Variance Analysis

Your Q2 gross margin came in 4.2 points below forecast, and the CFO wants to know why before the board call in three hours. Finance has the variances by line. What it doesn't have yet is whether the miss was volume (sold less than expected), price (sold at lower prices), or mix (sold more of the lower-margin products). Those are three different problems with three different responses. Variance analysis is the process of decomposing the difference between a planned or forecasted financial result and the actual result into its component causes, so you can see which factors drove the outcome and by how much. A simple budget-vs-actual comparison tells you that something went wrong. Variance analysis tells you what went wrong, and that determines whether the right response is a sales strategy change, a pricing adjustment, a cost reduction, or a product mix change. The method applies across the income statement: revenue variances break down into volume, price, and mix components; cost variances break down into rate (what was paid per unit) and usage (how much was consumed); and efficiency variances in manufacturing separate the rate of resource consumption from the volume of output. The value of variance analysis is the discipline it imposes: a structured conversation about causes instead of accepting the gap as an unexplained result.

How it works

A revenue or gross margin miss can come from three independent causes, and each calls for a different management response. A volume variance means the company sold fewer units or generated less revenue volume than planned; the explanation sits in the sales pipeline, market demand, or execution. A price variance means the company sold at a lower average price than planned, which might reflect discounting, competitive pressure, or a shift in customer negotiating power. A mix variance means the company sold a different combination of products or customer segments than planned: more of the lower-margin items and fewer of the higher-margin ones, even if total volume hit target. The arithmetic: total revenue variance equals the volume variance plus the price variance plus the mix variance. Each component is calculated separately and then summed. Without this decomposition, a gross margin miss looks like one problem. With it, Finance can tell the CFO that $0.8M of the miss was volume (fewer deals closed), $0.3M was price (higher average discount), and $0.2M was mix (more SMB revenue and less enterprise), and each of those has a different owner and a different fix. The decomposition doesn't change the outcome; it changes the quality of the conversation about what to do next.

When variance analysis stops being useful

Variance analysis turns into storytelling instead of diagnosis when it is done after the quarter has closed with the unstated goal of explaining away the miss. The usual symptom is a variance explanation made of 'timing' and 'one-time items' with no evidence behind either label. When Finance calls a revenue shortfall 'timing' but can't say which deals slipped, to which period, and with what probability of closing, it has written a narrative, not an analysis. The second failure is over-decomposition: splitting variances into so many sub-components that the analysis becomes unreadable and the main drivers get buried. A useful variance analysis surfaces the two or three largest contributors to the gap and explains them specifically enough to support a decision. The third failure is doing variance analysis only after the quarter closes, instead of during the quarter when corrective action is still possible. Midquarter variance tracking, which compares actual performance to the period forecast instead of the annual budget, gives Finance and leadership a chance to adjust before the period ends.

How platforms handle it

FP&A platforms routinely feature variance analysis in sales demos. In a demo, 'drill-down' usually means clicking a variance number to see the transactions or sub-categories behind it. In practice, the quality of that drill-down depends entirely on how well the underlying data is structured. If the general ledger doesn't have consistent department tags and cost categories, drilling down produces a list of transactions with no meaningful grouping. If the revenue data doesn't carry deal-level attributes like product line, customer segment, and sales rep, the volume-price-mix decomposition can't be done inside the platform and has to happen in a separate spreadsheet. When evaluating FP&A and forecasting tools, bring a real variance scenario from a recent period and ask the platform to decompose it. Can it split a revenue miss into volume and price components using existing data? Can it show which cost center drove the largest opex variance without manual configuration? The answers show whether the drill-down is real analysis or a formatted list of ledger entries.

Questions to ask when evaluating a variance analysis process

• Are revenue variances decomposed into volume, price, and mix components, or reported as a single net number?
• Is variance analysis performed during the quarter (against a rolling forecast) or only after close (against the original budget)?
• Are variance explanations specific, naming the deal, cost, or driver, or do they stay at the level of 'timing' and 'one-time items'?
• Does the FP&A platform support drill-down to the transaction or deal level with consistent dimension tags?
• Is there a defined threshold for which variances require a written explanation vs which can be acknowledged without commentary?
• Is each variance explanation tied to a corrective action and an owner, or does it end with the explanation?

Common mistakes

The most persistent variance analysis mistake is calling a gap 'timing' without evidence. 'Timing' is a legitimate driver when a deal slipped from Q3 to Q4 and Finance can show the specific deal, its new expected close date, and its probability of closing. It becomes a shield when it means 'we don't know why the number was low and we'd like to move on.' Finance teams that accept 'timing' without checking it against specific deals or invoices are covering for a miss, not diagnosing it. The second common mistake is running variance analysis only against the annual budget instead of the most recent forecast. If the annual budget was set 10 months ago and conditions have changed significantly, the budget variance tells Finance how far the world has moved from last October's assumptions, not whether Finance correctly anticipated the current quarter. Variance against the 90-day-prior forecast is a much better test of forecasting quality. The third mistake is not assigning owners to variance explanations. Once the gross margin miss has been decomposed into volume, price, and mix, each component needs a named owner (sales leadership for volume, revenue operations for price, product management for mix) who is responsible for the corrective action.

Not statistical variance

Search results mix this with variance in statistics and with analysis of variance, which tests whether group means differ. Neither is what finance means. Budget variance analysis compares one period's actuals against a plan and explains the gap by driver. A favourable variance is not automatically good news: revenue ahead of plan because a deal closed a quarter early is a timing variance that reverses next period.

Comparing planning platforms

Planning platforms separate on who can maintain the model after go-live, which is a staffing question as much as a software one. Once variance analysis is settled as a requirement, read the Forecasting Software category guide for how the market splits, then weigh the Anaplan profile against the Pigment profile on that requirement rather than on the feature list. Our Anaplan pricing breakdown records what the vendor publishes and what is quote-based. The Anaplan against Pigment comparison sets two of them side by side on the same criteria. Before you book demos, put your own numbers through the forecasting ROI calculator so the business case is written down before a vendor writes it for you.

Frequently asked questions

Related terms

Budget vs Actual Variance, Capital Expenditure (CapEx), and Cash Flow Forecasting come up in the same workflow as variance analysis, and the rest of the planning platforms cluster is indexed in the finance operations glossary.