The reason most forecasts miss is not bad math — it is using the wrong method for the business. A rep-committed forecast is great when your reps are disciplined and terrible when they sandbag. Historical run-rate is rock-solid for a high-volume SMB motion and useless for a team closing six enterprise deals a quarter. There is no single best method; there is only the right method for your deal shape, plus the discipline to cross-check it.
Here are the four methods that actually get used in the field, what each is good at, and where each one will burn you. If you want to move from “a guess” to something a board can plan against, this is the foundation of your revenue architecture.
1. Pipeline-stage forecasting
You assign each open deal a forecast category based on its stage — Commit, Best Case, Pipeline — and roll up the categories. It is judgment-driven but structured, and it is the workhorse method for repeatable B2B motions.
Use it when: your stages are well-defined with clear exit criteria and your reps update the CRM. It lies when: stages are vanity labels that do not reflect real buyer progress. Fix that first — see our guide to CRM pipeline hygiene, because this method is only as honest as your stage definitions.
2. Weighted pipeline forecasting
You multiply each deal by the historical win rate of its current stage and sum the results. A $100k deal sitting at a stage that historically converts 40% of the time contributes $40k. It is more objective than category forecasting because it uses real conversion data instead of rep opinion.
Weighted pipeline is only as good as your stage-conversion percentages. If you have not measured your actual win rate by stage — from a large enough sample — you are multiplying by made-up numbers. Build the conversion baseline first; without it, this method manufactures false precision.
Use it when: you have clean historical stage-conversion data and enough deal volume to make the averages meaningful. It lies when: deal sizes vary wildly, because one $2M outlier at 30% can swing the whole number.
3. Historical run-rate forecasting
You forecast forward from what the business has produced recently, adjusted for seasonality and growth trend. If you have closed $1.2M a quarter for four straight quarters and you are growing 15% a year, next quarter is roughly $1.25M.
Use it when: you have a stable, high-volume motion — SMB, transactional, PLG-assisted — where the law of large numbers smooths out individual deals. It lies when: the business is early, lumpy, or changing fast. Run-rate cannot see a market shift or a new competitor coming; it only extrapolates the past.
4. Rep-committed forecasting
The manager asks each rep what they will close, challenges it deal by deal, and rolls it up. Nothing beats it for capturing on-the-ground signal — a champion who went quiet, a budget that just froze — that no formula can see.
Use it when: deals are large and few, and rep-level judgment carries real information. It lies when: reps sandbag to beat expectations or happy-ears their way to a number to keep managers off their back. This method lives or dies on forecast discipline and a manager who inspects rather than accepts.
The fix for both failure modes is the same: inspect the deal, not the number. A rep who commits a deal should be able to name the economic buyer, the compelling event, the next scheduled step, and the paper process. If they cannot, the deal is not a commit no matter what category it sits in. Managers who run their forecast this way — deal by deal, evidence by evidence — get a rep-committed number that is worth trusting; managers who just ask “are you going to close it?” get theater.
The move that actually works: triangulate
Elite revenue teams do not pick one method. They run two or three, then reconcile the gap. When your weighted pipeline says $1.4M, your run-rate says $1.25M, and your reps commit $1.55M, the disagreement is the insight — it tells you exactly where to inspect. A forecast that three methods agree on is one you can take to the board. Consistently landing inside 5% is a learnable skill; our guide to forecasting accuracy within 5% walks through the operating rhythm, and the step-by-step CRM setup lives in the how-to library.
Forecast still a coin flip?
Get a free 30-minute revenue checkup. Walk us through how you forecast today and Kory White — a 25-year revenue exec, Maryland-based and working nationwide — will pinpoint why it misses. No pitch, no obligation.
Get your free revenue checkup Meet KoryFrequently asked questions
No single method is most accurate in every situation. Historical run-rate is best for stable, high-volume businesses; pipeline-stage forecasting suits repeatable mid-market motions; weighted pipeline works when your stage conversion data is clean. The most accurate forecasts triangulate two or three methods and reconcile the gap.
Weighted pipeline forecasting multiplies each open deal by the historical win probability of its current stage, then sums the results. A $100k deal at a 40% stage contributes $40k to the forecast. It is only as good as your stage conversion data — garbage probabilities produce garbage forecasts.
Run a formal forecast weekly during the quarter, with a tighter commit call in the final two weeks. Weekly cadence catches slippage early enough to act. Monthly is too slow for most B2B teams; daily creates noise without adding signal.