What a budget variance measures
A budget variance is the gap between what you said would happen and what did happen, on one line, for one period. It exists to answer a management question rather than an accounting one: where did the plan break, by how much, and who can do something about it. The dollar gap is arithmetic that a child could do. The value of a variance report lies entirely in the three things that surround it — the sign convention, the materiality rule, and the causal explanation.
Start with what the number is not. A variance is not a measure of performance on its own, because a budget is a forecast made months earlier, at an assumed volume, with assumed prices. If sales came in 20% above plan, every variable cost line will overspend, and every one of those overruns is the correct result of doing more business. That is why serious variance analysis flexes the budget to actual volume before judging anyone, a step covered in the sales volume and price variance calculator.
Nor is a variance the same as a forecast error, though the arithmetic is identical. A budget is a commitment used to authorise spending and to measure managers; a forecast is a best estimate updated as facts arrive. Most finance teams keep both, compare actual against both, and never confuse the two in the same column of the same report.
What the variance does give you is a triage list. On a fifty-line departmental report, four or five lines will carry almost all of the miss. Finding those four is the entire job of the monthly close review, and a percent-and-dollar materiality rule finds them in seconds.
The formula, and why the sign alone tells you nothing
Variance in dollars is actual minus budget. Keep that order and keep it everywhere: a positive number always means the actual figure landed above the plan. Reversing the subtraction on expense lines so that overruns come out negative is common, it feels intuitive, and it destroys your ability to add a column of variances together.
The percentage divides the dollar variance by the budget, not by the actual. Budget is the denominator because it is the fixed reference you agreed to; using actual as the denominator makes the same $10,000 miss produce two different percentages depending on which way it went. Take the absolute value of the budget in the denominator so that contra-revenue and credit-balance lines do not flip the sign of the percentage for no reason.
Then apply the account type. For revenue, other income and any line you want more of, actual above budget is favorable. For cost of sales, payroll, occupancy and every other expense, actual above budget is unfavorable. This is the single most common error in a hand-built variance sheet: a formula copied down a column labels a $50,000 revenue beat and a $50,000 payroll overrun the same way.
The clean fix is to report a second column that is already signed for profit: the effect on operating income. For revenue lines it equals the variance; for cost lines it equals the negative of the variance. Every line in that column can be summed, and the total ties to the operating income variance of the whole statement. When you can add your variance column and reconcile it to the bottom line, the report is right.
Worked example: a $250,000 six-month marketing budget
A marketing department has an approved annual budget of $500,000, spread evenly, so the June year-to-date budget is $250,000. Actual spend through June is $268,400. Finance policy requires written commentary on any variance above both 5% and $10,000.
- Dollar variance. $268,400 − $250,000 = $18,400. Positive: actual is above budget.
- Percentage variance. $18,400 ÷ $250,000 = 0.0736 = 7.36%.
- Apply the account type. Marketing is an expense line, so spending above budget is unfavorable.
- Effect on operating income. −$18,400. Operating profit is $18,400 lower than plan because of this line.
- Test materiality. 7.36% clears the 5% test and $18,400 clears the $10,000 floor. Both conditions are met, so the line needs commentary.
- Run rate. $268,400 ÷ 6 = $44,733 per month against a budget of $41,667 per month.
- Project the year. $44,733 × 12 = $536,800. Against the $500,000 annual budget that is a projected overrun of $36,800, or the same 7.36%.
Now do the part the arithmetic cannot. Two explanations fit these numbers and they demand opposite responses. If the department pulled a trade show forward from September into June, spend is on plan for the year and the projection is wrong — the correct commentary says “timing, no full-year impact” and the projection should be redone with the remaining periods at budget. If instead media rates rose 7%, the run rate is real, the $36,800 is real, and someone has to find it elsewhere or ask for more. Same variance, same percentage, entirely different meaning. Every variance you flag needs that sentence attached.
How to read the result: thresholds, run rates and timing
There is no universal acceptable variance, because tolerance depends on how controllable the line is. In practice a controller applies roughly three bands. Fixed, contracted lines such as rent, insurance and depreciation should land within 1–2% of budget; anything more means a posting error or a contract change nobody told finance about. Semi-variable lines such as utilities, repairs and travel routinely run ±5–10% and only matter as a trend. Genuinely variable lines such as materials, freight and commissions should be judged against a flexed budget, not the original one, because volume moves them mechanically.
Use two thresholds together, never one. A percentage rule on its own drags you into a 40% variance on a $2,000 subscription line while ignoring a 3% miss on a $4 million payroll. A dollar rule on its own ignores the small line that has quietly tripled. Requiring both a percentage and a dollar test filters a fifty-line report down to the handful that move the result.
Read the direction of travel before the size. A line that missed by 2% in each of the last four months is a broken assumption in the budget; a line that missed by 8% once is an event. The first needs the budget or the process changed, the second needs an explanation and a note that it will not recur. This is also why a year-to-date variance is more informative than a single month: the timing noise that dominates monthly numbers largely cancels.
Finally, treat the run-rate projection as a question, not an answer. Carrying a six-month rate forward assumes even seasonality and no corrective action, and both assumptions are usually wrong. It is still the fastest way to see whether a variance that looks small today lands as a serious full-year problem, which is exactly what a finance director wants flagged in month six rather than month eleven.
