Principle
Every individual rep's forecast can be defensible, cautious, and honestly submitted, and the team total can still be wrong by a wide margin. Forecast accuracy is not simply a function of individual honesty. It is a function of how correlated the underlying assumptions are across the whole team, and correlated assumptions do not cancel each other out. They compound.
Behaviour
A standard sales forecast rollup sums each rep's individually assessed Commit and Best Case figures into a single team or organisation-wide number, which is then presented to the board with an implicit assumption of statistical independence: that if one rep's number is a little optimistic, another rep's caution will roughly offset it, and the aggregate will land somewhere sensible.
This assumption fails whenever the individual forecasts share a common underlying driver, which in practice is most of the time. If every rep's Q4 number depends on the same macro condition, such as customers finalising annual budgets before a fiscal year-end, or every rep's forecast assumes the same new product feature ships on schedule, the individual forecasts are not independent bets. They are the same bet, repeated fifteen times, and if the shared assumption is wrong, every rep misses simultaneously rather than the misses averaging out across the team as the rollup methodology implicitly assumes.
They are not independent bets. They are the same bet, repeated fifteen times, and if the shared assumption is wrong, every rep misses simultaneously.
Evidence
Interview five to eight reps individually about the specific assumptions underlying their largest forecasted deal for the period, without revealing what other reps have said. Look for a shared, unstated dependency: a common product release date, a common seasonal buying pattern, a common competitor situation, or a common macro assumption about budget availability.
If three or more reps' numbers depend on the same specific external event occurring on schedule, the team forecast has dramatically less genuine diversification than the rollup arithmetic implies, and the true confidence interval around the total is far wider than the individual confidence intervals would suggest in isolation.
A useful supplementary test: ask the CRO directly what would need to happen for the entire team forecast to miss simultaneously, rather than for one or two individual deals to slip. A CRO who can answer this specifically, naming the shared dependency, is managing correlation risk consciously. A CRO who has never considered the question is managing a rollup number without understanding its actual structure.
Psychology
Nobody in the forecasting chain is being dishonest. Each rep genuinely believes their own number, and each number may well be an entirely fair, well-reasoned individual assessment given what that rep knows. The distortion happens at the aggregation layer, not the individual layer, and it happens because forecast rollup tools are built to sum numbers, not to test whether the assumptions behind those numbers are actually independent of one another.
CROs reinforce this blind spot because a rollup that simply sums individual numbers is far easier to produce, explain, and defend to a board than one that attempts to model correlation across the team, which requires a genuinely different and more sophisticated kind of forecasting discipline that most sales organisations have never built. The result is a forecasting culture that is highly rigorous at the individual deal level and almost entirely unexamined at the level where the numbers that actually matter to the board are produced.
Commercial Risk
Correlated forecast risk is invisible in a standard rollup and only becomes visible when the shared assumption actually breaks, at which point the entire team misses simultaneously rather than the expected handful of individual deals slipping in an uncorrelated, easily absorbed pattern. This produces forecast misses that are larger and more sudden than historical forecast accuracy would suggest is plausible, precisely because historical accuracy was measured in periods where the shared assumption happened to hold.
For an investor, this means historical forecast accuracy is a weaker predictor of future forecast reliability than it appears, specifically in any period where the underlying market or product conditions differ meaningfully from the conditions of the historical track record being cited.
Investment Committee Note