A variance is the distance between what the organisation expected and what actually happened.
The arithmetic is easy.
The learning is difficult.
Revenue can be below budget because fewer units were sold, prices fell, product mix changed, a launch was delayed, a market contracted or foreign exchange moved. Cost can be above budget because input prices rose, more material was used, activity was higher than planned, staffing changed, a project accelerated or the original standard was unrealistic.
“Unfavourable variance” does not tell us which of those things happened.
Variance analysis exists to open the difference until the mechanism becomes visible.
This article is part of Batch 018 of the eduKateSG Finance Authority 400. Budgets owns the planned resource boundary. Forecasts owns the current expectation. This page owns the plan-to-actual learning loop. Scenario Planning owns alternative futures. The canonical owner remains How Finance Works.
A variance is not a verdict. It is a location where the plan and the world stopped agreeing.
Educational boundary: this article explains general management-accounting and financial-analysis concepts. It is not accounting, audit, tax, legal, investment or business advice for a particular organisation.
The Short Answer: What Is Variance Analysis?
Variance analysis compares an actual result with a reference result—such as a budget, forecast or standard—and decomposes the difference into causes that can be interpreted and acted on.
The reference matters.
- Actual vs budget asks how execution differed from the approved plan.
- Actual vs forecast asks how reality differed from the latest expectation.
- Actual vs prior period asks how performance changed through time.
- Actual vs standard asks how actual price, usage or efficiency differed from a defined standard.
The same actual number can produce different variances against different baselines. A variance should therefore never be quoted without saying “variance from what?”
The Basic Formula Is Only the Door
At the simplest level:
VARIANCE = ACTUAL − REFERENCE.
But interpretation depends on the item.
Higher revenue than budget may be favourable. Higher cost than budget may be unfavourable. Yet even those labels can fail if the extra cost supported profitable growth or the revenue beat came from unsustainable discounting.
The purpose of the formula is to identify the gap. The purpose of analysis is to explain the gap.
Favourable Does Not Always Mean Good
A favourable labour-cost variance can occur because the company failed to hire enough people.
A favourable maintenance-cost variance can occur because servicing was postponed.
A favourable inventory variance can occur because the business ran out of stock and lost sales.
A favourable capex variance can occur because a critical project is six months late.
The financial label describes the direction relative to the reference. It does not settle the operating judgement.
A cost underspend is only good if the capability the cost was meant to support still exists.
Unfavourable Does Not Always Mean Bad
A company may spend more because demand was stronger than budget and production rose 30%.
A customer-acquisition budget may be exceeded because an unexpectedly efficient channel became available and produced highly profitable new customers.
A technology project may accelerate because an earlier launch has high strategic value.
The correct question is not whether the variance is red or green.
It is whether the organisation received enough additional value, capability or risk reduction for the additional resources used.
Revenue Variance: Start With Price and Volume
A simple revenue variance can be separated into price and volume effects.
Suppose the budget expected 10,000 units at $100 each: $1.0 million revenue.
Actual performance is 9,000 units at $110 each: $990,000 revenue.
The total revenue variance is only $10,000 unfavourable.
But underneath it, price is favourable and volume is unfavourable.
Those two forces nearly cancel in the headline.
The cancellation is why decomposition matters.
Price Variance: Did the Business Charge More—or Discount Less?
A price variance asks how actual selling price differed from the planned or standard selling price for the actual volume.
A favourable price variance can result from genuine pricing power, lower discounting, better customer mix, premium product mix or temporary market tightness.
It can also come with lower volume.
Price should therefore be interpreted beside volume, customer retention and margin—not in isolation.
Volume Variance: More Units Are Not Automatically Better Units
Higher volume can improve revenue and absorb fixed operating capacity more efficiently.
But additional units can have lower margin, require discounts, create more working capital or push the system into overtime and expensive capacity.
The variance should therefore ask not only how many more units were sold, but what economics came with them.
The later Contribution Margin and Operating Leverage articles retain those deeper volume-economics mechanisms.
Mix Variance: The Total Can Move Even When Individual Products Do Not
Suppose a company sells a high-margin premium product and a lower-margin entry product.
