Sales Management & Coaching
Sales Forecasting Accuracy
Most forecasting problems get treated as a data problem: better stages, better fields, better tooling. The forecast is usually wrong for a much simpler reason. It's built from what sellers are willing to say out loud.
By Shane McGrath · Published 24 August 2026
Every leader who's chased forecast accuracy has tried the same three fixes. Tighter stage definitions. Mandatory fields. A weighted model. The forecast improves for a quarter, then drifts back.
A forecast is a by-product, not an input
It's the output of how well deals are understood. If nobody knows what the buyer is actually trying to change or who has to approve it, no model fixes that. You're averaging guesses more precisely.
Why close dates are optimistic
Because the close date is usually the seller's hope, mapped onto the end of a quarter, and then defended. Ask where a date came from and you'll get one of two answers: the buyer said something specific about their own timeline, or the seller needed the deal in this quarter. Only one of those is a forecast.
- Ask what the buyer has to do internally before they can sign, and how long that has taken them before.
- Treat any date with no buyer-side event behind it as unforecastable.
- Track slipped deals by seller and by reason, not just by amount.
- Separate "this will close" from "I want this to close" and make it safe to say the second.
Pressure makes it worse
A manager who reacts badly to a deal moving out teaches sellers to move it out later, quietly, when there's no time left to help. Forecast accuracy is downstream of how safe it is to be honest, and most teams underrate how quickly sellers learn what's safe.
Where accuracy actually comes from
From evidence gathered in the deal, surfaced in a deal review that rewards honesty, by managers who coach rather than interrogate. The forecast then becomes a summary of things you already know, which is the only version that holds up.
The question worth asking your team
For each deal in this quarter's commit: what has the buyer done that they wouldn't have done if they weren't going to buy? If nobody can answer for a deal, it isn't a commit. It's a hope with a date on it.
FAQs
Common questions
- What's a realistic level of forecast accuracy for a B2B team?
- We won't put a number on it, because it depends on deal size, cycle length and how many deals make up the quarter. The more useful measure is whether the same deals slip repeatedly and whether anyone knew before it happened.
- Do weighted pipeline models help?
- They smooth the maths, they don't improve the information. A weighted number built on stages that sellers self-select is precise and still wrong. Fix what's known about the deals first.
- How do we get sellers to be honest about risk?
- Watch what happens the first time someone is. If naming a risk gets them interrogated, nobody does it twice. If it gets them help, it becomes normal within a couple of months.
Want this working inside your team?
Book a free sales review and we'll find the simplest change that moves your numbers.