Forecast Variance

What Forecast Variance Is

Forecast variance is defined as the gap between what was predicted and what actually happened, measured in dollars and as a percentage of the forecast. It is the report card for your forecasting process. According to Clari (2024), the median B2B SaaS company has a quarterly forecast variance of 12-18%, meaning they miss their own prediction by roughly one-sixth of the total number every quarter.

Variance is distinct from forecast bias. Bias tells you the direction of the miss. Variance tells you the size. A company with low bias but high variance has forecasts that are right on average but wildly off in any given quarter. A company with high bias but low variance misses predictably in the same direction, which is easier to correct.

How is forecast variance analyzed?

Forecast Variance % = (Actual - Forecast) / Forecast x 100

Effective variance analysis breaks the total variance into components:

Component	What It Measures	Example
Volume variance	Difference in number of deals closed	Forecast: 50 deals. Actual: 42 deals. Volume variance: -16%
Price variance	Difference in average deal size	Forecast ASP: $85K. Actual ASP: $92K. Price variance: +8.2%
Timing variance	Deals that closed in a different period than forecast	6 deals slipped from Q1 to Q2
Win rate variance	Difference in conversion rate from forecast	Forecast win rate: 28%. Actual: 23%

Decomposing variance this way tells you what specifically went wrong. "We missed by $800K" is a fact. "We missed because 6 enterprise deals slipped and win rates dropped 5 points in mid-market" is actionable intelligence.

Why forecast variance matters for revenue teams

High forecast variance forces every downstream function into reactive mode. When the revenue number swings 15-20% from quarter to quarter, finance cannot commit to hiring plans. Marketing cannot allocate budget with confidence. The board cannot set expectations with investors. Revenue predictability is the foundation of organizational confidence, and variance is its enemy.

The compounding effect is significant. One quarter of 20% variance creates a one-time miss. Four consecutive quarters of 20% variance creates a pattern that damages credibility, delays fundraising, and forces the company into a defensive posture.

How to reduce forecast variance

- Decompose variance every quarter. Do not accept "we missed by 10%" as an answer. Break it into volume, price, timing, and win rate components. Each has a different root cause and a different fix. - Track variance at the segment and rep level. Total company variance hides where the problem originates. You may have 5% variance in mid-market and 25% variance in enterprise. Fix the enterprise forecasting process specifically. - Tighten forecast windows progressively. At the start of the quarter, allow 15% variance tolerance. By mid-quarter, tighten to 10%. By the last three weeks, the commit should be within 5% of the final number. This progressive tightening forces earlier deal inspection. - Use bottom-up forecasting for near-term and top-down for longer horizons. Each method has different variance profiles. Blending them produces more stable forecasts across time horizons.

Common mistakes with forecast variance

Treating all variance as equal. Missing by +10% (beating forecast) and missing by -10% (under-delivering) are not the same. Under-forecasting wastes resources that could have been invested. Over-forecasting breaks promises. Both are problems, but they have different root causes and different consequences. Not tracking variance trends. A single quarter of high variance could be an anomaly. Three consecutive quarters of increasing variance signals a deteriorating forecast process. Track the trend, not just the snapshot. If variance is growing, the underlying causes are getting worse, not better.