Total unit volume can hit budget while profit misses because customers bought more of the lower-margin product.
That is a mix problem.
A revenue or margin variance that does not separate mix can hide a meaningful shift in customer demand or product economics.
Market Size and Market Share Can Explain Volume More Deeply
Sales volume can miss budget because the whole market was smaller than expected or because the company’s share of that market fell.
Those are different problems.
If the market contracted but the company gained share, operating execution may be strong despite an unfavourable sales variance. If the market grew faster than expected but the company still missed its own budget, competitive position may have weakened.
Variance analysis becomes stronger when it places internal results inside the external environment.
Cost Variance: Separate Price From Usage
Direct input costs can often be decomposed into how much was paid per unit of input and how much input was used.
A materials cost miss can arise because:
- supplier prices were higher;
- more material was consumed;
- waste increased;
- product mix changed;
- quality standards changed;
- volume changed;
- the original standard was wrong.
Blaming procurement for total material cost without separating price and usage can attribute the variance to the wrong owner.
Labour Variance: Rate, Hours and Productivity Tell Different Stories
Labour cost can exceed budget because wage rates are higher, more hours are used, overtime rises, staffing mix changes or productivity falls.
It can also exceed budget because the business deliberately hired more experienced staff to improve quality or accelerate growth.
The variance should therefore connect wage cost to output, quality, service and capacity.
Lower labour cost with lower output is not an efficiency gain merely because the cost line is favourable.
Flexible-Budget Variance: Adjust for the Activity That Actually Happened
Comparing actual costs at 120% of planned activity with a fixed budget built for 100% activity can create a misleading cost variance.
A flexed budget asks what costs should have been at the actual level of activity, given the cost behaviour assumptions.
This helps separate normal volume-driven cost from genuine price, usage or efficiency changes.
ACCA’s budgeting material explicitly distinguishes flexible budgeting across activity levels from a flexed budget created at actual activity for financial control.
Planning Variance vs Operational Variance: Was the Standard Wrong—or Was Execution Wrong?
This is one of the most important distinctions in variance analysis.
A planning variance arises because the original budget or standard was based on assumptions that later prove inappropriate.
An operational variance asks how actual performance differed from a revised, more realistic standard once the planning change is recognised.
ACCA’s technical material uses this distinction to separate inaccurate standards from operational decisions and performance.
This prevents managers from being blamed for inflation, market movement or another external change that made the original standard unrealistic.
It also prevents a bad original plan from creating apparently favourable operating performance simply because the standard was easy to beat.
Timing Variance: The Annual Total Can Hide a Delayed Business
A project can be $5 million under budget because the work has not happened yet.
Revenue can be below budget because a launch shifted from December to January.
Marketing can be under budget because a campaign moved one quarter.
These timing differences can reverse later.
A variance analysis should distinguish permanent economic differences from timing differences that merely move the result between periods.
Structural Variance: Some Gaps Reveal a New Business Reality
If customer churn doubles, energy cost resets permanently, regulation requires a new compliance layer or a product becomes obsolete, the variance may not reverse.
The forecast and perhaps the next budget need to change.
Variance analysis is therefore a bridge from history into forecasting.
ACTUAL RESULT → VARIANCE → ROOT CAUSE → TRANSIENT OR STRUCTURAL? → ACTION → UPDATED FORECAST.
Cash-Flow Variance: Profit Can Be on Plan While Liquidity Is Not
A company can hit operating profit and miss operating cash flow.
Customers may pay later. Inventory may rise. Suppliers may be paid earlier. Tax timing may change.
Cash variance should therefore be analysed separately from earnings variance.
The earlier Profit Quality article owns the earnings-to-cash bridge. Variance analysis asks which planned cash assumptions failed in this period.
Working-Capital Variance: Follow Receivables, Inventory and Payables
A working-capital variance can reveal deterioration before the income statement looks weak.
- Receivable days are ten days worse than plan.
- Inventory days are fifteen days worse.
- Supplier days are five days shorter.
The cash consequence can be large even if revenue and margin are near budget.
The earlier Cash Conversion Cycle provides the operating clock. Variance analysis asks why the actual clock moved away from the planned clock.
