Here's a number that should bother you more than it does. Most B2B sales teams forecast with about the accuracy of a coin toss.
Not because they're lazy. Not because the tools are bad. Because the forecast is built on the wrong foundation, and almost nobody questions the foundation.
I've been in B2B sales for 20 years. I've sat in hundreds of forecast calls, on both sides of the table. The conversation is nearly always the same. A rep says "I'm feeling good about this one." A manager nods and moves it to commit. Three weeks later the deal slips, or it dies, and everyone acts surprised.
It wasn't a surprise. It was never measurable in the first place.
The forecast is a story, not a calculation
Walk into most pipeline reviews and you'll hear the forecast described in the language of feelings. Strong relationship. Champion's on board. They love us. Should land this quarter.
None of that is data. It's narrative. And narrative is exactly what a forecast should not be made of.
The problem isn't that reps are dishonest. Most of them genuinely believe their deals. The problem is that human judgement, on its own, is a terrible instrument for predicting which deals close. We're optimistic by nature. We remember our wins more vividly than our losses. We're under quota pressure to call deals up rather than down. So the forecast drifts upward, quarter after quarter, slowly coming loose from what the deals actually look like.
Three reasons forecasts go wrong
When I look at why a forecast missed, it almost always comes back to one of three things.
Get pipeline intelligence insights weekly
Join sales leaders learning to turn CRM data into pipeline confidence. No spam. Unsubscribe anytime.
The first is gut feel dressed up as confidence. A deal gets a high probability not because it resembles deals you've won, but because the rep likes the prospect and the conversations feel warm. Warmth is not a buying signal. Plenty of deals feel great and never close. The forecast is treating a feeling as a fact.
The second is optimism bias, and it's baked in at every level. The rep is optimistic about the deal. The manager is optimistic about the rep. The VP is optimistic about the quarter. Each layer rounds up a little, and by the time the number reaches the board it's a tower of small hopes stacked on top of each other. Nobody lied. The system just has nothing pushing the number back down.
The third is the big one. Most teams treat all pipeline as equal. They weight deals by stage and value, so a £100K deal at proposal counts more than a £40K deal at discovery. Fair enough. But stage and value tell you nothing about whether this deal looks like the ones you actually win. Two deals can sit at the same stage, same value, and have wildly different odds of closing. The forecast can't see the difference, so it treats them the same. That's where it breaks.
What good forecast accuracy actually looks like
People ask me what a good accuracy number is. The honest answer is that if you're consistently landing within about 10% of your called number, you're doing well. Most teams aren't close, and the ones who think they are usually haven't measured it properly. They've just gotten used to the miss.
But chasing an accuracy percentage misses the point. Forecast accuracy isn't really a forecasting problem. It's a pipeline quality problem wearing a forecasting mask. This is the gap Pipeline Intelligence is built to close. If your pipeline is full of deals nobody can objectively assess, no forecasting technique on earth will save the number. You can't accurately predict the outcome of something you can't measure.
So the fix isn't a better spreadsheet or a tighter commit process. It's giving every deal an objective, evidence-based measure of how likely it is to close, before anyone calls it.
The fix: forecast on evidence, not optimism
Here's the shift. Instead of asking how you feel about a deal, ask how closely it resembles the deals you've already won.
That's a question your own data can answer. Every deal you've ever closed-won is a record of what a winnable deal looks like at your company. The industry. The size. The buying behaviour. The number of stakeholders. The speed it moved through each stage. Your wins are a pattern, and every open deal can be measured against that pattern.
Here's the detail most people get wrong. You have to compare deals stage-relative. A deal sitting at proposal should be benchmarked against deals that won from the proposal stage, not against your closed-won deals as a whole. If your winners typically had three engaged stakeholders by the time they reached proposal and this one has a single contact, that isn't a hunch. It's a signal you can act on, weeks before the forecast call where it would otherwise blow up in your face.
Do that across the whole pipeline and the forecast changes character completely. It stops being a story your reps tell and becomes a number grounded in what actually closes. The CRO walks into the board meeting and says "64% of our pipeline matches our winning pattern this quarter" instead of "we think we'll hit it, but I can't be sure."
What a revenue leader can do this week
You don't need to buy anything to start. Three steps, all doable now.
First, pull your last 50 to 100 closed-won deals and actually look at them. What do they have in common? Industry, size, region, deal size, sales cycle length, lead source. Write the pattern down. You'll probably find it isn't one pattern but two or three distinct ones.
Second, score your current pipeline against that pattern, even roughly. A manual high, medium or low against your real win profile beats stage-and-gut every time. You'll spot the deals everyone's been optimistic about that look nothing like a winner.
Third, change the question you ask in the pipeline review. Ban "how do you feel about it" and replace it with "what about this deal matches the deals we've won, and what's missing?" Watch how fast the conversation gets honest. (If your pipeline reviews have quietly stopped working, here's why that happens.)
That alone will tighten your forecast. The patterns are already sitting in your CRM. The only thing missing is the discipline to forecast on them instead of on hope.
Telepath Pro builds your winning pattern from your closed-won data and scores every open deal against it, stage by stage, so your forecast is grounded in evidence rather than optimism. You can run a free ICP analysis on your own won deals in three minutes. No signup, no credit card.