Favorable or unfavorable, by account type
| Line type | Actual above budget | Actual below budget | Effect on operating income |
|---|---|---|---|
| Revenue, other income | Favorable | Unfavorable | + variance |
| Cost of goods sold | Unfavorable | Favorable | − variance |
| Operating expense, payroll | Unfavorable | Favorable | − variance |
| Gross profit, contribution, EBITDA | Favorable | Unfavorable | + variance |
| Contra-revenue: returns, discounts | Unfavorable | Favorable | − variance |
| Capital expenditure | Neither by itself | Neither by itself | None until depreciated |
A favorable expense variance is not automatically good news. Underspending on maintenance, training or lead generation borrows from a later period.
Where a 5% rule and a $10,000 floor disagree
| Budget for the line | 5% of budget | Dollar floor | Which test binds | Variance needed to trigger review |
|---|---|---|---|---|
| $25,000 | $1,250 | $10,000 | Dollar floor | $10,000 (40.0%) |
| $100,000 | $5,000 | $10,000 | Dollar floor | $10,000 (10.0%) |
| $200,000 | $10,000 | $10,000 | Both, equally | $10,000 (5.0%) |
| $500,000 | $25,000 | $10,000 | Percentage | $25,000 (5.0%) |
| $2,000,000 | $100,000 | $10,000 | Percentage | $100,000 (5.0%) |
| $10,000,000 | $500,000 | $10,000 | Percentage | $500,000 (5.0%) |
The crossover sits where 5% of the budget equals the floor, at a $200,000 line. Below it the floor governs; above it the percentage governs.
Mistakes that make a variance report useless
- One formula down the whole column. Copying a single variance formula across revenue and expense lines mislabels half the report. Drive the label from the account type.
- Comparing actual against the wrong plan. Actual versus original budget answers a different question from actual versus latest forecast. Label the column and keep the two apart.
- Not flexing for volume. Judging variable costs against a fixed-volume budget punishes managers for selling more. Flex the budget to actual output first.
- Chasing percentages on small lines. A 300% variance on a $400 line is noise. Apply a dollar floor before anyone writes commentary.
- Treating timing as performance. Accrual cut-offs, prepayments and invoice slippage produce large monthly variances that reverse next month. Check the sub-ledger before you write a cause.
- Netting variances that offset. A department that is $40,000 under on payroll and $38,000 over on contractors shows a $2,000 net variance and has a real substitution story hidden inside it.
- Explaining the number instead of the cause. “Marketing is over by $18,400 because we spent more than budget” is a restatement, not an explanation. Name the driver, the owner and the action.
No accounting standard governs a budget variance
Budgets are internal management information. Neither US GAAP nor IFRS prescribes a budget, a variance format, or a materiality threshold for one, so your policy is whatever your finance function writes down. The Institute of Management Accountants publishes guidance on planning and variance practice through its Statements on Management Accounting, and the framework used here — static budget, flexible budget, and the variances between them — is the standard treatment in managerial accounting texts such as Horngren.
Two exceptions matter. Governmental funds in the United States do report budgetary comparisons: GASB requires a budgetary comparison presenting original budget, final budget and actual results for the general fund and major special revenue funds. And where standard costs feed inventory valuation, the variances stop being purely internal, because ASC 330 only permits standard costs in inventory if they approximate actual cost — which means material variances have to be cleared to cost of sales or allocated back to inventory at period end.
Budget variance against standard-cost variance analysis
A budget variance tells you a line missed. It cannot tell you why, because it collapses price, quantity, efficiency and mix into one number. Standard costing exists to split that number apart, and the split follows the same pattern everywhere: hold one factor at standard and let the other move.
For materials, the direct material price and quantity variance calculator separates paying the wrong price from using the wrong amount. For labour, the direct labor rate and efficiency variance calculator separates the wage rate from the hours taken. For the top line, the sales volume and price variance calculator separates selling fewer units from selling them cheaper. Together those three account for most of a manufacturer's operating income miss, and each one lands on a different manager's desk.
Above the individual lines, the ratios do the summarising. Whether a cost overrun actually threatens the year depends on how much contribution each incremental sale brings, which the contribution margin calculator gives you, and on how far above break-even you are operating — see the break-even sales revenue calculator and the margin of safety calculator. If the variance you are explaining sits in cost of sales, reconstructing the number from the stock ledger with the cost of goods sold calculator is usually faster than arguing about the general ledger balance.
Key terms
- Static budget
- The budget as approved, at the volume assumed when it was built. Comparing actual against it mixes volume effects with everything else.
- Flexible budget
- The budget recalculated at actual volume, holding standard prices and rates. The correct benchmark for variable costs.
- Favorable variance
- A variance that increases operating income relative to plan: revenue above budget, or cost below it. Abbreviated F.
- Unfavorable variance
- A variance that reduces operating income relative to plan: revenue below budget, or cost above it. Abbreviated U, and sometimes shown in brackets.
- Materiality threshold
- The policy rule that decides which variances need written explanation. Best expressed as a percentage and a dollar floor that must both be cleared.
- Run rate
- Actual to date divided by periods elapsed. Multiplied by the periods in the year, it projects a full-year outcome on the assumption that nothing changes.
- Timing variance
- A gap caused by when a transaction was recorded rather than by how much was spent. It reverses in a later period and carries no full-year impact.