# A forecast is evidence, not optimism URL: https://www.coltasa.com/insights/forecast-evidence Author: Colin Constantine Published: 2026-10-05 Category: Forecasting > Why most early-stage forecasts miss, and how exit criteria, deal reviews and a few honest questions make the number something you can trust. Photo: [RDNE Stock project](https://www.pexels.com/photo/a-magnifying-glass-on-a-page-of-various-charts-7947669) on Pexels Ask an early-stage sales team how their forecast is built and you will often hear some version of: "Each rep tells us what they think will close." That is not a forecast. It is a collection of hopes, weighted by each seller's optimism. A good forecast is a judgement built on evidence. It should tell you what will close, what might, and why, in a way anyone on the leadership team can check. Getting there does not need expensive software. It needs discipline about what counts as proof. The problem is widespread: in Gartner's State of Sales Operations survey, [only 45 percent of sales leaders and sellers](https://www.gartner.com/en/newsroom/press-releases/2020-02-12-gartner-says-less-than-50--of-sales-leaders-and-selle) said they had high confidence in their organisation's forecast accuracy. ## Why forecasts miss - Stages are based on seller activity. A deal moves to "proposal" because a proposal was sent, not because the buyer agreed to anything. - Probability is assigned by stage. Every deal in stage three gets 40 percent, whatever the evidence says. - Close dates come from the quarter. The date is the end of the period, not a date the buyer has agreed to. - Nobody challenges the call. Pipeline reviews become status updates, so weak deals are never questioned until they slip. ## Define stages by buyer evidence Rewrite every stage so it can only be passed when the buyer has done something verifiable, using [exit criteria written as buyer actions](https://www.coltasa.com/insights/stage-exit-criteria). For example: - Qualified: the buyer has confirmed a problem worth solving and who else is involved. - Validated: the economic buyer has agreed the business case and a budget range. - Committed: the buyer has agreed a [mutual success plan](https://www.coltasa.com/insights/mutual-success-plans) with named owners and dates. - Contracting: procurement and legal are engaged and working from your paper. Now a stage means something. When a deal is in "committed", everyone knows what has actually happened. ## Use three categories, not percentages For each deal expected this period, ask the seller to call it one of three things: - Commit: you would bet your bonus on it. Every step to signature is known and agreed. - Best case: it can close this period if specific, named things happen. - Pipeline: real, but not this period. Track how accurate each seller's commit calls are over time, and use the categories to keep [the weekly forecast call](https://www.coltasa.com/insights/weekly-forecast-call) short. Within two or three quarters you will know whose forecast to trust, and who needs coaching. ## Run deal reviews, not pipeline reviews A pipeline review walks through every deal and asks "any update?" A deal review takes the few deals that matter most and tests them properly. Five questions do most of the work: - What problem is the buyer solving, and what happens if they do nothing? - Have we met the person who signs, and what did they say? - What are the remaining steps to signature, and who owns each one? - What could stop this deal, and what are we doing about it? - Where did the close date come from? If the seller cannot answer one, the deal is not in commit. That is not a criticism of the seller. It is useful information, found early enough to act on. ## Watch the trends, not just the number Beyond the deals themselves, a few simple measures show whether the forecast is healthy: how many deals slip each period, how long deals spend in each stage, how much [pipeline you have against target](https://www.coltasa.com/insights/pipeline-coverage), and how accurate last quarter's commit was. A sudden rise in slipped deals or time in stage is an early warning, usually weeks before the number misses. Slippage is more common than most teams assume: [Ebsta’s 2024 benchmark data](https://ebsta.com/wp-content/uploads/2024/07/H1-Update-2024-B2B-Sales-Benchmarks.pdf) found that between 34 and 46 percent of deals slipped past their close date each quarter in the first half of that year. ## Make honesty safe None of this works if sellers are punished for bad news. If moving a deal out of commit leads to a difficult conversation, sellers will keep it there until the last day. Reward accurate calls, including the ones that come down. A team that tells the truth early gives leadership time to fix things, and that is the whole point of a forecast. ## Sources and further reading - [Gartner (2020): Less than 50% of sales leaders and sellers have high confidence in forecasting accuracy](https://www.gartner.com/en/newsroom/press-releases/2020-02-12-gartner-says-less-than-50--of-sales-leaders-and-selle) - [Ebsta: H1 2024 B2B sales benchmarks (PDF)](https://ebsta.com/wp-content/uploads/2024/07/H1-Update-2024-B2B-Sales-Benchmarks.pdf)