Back to Blogs
FP&A Excellence

Why Sales Forecasts End Up Always Wrong

7 min read19 February 2026
Share:

Sales forecasts are never fully accurate. They tend toward optimism more often than caution. Yet the quarterly ritual continues. Rolling up pipeline, applying conversion rates, adjusting for management judgment, and presenting numbers that most people in the room quietly doubt.

The goal is not perfection. It is making the forecast useful despite its limitations.

The Pipeline Coverage Delusion

Sales leadership presents three times pipeline coverage for the quarter. Finance takes comfort in the ratio that surely a third will close. Except that half the pipeline should already be marked as lost as they are stale opportunities, exploratory conversations logged as deals, and aspirational entries with no real basis.

Pipeline coverage is a weak predictor of results unless pipeline quality is assessed honestly.

Shift the focus from coverage ratios to pipeline quality indicators. What proportion of the pipeline is in later stages? What is the historical conversion rate by stage and deal size? How long have these opportunities been open?

A 2x pipeline of late-stage, well-qualified deals is more reliable than a 4x pipeline of early-stage maybes. Most forecasting processes treat them the same.

Stage Duration Blindness

CRM systems show deals progressing through stages. Most forecasting models assume a deal in the proposal stage will close within a standard number of days because that is the average.

Averages are misleading here. A deal stuck in proposal stage for 90 days is very unlikely to close next month. A deal that moved from qualification to proposal in a week may close faster than the model suggests.

Track velocity as well as stage position. How long has a deal been in its current stage? How quickly did it move through earlier stages? Deals that progress quickly tend to close and deals that stall tend to die.

Build stage duration into your forecasting logic. A deal that arrived at proposal stage last week warrants different weighting than one that has been there for ten weeks.

The Rep Variability Problem

Forecasting models typically assume all sales representatives are equally reliable forecasters. In practice, one person's "committed" means something very different from another's.

Track forecast accuracy at the individual level and adjust probabilities based on historical reliability. Make it visible as it creates accountability and gives individuals useful feedback on their own judgment.

For new representatives without sufficient history, apply team averages until a personal track record develops.

The Forecast vs. Target Confusion

Sales forecasts are regularly distorted by their relationship to targets. If the target is a specific number, the forecast tends to land suspiciously close to it even when pipeline dynamics suggest something quite different.

This happens because forecasts are used for resource planning, incentive calculations, and external guidance. The forecast becomes a negotiation rather than a prediction.

Separate forecasting from target-setting deliberately. Make it clear that revising a forecast is updating a prediction, not declaring a shortfall against plan. Create the conditions where honest forecasting is genuinely rewarded over optimistic forecasting.

The best sales forecasting I have seen came from teams that kept a strict separation between "what we are committing to" and "what we genuinely expect to happen" and measured themselves on the accuracy of the latter.

Share:
SA

Shaheen Ali, FCCA

15+ years in FP&A, audit, and finance transformation. Based in Qatar.

Connect on LinkedIn

Stay close to the thinking

Practical perspective on FP&A, finance transformation, and AI. Published when there's something worth saying, not on a schedule.

Related articles

FP&A Excellence

Why Your Insights Aren't Landing

You've spent three days analyzing the data. You've found something genuinely important. You write it up, send it out, and nothing happens. Welcome to the insight graveyard.

7 min read·16 April 2026
Read
FP&A Excellence

The Variance Commentary Nobody Reads

Every month actuals finalised, variance reports produced, detailed commentary written. It took two days to write. Business unit leaders skim their own numbers. Nobody else engages with it.

6 min read·9 April 2026
Read
FP&A Excellence

Why Your Analysis Keeps Getting Ignored

You pulled the data, built the model, validated the output, and wrote a clear summary. You shared it with the leadership team. And then, nothing. What makes the difference?

8 min read·2 April 2026
Read