Variance

In finance and financial planning and analysis (FP&A), variance is the difference between a budgeted or forecast number and the actual number a business posts for the same line item over the same period. Set a marketing budget of a certain size, spend a different amount, and the gap between the two is the variance. A variance is called favorable when the actual result is better for the business than the plan — lower cost, higher revenue — and unfavorable when it is worse. The sign alone is not the story, though. The purpose of variance is diagnostic: it flags where reality drifted from expectation so someone can ask why, and then decide whether the plan, the execution, or the assumptions need to change. A budget without variance analysis is a wish; variance is what turns it into a control loop.

Variance rarely arrives as one clean number. Analysts break a total variance into its parts to locate the cause. A sales variance splits into a volume component (you sold more or fewer units than planned) and a price or rate component (each unit sold for more or less than planned). A cost variance splits similarly into how much you used and what you paid per unit. This decomposition matters because a favorable-looking total can hide two offsetting problems, and a small total can mask a large volume miss cancelled by a lucky price. Good variance analysis does not stop at the headline gap. It walks each line back to a driver a manager can act on — a pricing decision, a supplier change, a demand shift, a stale forecast assumption — so the next budget is sharper than the last.

The word variance carries a second, unrelated meaning in statistics, and confusing the two is a common trap. In statistics, variance measures dispersion — the average of the squared distances of each value from the mean of a data set. It answers how spread out the numbers are, and its square root is the standard deviation. A marketing team analyzing conversion-rate stability across landing pages uses the statistical variance; a marketing team explaining why the media budget overspent uses the budget variance. Same word, different math, different question. This page leads with the finance sense because that is the one you meet in budgets, P&Ls, and board reviews, but it is worth naming the statistical sense so the term is never read in the wrong register.

Keeping the senses apart is more than pedantry. The finance variance is a single subtraction — plan minus actual — with a direction that is favorable or unfavorable to the business. The statistical variance is always a non-negative number describing a whole distribution, with no notion of favorable or unfavorable. You would never say a data set's dispersion was favorable, and you would never describe a budget miss in squared units. When someone in a meeting says variance, the surrounding words tell you which they mean: budget, actual, forecast, favorable, and unfavorable point to finance; mean, dispersion, standard deviation, and distribution point to statistics. Read the context, pick the right sense, and the arithmetic follows without ambiguity.

Use variance as a running control, not a post-mortem. Compare actuals to plan on a regular cadence — monthly for most budgets, sometimes weekly for volatile media spend — and set a threshold below which small gaps are noise and above which a variance earns an explanation. Decompose every material variance into its drivers rather than reporting one blended figure, because volume and price move for different reasons and demand different responses. Write the why next to the number, in plain language a decision-maker can act on. A variance report that lists gaps without causes is a spreadsheet; a variance report that names causes is a management tool.

The discipline also means treating the plan itself as testable. If a line shows the same unfavorable variance quarter after quarter, the problem may be an unrealistic budget rather than poor execution, and the honest fix is to reforecast, not to keep flagging the miss. Guard against two failure modes: celebrating a favorable total that hides an offsetting problem underneath, and drowning managers in tiny variances that carry no signal. Set thresholds, decompose, explain, and feed what you learn back into the next budget. Done this way, variance analysis tightens the loop between planning and reality instead of merely recording the distance between them.

Variance — the gap between a budgeted figure and the actual result — is analyzed in finance to explain why performance beat or missed plan, and is distinct from the statistical measure of dispersion that shares its name.

Etymology: source.

Search interest for this term over the last five years:

View interest-over-time on Google Trends →

Areas of marketing where variance is a core concern: