Budget vs. Actual Variance Formula: Dollar and Percent, with Sign Rules

The budget vs. actual variance formula is Dollar variance = Actual − Budget, and Percentage variance = (Actual − Budget) ÷ |Budget| × 100. For revenue and profit lines, a positive result is favorable. For expense lines, a positive result is unfavorable, because spending more than plan reduces profit. Every variance should carry both its number and an F (favorable) or U (unfavorable) label so the reader never has to work out the sign.

This post covers the formulas and the conventions around them: sign rules, edge cases in the percentage formula, how to split a cost variance into volume and rate, and how actual vs. forecast differs from budget vs. actual. For the monthly process of investigating variances and writing the narrative, see our guide to variance analysis.

The Two Formulas

Dollar variance = Actual − Budget

Percentage variance = (Actual − Budget) ÷ |Budget| × 100

The dollar variance tells you the size of the miss in money. The percentage variance tells you the size relative to the plan, which is what lets you compare a $25,000 miss on a $520,000 payroll line with a $22,000 miss on a $90,000 freight line. The first is 4.8% off plan. The second is 24.4% off plan. Same rough dollar size, very different signals.

The vertical bars around Budget mean absolute value. It matters only when the budget figure is negative, such as a planned operating loss, a budgeted FX loss, or a contra-revenue line. Dividing by a negative budget flips the sign of the percentage and makes an improvement look like a deterioration. Using the absolute value keeps the direction of the percentage consistent with the direction of the dollar variance.

Example: a business unit budgeted an operating loss of $200,000 and actually lost $150,000. Dollar variance = −150,000 − (−200,000) = +$50,000. Percentage variance = +50,000 ÷ 200,000 = +25.0%. The loss was smaller than planned, so the variance is favorable, and the positive sign says so. Without the absolute value, the formula would return −25%.

Favorable and Unfavorable: The Sign Rules

A variance is favorable when it helps profit and unfavorable when it hurts profit. AccountingCoach puts it in those terms: actual revenues above budget are good for profit, and actual expenses below budget are good for profit.

Applied to Actual − Budget:

There are two common ways to present this, and either works if you are consistent.

Convention A: raw sign plus a label. Always compute Actual − Budget. Expense overruns show as positive numbers with a U label. This is the simplest to build in a spreadsheet, and it matches the way the general ledger stores expense balances.

Convention B: profit-signed. Compute Actual − Budget for revenue lines and Budget − Actual for expense lines, so that a positive number always means favorable and a negative number always means unfavorable. Many reporting packages present it this way, showing unfavorable variances in parentheses. AccountingCoach's note on negative variances describes that presentation: negative variances are unfavorable to profit and are shown in parentheses.

Convention B reads faster for a CEO or a board, because the sign alone tells the story. Convention A is easier to audit, because every cell uses the same formula. Whichever you choose, write it in the footnote of the report and apply it to every line, including subtotals. The most common error in a variance report is a subtotal that uses one convention while its components use the other.

Worked Example (Illustrative)

The table below is an illustrative month for a mid-market company, in thousands of dollars. The figures are assumptions chosen to show the mechanics, not data from a real business. It uses Convention A: Actual − Budget, with an F or U label.

Check the arithmetic. Revenue: 4,750 − 5,000 = −250, and −250 ÷ 5,000 = −5.0%. Gross profit: 1,710 − 1,900 = −190, and −190 ÷ 1,900 = −10.0%. Operating expenses: 545 + 112 + 48 = 705 against 520 + 90 + 60 = 670. EBITDA: 1,710 − 705 = 1,005 against 1,900 − 670 = 1,230, a variance of −225, or −18.3%.

Two things in this table are worth noticing.

The percentage variance grows as you move down the P&L. A 5.0% revenue miss became a 10.0% gross profit miss and an 18.3% EBITDA miss. That is operating leverage at work: fixed costs do not shrink when revenue does. A lender testing a leverage or DSCR covenant cares about the EBITDA line, so a revenue variance that looks modest at the top can be a covenant question at the bottom.

