Revenue Variance (Forecast Variance)

The Number Your CFO Watches Most Closely

Revenue variance is the difference between what you said would close and what actually closed. Expressed as a dollar amount or percentage, it is the single metric a CFO uses to judge whether the revenue team's predictions are worth anything. A team that forecasts $2M and delivers $1.6M has a -20% variance. A team that forecasts $2M and delivers $2.1M has a +5% variance. Both directions matter — chronic over-forecasting erodes trust, and chronic under-forecasting creates resource allocation problems.

Why Variance Matters More Than Growth Rate

A company growing 40% with plus-or-minus 5% variance is worth significantly more than one growing 50% with plus-or-minus 20% swings. Significant forecast variance results in material ARR multiple compression during diligence. Investors model future cash flows based on the assumption that the revenue team can predict its own outcomes. When variance is wide, that assumption breaks, and the risk premium goes up. Revenue predictability is worth more than growth in every board conversation and every investor model.

Measuring Variance Correctly

Track variance at three levels: total company, by segment, and by rep.

Level	What It Reveals	Cadence
Total company	Overall forecast reliability	Monthly and quarterly
By segment	Which motions are predictable and which are not	Quarterly
By rep	Individual forecasting accuracy and coaching needs	Monthly

Blended company variance can mask segment-level problems. You might be plus-or-minus 5% in aggregate because your enterprise team over-forecasts by 15% and your mid-market team under-forecasts by 10%. That is not predictability — that is offsetting errors.

The Root Causes of Variance

Most variance comes from three sources: optimistic deal staging, slippage that is not flagged early enough, and inconsistent forecast methodology. Reps who categorize deals as "commit" based on feel rather than data produce unreliable forecasts. Deals that slip from one quarter to the next without early warning inflate the current quarter and deflate the next. And when every rep uses a different standard for what "commit" means, the rolled-up forecast is noise.

Reducing Variance Systematically

Fix methodology first, then calibration. Standardize what each forecast category means — commit, best case, upside — with specific criteria tied to deal signals, not rep judgment. Require that every commit-level deal has a confirmed close date, an identified economic buyer, and recent activity within 14 days. Then track variance by rep over time and use the pattern to calibrate: if a rep consistently over-forecasts by 15%, adjust their commits accordingly until their accuracy improves.