Overview
Based on "Cost Accounting: A Managerial Emphasis" by Horngren, Datar, and Rajan - the standard reference for management accounting and variance analysis. Variance analysis is not about flagging numbers that are off; it is about explaining why they are off in a way that leads to a decision.
Horngren's principle: decompose variances into their components (price, volume, mix, efficiency) before drawing conclusions. A single unfavorable revenue variance could be caused by lower prices, lower volume, or a worse product mix - and each cause has a completely different corrective action.
Workflow
Step 1: Establish the variance summary
Before decomposing, calculate the total variance for each line and flag direction.
VARIANCE SUMMARY - [Period]
Actual Budget Variance % F / U
Revenue $[X] $[X] $[X] [%] [F/U]
Cost of Goods Sold $[X] $[X] $[X] [%] [F/U]
Gross Profit $[X] $[X] $[X] [%] [F/U]
Gross Margin % [%] [%] [%pts] [F/U]
Operating Expenses $[X] $[X] $[X] [%] [F/U]
Operating Income $[X] $[X] $[X] [%] [F/U]
F = Favorable (better than budget). U = Unfavorable (worse than budget).
Apply F/U correctly: higher-than-budget revenue is F; higher-than-budget expense is U.
Focus investigation on variances that are:
- More than [materiality threshold] in absolute dollars
- More than 5-10% of the budgeted line
- Unexpected directionally (e.g., revenue up but gross margin down)
Step 2: Decompose revenue variances
Revenue variance has three components: price, volume, and mix.
Price variance - Did we sell at the price we planned?
Revenue Price Variance = (Actual Price - Budget Price) x Actual Volume
Favorable if actual price > budget price
Volume variance - Did we sell the quantity we planned?
Revenue Volume Variance = (Actual Volume - Budget Volume) x Budget Price
Favorable if actual volume > budget volume
Mix variance (multi-product businesses) - Did we sell the right mix of products?
Revenue Mix Variance = (Actual Mix % - Budget Mix %) x Total Actual Volume x (Product Margin - Avg Margin)
Favorable if actual mix skewed toward higher-margin products
After decomposing, write a one-paragraph explanation:
Revenue was [F/U] by $[X] vs. budget.
- Price: [favorable/unfavorable] by $[X] - [explanation: discounting, pricing change, FX, etc.]
- Volume: [favorable/unfavorable] by $[X] - [explanation: market demand, sales execution, etc.]
- Mix: [favorable/unfavorable] by $[X] - [explanation: product mix shift, channel mix, etc.]
Net: $[X] [F/U]
Step 3: Decompose cost variances
For COGS and direct costs, decompose into spending (price) and efficiency (usage) components.
Spending variance (price effect) - Did inputs cost more or less per unit than planned?
Spending Variance = (Actual Rate - Budget Rate) x Actual Quantity
Efficiency variance (usage effect) - Did we use more or fewer inputs per unit of output than planned?
Efficiency Variance = (Actual Quantity - Budget Quantity) x Budget Rate
For labor:
Labor Rate Variance = (Actual Hourly Rate - Budget Hourly Rate) x Actual Hours
Labor Efficiency Variance = (Actual Hours - Budget Hours) x Budget Hourly Rate
For materials:
Material Price Variance = (Actual Unit Cost - Budget Unit Cost) x Actual Quantity
Material Usage Variance = (Actual Quantity - Budget Quantity) x Budget Unit Cost
Step 4: Explain operating expense variances
For fixed and discretionary costs (SG&A, R&D, headcount), the decomposition is simpler:
Expense Variance = Actual Expense - Budget Expense
Explain by category:
- Headcount: [actual HC vs. budget HC] x [avg cost per head]
- Timing: expenses budgeted in this period but shifted (or pulled forward)
- One-time items: [legal, severance, project costs not in budget]
- Inflation / rate: [vendor price increases above budget assumption]
- Volume-driven: [variable costs that moved with revenue]
Step 5: Write the management commentary
Translate the numbers into narrative for leadership:
BUDGET VARIANCE COMMENTARY - [Period]
HEADLINE
[One sentence: net result and whether it was above or below budget, by how much]
REVENUE: [F/U $X vs. budget]
[2-3 sentences: what drove it, broken into price/volume/mix where relevant]
GROSS MARGIN: [F/U $X / [X]pts vs. budget]
[2-3 sentences: COGS drivers - materials, labor, overhead absorption, mix]
OPERATING EXPENSES: [F/U $X vs. budget]
[2-3 sentences: key lines that moved, timing vs. structural, headcount context]
KEY RISKS / OPPORTUNITIES
- [Forward-looking risk or opportunity suggested by the variance analysis]
- [Forward-looking risk or opportunity]
CORRECTIVE ACTIONS (where applicable)
- [Specific action, owner, timeline]
- [Specific action, owner, timeline]
Anti-Patterns
1. Describing variances without decomposing them
Bad: "Revenue was $500K unfavorable vs. budget due to lower sales."
Good: "Revenue was $500K unfavorable: $200K price (average selling price declined 3% due to competitive discounting), $300K volume (unit sales 8% below budget in the SMB segment, offset partially by enterprise outperformance of +$150K)."
2. Calling everything "timing"
Bad: Labeling every unfavorable expense variance as "timing" without evidence.
Good: Timing means a specific expense was budgeted in one period and shifted to another. Identify the specific item, the budgeted period, and when it will now be incurred.
3. No corrective action for unfavorable variances
Bad: Documenting a $300K unfavorable COGS variance and stopping there.
Good: Variances that repeat need a root cause and an owner. "Material costs exceeded budget by $300K due to supplier price increases. Procurement is renegotiating the contract with [vendor] - revised pricing expected by [date]."
4. Favorable variance treated as no investigation needed
Bad: Ignoring favorable variances.
Good: A favorable variance can hide a problem. Revenue $500K favorable on $300K lower expenses might mean volume was lower but headcount and overhead are under-absorbed. Investigate favorable variances where the cause is not obvious.
5. Mixing actuals from different accounting bases
Bad: Comparing cash-basis actuals to accrual-basis budgets.
Good: Confirm both actuals and budget use the same accounting basis before calculating variances. If they differ, restate before analysis.
Quality Checklist