Cost of sales shows as favorable, and that is misleading. It came in under budget because the company sold less. The next section shows how to separate that effect.

Splitting a Variance: Volume vs. Rate

A static budget compares actual costs with the costs planned for the original volume. When volume changes, variable costs should change too, so a static comparison mixes two effects: how much you sold, and how efficiently you produced it.

A flexible budget restates the variable cost at actual volume. In the illustrative budget, cost of sales was 62% of revenue (3,100 ÷ 5,000).

Flexed cost of sales = Budgeted cost rate × Actual revenue = 62% × 4,750 = 2,945

Now split the −60 static variance:

The static report said cost of sales was $60,000 favorable. The split says the company spent $95,000 more than it should have for the volume it actually shipped, and the lower volume hid it. That $95,000 is the number to investigate: material prices, scrap, overtime, or mix. The Corporate Finance Institute's variance analysis guide works through the standard-cost version of the same idea, splitting cost variances into price and quantity components.

Using a single cost rate is a simplification. In a real flex budget, each variable cost line gets its own driver: units for materials, labor hours for direct labor, shipments for freight. The arithmetic is the same.

Actual vs. Forecast vs. Budget vs. Actual

The formulas do not change when you compare against a forecast instead of a budget. What changes is the question you are answering.

Budget vs. actual compares results with the plan approved at the start of the year. It answers: are we delivering what we committed to? The budget usually does not move, so the variance accumulates. By the third quarter, a budget built the previous autumn may be describing a business that no longer exists.

Actual vs. forecast compares results with the most recent reforecast. It answers: is our current view of the business accurate? Forecast variances should be small. Large or one-directional forecast variances say the forecasting process has a problem, which matters as much as the operating result itself.

Extending the illustrative example with a forecast prepared mid-month:

Check: 4,750 − 4,800 = −50, and −50 ÷ 4,800 = −1.0%. 1,005 − 1,040 = −35, and −35 ÷ 1,040 = −3.4%.

Read together, the two columns say different things. The company is well behind its annual plan, which is a performance conversation. The forecast was close, which means management saw the shortfall coming. For a lender or a sponsor, that second fact carries real weight. A company that misses its budget but forecasts accurately is a company whose numbers can be relied on. A company that misses both is one where the lender will start asking for more frequent reporting.

Show both columns in a monthly package. If you only have room for one, a board usually wants budget vs. actual and a lender usually wants actual vs. forecast plus the updated full-year outlook. Our post on driver-based forecasting vs. line-item budgeting covers how to build a forecast that holds up to that comparison.

Edge Cases in the Percentage Formula

Budget of zero. The percentage is undefined. Show the dollar variance and mark the percentage "n/m" (not meaningful). Do not show a percentage of infinity or a spreadsheet error.

Very small budgets. A $3,000 actual against a $500 budget is +500%. That is arithmetically correct and practically useless. Set a materiality threshold, such as a dollar floor and a percentage, and only comment on variances that clear both.

Sign changes. When the budget is a profit and the actual is a loss, or the reverse, the percentage can exceed 100% and is hard to read. Lead with the dollar variance and describe the swing in words.

Year-to-date vs. month. Compute YTD variance from YTD actual and YTD budget. Do not add up monthly percentages; percentages are not additive.

Putting the Formulas to Work

A clean variance report starts with the formula choices above written down: which convention, how negative budgets are handled, which lines get flexed, and the materiality threshold for commentary. With those fixed, the report becomes a mechanical output of the close, and the analysis time goes to explaining the variances instead of arguing about their signs. The explaining part is covered in Why Variance Analysis Matters and the step-by-step variance analysis guide.

Want a repeatable checklist for the monthly review? Get our Monthly Financial Analysis Checklist →