Capital-Expenditure Variance: Underspend Can Mean Delay, Scope Change or Efficiency
A capex project that spends $20 million against a $30 million budget has a $10 million underspend.
That number is useless without the project state.
- Was the project completed for less?
- Was scope reduced?
- Were invoices delayed?
- Did construction slip?
- Was part of the project cancelled?
- Did supplier pricing improve?
The same favourable financial variance can describe excellent procurement or a failed implementation.
Headcount Variance: Vacancy Savings Can Hide Capacity Loss
Personnel expense can be below budget because hiring was efficient.
It can also be below budget because roles could not be filled.
If the vacancy reduces product development, sales, service or control capacity, the salary underspend has an operating cost elsewhere.
Variance analysis should follow the resource gap to the work it was supposed to perform.
Materiality: Not Every Variance Deserves the Same Attention
An organisation can generate thousands of variances.
Investigating all of them equally can cost more management time than the variances are worth.
Useful triage considers:
- absolute financial size;
- percentage size;
- recurrence;
- cash impact;
- risk significance;
- strategic importance;
- whether the variance is controllable;
- whether it signals a broader problem.
A small cybersecurity variance can matter more than a much larger office-supplies variance if it reveals a control failure.
Controllable vs Uncontrollable: Accountability Needs a Fair Boundary
Managers should be evaluated on decisions they can influence.
A production manager may control material usage but not the global commodity price. A local country manager may control sales execution but not the translation effect of group reporting currency. A department head may control hiring decisions but not a centrally imposed salary increase.
Separating controllable and uncontrollable drivers improves diagnosis and reduces the incentive to manipulate the explanation.
Root Cause vs Symptom: Stop One Layer Deeper
“Revenue missed because volume was lower” is an explanation, but it may not be a root cause.
Why was volume lower?
- Sales capacity?
- Customer churn?
- Competitor price?
- Product quality?
- Supply shortage?
- Market contraction?
- Launch delay?
Variance analysis should descend until the cause is specific enough to change a decision.
If the explanation does not suggest what to monitor, repair or update, it may still be too shallow.
Variance Waterfalls Make the Bridge Visible
A waterfall or bridge starts with the reference result and shows the major drivers that reconcile to actual.
For example:
| Operating profit bridge | Effect |
|---|---|
| Budget operating profit | $10.0m |
| Lower sales volume | −$1.8m |
| Higher selling price | +$0.9m |
| Adverse product mix | −$0.6m |
| Lower material cost | +$0.7m |
| Hiring delay | +$0.5m |
| Unplanned logistics cost | −$0.4m |
| Actual operating profit | $9.3m |
The bridge shows that the $0.7 million total miss contains several opposing forces.
Management now has something to discuss beyond “profit was 7% below budget.”
One-Off vs Recurring: Will the Variance Follow Us Into the Next Period?
An unplanned legal fee may be one-off.
A permanently higher wage structure is recurring.
A launch delay may shift revenue into the next quarter. A lost customer may permanently reduce the base.
The forecast should change only for the part of the variance that changes the future.
This is where variance analysis must hand off to Forecasts.
Repeated “One-Off” Variances Are Usually Telling You Something
If restructuring cost, expedited freight, contractor spend, customer credits or “unexpected” maintenance recur every year, the planning model may be excluding a normal feature of the business.
Repeated exception is evidence that the exception may belong in the base case.
The earlier Adjusted Earnings applies the same discipline to external performance measures.
Variance Analysis Can Become a Blame System
If every unfavourable variance is treated as failure, managers learn to defend the number rather than explain it.
They may delay bad news, shift costs between categories, negotiate easy budgets or construct explanations around organisational politics.
A learning-oriented system asks:
- What changed?
- Why?
- Was it controllable?
- Was the original assumption reasonable?
- What should we do now?
- What should the next forecast learn?
Accountability still matters. But accountability without diagnosis produces defensive reporting rather than better decisions.
Variance Analysis Can Also Be Gamed by Classification
A cost can sometimes be shifted between departments, periods or categories without changing the underlying economics.
Revenue can be pulled forward or pushed back within accounting constraints. Capex and opex classification can change where cost appears. Central allocations can move profit between business units.
