How to Calculate Sales Forecast Accuracy (And Why Most Teams Do It Wrong)
How to calculate sales forecast accuracy is a simpler question than most RevOps teams make it. The answer only matters if you are measuring the right thing. A lot of companies celebrate hitting 100% of quota in Q3 while their forecast was off by 40%. Those are two different problems, and conflating them is how you end up flying blind into Q4.
The standard formula is straightforward. Divide your actual revenue by your forecasted revenue, then multiply by 100 to get a percentage. If you forecasted $500,000 and closed $430,000, your forecast accuracy is 86%. Most organizations consider anything above 90% acceptable. Anything below 80% consistently means your pipeline data, your stage definitions, or your reps are telling you a story that is not true.
How to Calculate Sales Forecast Accuracy: The Core Formula
The math itself takes thirty seconds. The hard part is deciding what numbers go into it.
You need two clean inputs: the forecast number you committed to at the start of the period, and the actual closed-won revenue at the end. The forecast has to be locked. If you let reps or managers revise the number on day 28 of a 30-day month, you are measuring how good people are at editing a spreadsheet after the fact.
Lock the forecast on day one. Measure actuals on the last day. Run the formula. That number is your accuracy rate for the period.
Where this gets more useful is when you break it down. Forecast accuracy at the company level tells you something. Forecast accuracy by rep, by segment, or by deal source tells you a lot more. A VP of Sales who is consistently 95% accurate is hiding something if two of her reps are at 60% and two are at 130%. The average is fine. The variance is the real data.
What Kills Forecast Accuracy Before You Even Run the Numbers
Bad stage hygiene is the usual culprit. If your CRM has deals sitting in Proposal Sent for six weeks because nobody updated them, your forecast model is building on garbage. Reps stop updating stages after the third call on a Friday, and then managers pull a pipeline report on Monday and treat those numbers as real. The formula cannot save you from that.
The second killer is inconsistent close date discipline. A deal with a close date that has been pushed four times is a hope, not a Q2 deal. If your RevOps setup has no rules around how many times a close date can slip before a deal gets flagged or moved out of the forecast, you are letting wishful thinking sit in your committed column.
Third: using one forecast methodology when you need two or three. Weighted pipeline works fine for early-stage visibility. It is a terrible way to call the current quarter. For near-term commits, you want rep-called forecasts validated against historical close rates by stage and rep. For longer horizons, weighted pipeline or AI-assisted models in HubSpot or your CRM of choice fill the gap better.
How to Actually Improve the Number Over Time
Run accuracy reporting weekly, not just at quarter end. The point is to catch drift early. If you are three weeks into a quarter and your forecast accuracy on deals that should have closed already is 65%, act on it now instead of writing it up in a postmortem.
Hold a short forecast review every week, 20 minutes. Its job is to pressure-test the assumptions, not to update the number. Which deals moved? Which stalled? Did the close date change, and why? Do this consistently for two or three quarters and your managers get genuinely better at reading their pipeline. The accuracy rate climbs because the inputs got honest, not because the formula changed.
AI automation helps here too, particularly for flagging deals where the activity data contradicts the stage. If a deal is marked Negotiation but the last email was 22 days ago and there has been no meeting booked, an automated alert beats hoping a rep volunteers that information in a call review.
Frequently Asked Questions
What is a good sales forecast accuracy rate? Above 90% is the benchmark most revenue leaders use for a committed forecast. For a pipeline-level forecast covering a full quarter, 80 to 85% is more realistic, especially in deal cycles longer than 60 days. The target matters less than the trend. If you were at 72% last quarter and 79% this quarter, you are moving in the right direction.
Should you measure forecast accuracy by revenue or by deal count? Revenue, almost always. Deal count accuracy can look fine while revenue accuracy is terrible if one large deal swings the quarter. Track both if you have a highly variable deal size distribution, but revenue is the number that matters to the business.
How often should we recalculate forecast accuracy? Lock your forecast at the start of each period and measure accuracy at the end. Do not recalculate mid-period using a revised forecast. That defeats the purpose. Weekly pipeline reviews are separate from the accuracy measurement itself.
If your forecast numbers keep surprising you at quarter end, the problem is almost never the formula. Let us look at your pipeline data and your process, and figure out where the honesty breaks down. A fractional GTM leadership engagement can put someone in the room who has fixed this before.
