevOps Foundations course covers weighted pipeline in Lesson 6, "Forecasting basics".