The earlier Capital Expenditure vs Operating Expense shows one important classification boundary.
Variance analysis should follow economic substance, not merely the account code.
A Good Variance Review Ends With an Updated Action
A meeting that spends an hour explaining last month’s variance and changes nothing has completed only half the job.
The variance should produce one of several outcomes:
- no action because the difference is immaterial or self-correcting;
- operating repair;
- forecast update;
- resource reallocation;
- control change;
- assumption change;
- risk escalation;
- new monitoring indicator.
The analysis is complete when the organisation knows what the difference changes about the future.
A Worked Variance Case: Profit Beat, Cash Miss
Consider a business with the following monthly result:
| Item | Budget | Actual | Headline variance |
|---|---|---|---|
| Revenue | $5.0m | $5.2m | +$0.2m |
| Operating profit | $0.6m | $0.7m | +$0.1m |
| Operating cash flow | $0.5m | −$0.2m | −$0.7m |
| DSO | 45 days | 68 days | +23 days |
| Inventory | $1.2m | $1.8m | +$0.6m |
The business “beat budget” on revenue and profit.
It also consumed cash because customers paid more slowly and inventory rose.
The correct conclusion is not that performance was simply favourable or unfavourable.
The business gained accounting performance while increasing the financing burden of its operating cycle.
That is the kind of integrated statement a useful variance analysis should produce.
The Variance Failure Map
| Failure mode | What happens |
|---|---|
| Headline-only analysis | Total variance hides offsetting drivers |
| Red-green thinking | Favourable is assumed good, unfavourable bad |
| Wrong baseline | Actual compared with an irrelevant or stale reference |
| No flexing | Higher activity makes normal cost look inefficient |
| No planning/operational split | Managers blamed for broken assumptions |
| Timing blindness | Delayed spending mistaken for savings |
| Cash blindness | Profit variance hides working-capital stress |
| Blame culture | Explanations become defensive |
| Classification gaming | Accounts move while economics stay the same |
| No feedback | Variance explained but forecast and decisions remain unchanged |
Operating Test: Can the Variance Be Converted Into a Decision?
A useful variance explanation should reach one of three places:
- repair: something in execution should change;
- reforecast: the expected future should change;
- retain: the original plan remains appropriate and the variance is temporary or immaterial.
If the explanation cannot tell us which path is appropriate, it probably has not reached the decision layer yet.
The Variance-Analysis Diagnostic
- What baseline are we comparing with?
- What is the total variance?
- Which drivers offset one another?
- What part comes from price?
- What part comes from volume?
- What part comes from mix?
- What part comes from input rate or usage?
- Should the reference be flexed for actual activity?
- Was the original standard or assumption wrong?
- What part is operational performance?
- What part is timing?
- What part is structural and likely to recur?
- What cash consequence sits behind the profit variance?
- Who controls the driver?
- What action or forecast change should follow?
Observable Mastery Test
A factory budgets 100,000 units at $50 revenue per unit and $30 variable cost per unit. It actually sells 120,000 units at $48 while variable cost averages $32. Total revenue and total cost are both above budget.
You understand variance analysis if you can explain why a simple “actual minus budget” comparison is insufficient and identify the need to separate:
- higher activity;
- selling-price variance;
- volume effect;
- variable-cost rate or usage effects;
- the profit consequence;
- whether the original assumptions need to change.
The World Return: Did the Gap Produce Learning?
PLAN / FORECAST → ACTUAL RESULT → VARIANCE → DECOMPOSITION → ROOT CAUSE → CONTROL / ASSUMPTION / TIMING → ACTION → UPDATED FORECAST → NEXT RESULT.
The variance earns its place when the organisation becomes more accurate, more capable or more resilient because it investigated the gap.
A variance that is explained every month and never changes behaviour is only reporting.
The purpose of variance analysis is not to prove that reality disobeyed the plan. It is to discover what reality taught us that the plan did not know.
Research Anchors
ACCA’s Performance Management technical-article library covers variance analysis, materials mix and yield, market size and market share, budgeting and control. ACCA’s discussion of planning and operational variances distinguishes variances caused by inaccurate budgets and standards from those caused by operational decisions and performance.