Why is our sales forecast always wrong?

The forecast meeting goes well. The number feels achievable. The team knows which deals are expected to land.

Then the end of the month or quarter arrives. Deals slip. Values shrink. A supposedly committed opportunity disappears. Something unexpected closes. And once again, forecast and actual revenue don’t match.

The immediate reaction is often to blame the CRM, the forecast process, the seller, data hygiene or the forecasting tool. Any of those may contribute. But repeated forecast misses usually point further upstream.

The forecast may not be the problem. It may simply be where the problem becomes visible.

Forecast error is rarely one problem. It is usually a pattern of assumptions that keep proving wrong.

Check five things first.

If your forecast keeps missing, check these five things.

Before discussing probability or forecast category, validate the opportunity itself.

• Is there a genuine customer problem, and evidence they intend to act?
• Do we understand why they need to act now?
• Are the relevant stakeholders engaged?
• Is there a clear decision process?
• Is there a compelling commercial reason to move?
• Is the opportunity still active in the customer’s world?

A forecast cannot be more credible than the opportunities underneath it.

An opportunity should not become forecastable simply because it has been open a long time, or because a seller is optimistic. If the pipeline itself is unreliable, start by understanding how much of the reported opportunity is genuinely credible.

Enough pipeline but still missing target?→

Most stage inflation comes from counting effort as progress. The two are easy to separate once you look for the difference.

Seller activity shows what happened. Buyer evidence shows whether the deal progressed.

A proposal being sent is seller activity. A customer confirming that procurement starts next Tuesday and the decision meeting is scheduled is buyer evidence. This does not require a new methodology; the distinction works inside whatever sales process you already run.

The useful question is where the close date actually came from.

A close date is not a forecast because somebody entered a date into the CRM.

Look for the customer event driving timing, agreed milestones, procurement steps, legal or commercial dependencies, an implementation deadline and a mutually agreed next action. Then ask what has to happen between today and the proposed close date for the deal genuinely to land. If the sequence is unrealistic, the date probably is too.

Forecast accuracy is not only about whether a deal closes. It is also about what it closes for. Compare forecast value with booked value and ask whether early values are routinely optimistic, whether values shrink late in the cycle, whether unconfirmed expansion is included, whether discounts only appear at the end, and whether particular sellers consistently overstate value.

A deal forecast at £200k that closes at £120k is still an £80k forecast miss, even though it was won.

Winning the deal doesn’t automatically make the forecast accurate.

Seller judgement matters. Experienced sellers and managers carry context that no field in the CRM holds. The point is not to discount that judgement, but to test it.

Judgement plus evidence is stronger than judgement alone.

Confidence is useful context. Evidence should carry the forecast.

“The forecast was wrong” isn’t a diagnosis.

Forecast misses are not all the same event, and the type of miss narrows the investigation considerably. Categorise them.

The pattern of the miss tells you where to investigate. High slippage points towards timing assumptions. Losses from late-stage forecast categories point towards qualification or opportunity strategy. Repeated shrinkage suggests weak value estimation. Unsupported deals suggest optimism, or weak manager inspection. Surprise revenue suggests the process is not capturing real progression.

Slippage is data. Use it.

The common failure is administrative: the slipped deal moves into the next period and the question of why is never asked. Instead, look at how often deals slip, how many times, from which stages, which sellers carry the most slipped value, which reasons recur, and what was believed before the deal moved.

Recurring causes tend to be seller-created dates, unclear customer urgency, weak stakeholder access, unknown procurement steps, unrealistic next actions or thin qualification. Avoid universal slippage thresholds — different sales motions behave differently — and compare against your own history instead.

Every slipped deal should teach you something about your forecasting assumptions.

Do stages mean the same thing to everyone?

If one seller’s stage four is another seller’s stage two, weighted forecasting stops being arithmetic and becomes an average of different opinions. Ask whether entry and exit criteria are clear, whether two managers inspecting the same deal would reach a similar conclusion, whether progression requires buyer evidence, and whether opportunities can move backwards when the evidence changes.

If the stage isn’t consistent, the probability attached to it isn’t consistent either.

This rarely requires a new methodology. Clearer definitions and consistent manager reinforcement are often enough.

Forecast meetings should inspect evidence, not negotiate optimism.

The pattern to avoid is familiar: “you said £500k last week, why is it £420k now?” It creates pressure to defend a number rather than examine a deal, and the predictable response is a number that stops moving whether or not reality does.

• What changed?
• Which customer action supports the close date?
• What new evidence increased or reduced confidence?
• What still needs to happen, and what could stop the deal?
• Has the value changed?
• Should this still be in the forecast at all?

Who consistently calls the number well?

Team-level forecast accuracy hides individual patterns. Look at forecast versus actual by seller, then at slippage, value shrinkage, unsupported deals and surprise revenue the same way, and by manager as well as by seller.

Where someone consistently forecasts accurately, the question worth answering is what they do differently. Usually they qualify harder, understand customer timing better, update close dates earlier, carry fewer weak opportunities, involve managers sooner and rely on stronger buyer evidence.

Forecasting can expose capability and management gaps, not just data problems.

A practical sales forecast diagnostic.

If the forecast keeps missing, use the pattern to identify which assumption repeatedly proves wrong.

Will better forecasting software solve it?

Forecasting technology genuinely improves visibility, aggregation, historical comparison, pattern recognition and opportunity signals. What it cannot do is manufacture customer evidence. Where qualification is weak, values are unrealistic, dates are optimistic or stages mean different things to different people, better technology processes weak assumptions more efficiently.

Improve the inputs and the process before assuming the model is the constraint.

What should I do if our sales forecast is consistently inaccurate?

1. 1Measure the miss. Compare forecast with actual, period by period.
2. 2Categorise it. Slipped, lost, shrunk, unsupported or surprise.
3. 3Identify the assumption. Quality, stage, timing, value or evidence.
4. 4Find the pattern. By seller, manager, segment, stage and opportunity type.
5. 5Tighten the weakest assumption. Don’t rebuild the whole forecast process if one factor causes most of the error.
6. 6Measure again. Did slippage fall, forecast versus actual improve, unsupported deals reduce, values become more accurate and seller calls more consistent?

The goal isn’t perfection. It’s predictability.

And if the shortfall runs wider than the forecast, the broader diagnostic starts one level up:

Why is my sales team missing target?→

A forecast is only as credible as the opportunities underneath it.

Forecasting isn’t about predicting the future perfectly. It’s about understanding the present accurately enough to make a credible commercial call.

When the forecast repeatedly misses, don’t only ask why the number was wrong. Ask which assumption underneath the number was wrong: quality, stage, timing, value or evidence. Then fix the weakest one. Inspect the evidence, understand the pattern, improve the process, measure the accuracy.

Better forecasting starts upstream.

Frequently asked questions.

James King

Founder, Revenue Performance Partners

James works across sales performance, enablement and commercial operations, helping organisations diagnose performance gaps and turn them into measurable improvement.

More about James→