Conducting Variance Analysis

Conducting Variance Analysis

When To Use

• Monthly, quarterly, or annual close when comparing actuals against budget or forecast
• Preparing management commentary for board decks, earnings packages, or operating reviews
• Investigating unexpected P&L or balance sheet movements flagged by finance or business owners
• Rolling forecast updates that require rebaselining assumptions
• Ad hoc deep dives requested by CFO, controller, or business unit leads

Inputs To Gather

• Budget/forecast data: Approved budget or latest forecast by line item, cost center, and period
• Actual results: General ledger trial balance or sub-ledger detail for the analysis period
• Prior period actuals: Prior year or prior quarter for trend context
• Chart of accounts mapping: Ensure budget and actuals align to the same account hierarchy
• Volume/operational metrics: Units sold, headcount, hours billed, transactions processed — whatever drives the line items
• Known events log: One-time items, reorgs, timing shifts, or reclassifications already identified by the business
• Materiality threshold: Dollar and percentage thresholds for which variances require narrative explanation (e.g., >$50K or >10%) [VERIFY — thresholds vary by organization and reporting level]

Workflow

1. Align data structure

   • Map actuals to the budget hierarchy (cost center, department, GL account)
   • Reconcile total actuals to the posted trial balance before proceeding
   • Confirm the reporting period matches (calendar month vs. 4-4-5, fiscal vs. calendar year) [VERIFY]

2. Compute raw variances

   • Calculate dollar variance: Actual − Budget (favorable/unfavorable sign convention per company policy)
   • Calculate percentage variance: (Actual − Budget) / |Budget|
   • Flag any line where budget = 0 but actuals exist (new activity, misclassification, or timing)

3. Apply materiality filter

   • Rank variances by absolute dollar impact
   • Isolate items exceeding the materiality threshold for detailed analysis
   • Group immaterial variances into an "other" category with a brief roll-up note

4. Decompose drivers

   • Volume variance: (Actual volume − Budget volume) × Budget rate/price
   • Rate/price variance: (Actual rate − Budget rate) × Actual volume
   • Mix variance: Impact of product/service/channel mix shift on blended margins
   • Timing variance: Identify spend or revenue recognized in a different period than budgeted
   • One-time / non-recurring items: Isolate discrete events (severance, legal settlements, asset write-downs) from run-rate trends
   • For cost lines, distinguish between controllable variances (hiring pace, discretionary spend) and non-controllable variances (FX, commodity prices, allocated overhead)

5. Build management narrative

   • Lead each variance explanation with the dollar impact and direction (favorable/unfavorable)
   • State the primary driver in one sentence, then provide supporting detail
   • Connect variances to operational actions: "Revenue was $1.2M favorable driven by 8% higher unit volume in the Southeast region following the Q2 channel expansion"
   • Quantify offsetting variances explicitly — avoid netting without disclosure
   • Flag any variance expected to persist into future periods vs. one-time catch-ups

6. Prepare forecast implications

   • For each material variance, indicate whether the current full-year forecast should be adjusted
   • Note risks and opportunities with estimated dollar ranges
   • Recommend specific actions where a variance is unfavorable and controllable

Output

• Variance summary table: Line item, budget, actual, $ variance, % variance, favorable/unfavorable flag — sorted by materiality
• Driver decomposition detail: For each material variance, a breakdown into volume, rate, mix, timing, and one-time components
• Management narrative: Plain-language explanations suitable for executive and board audiences, with each material line item addressed
• Forecast impact section: Adjustments recommended to the rolling forecast or full-year outlook, with risk/opportunity flags
• Appendix: Data reconciliation notes, threshold definitions, and any items marked [VERIFY]

Quality Checks

• Total actuals in the variance report reconcile to the posted GL trial balance — no unexplained gaps
• All variances exceeding the materiality threshold have a narrative explanation with a named driver
• Volume and rate/price variance components sum back to the total line-item variance (no residual)
• Favorable/unfavorable sign convention is consistent throughout (revenue favorable = actual > budget; cost favorable = actual < budget)
• One-time items are explicitly separated from recurring run-rate variances
• Narrative avoids circular language ("costs were higher because spending increased") — every explanation ties to an operational or external cause
• Prior period trends are referenced where a variance represents an acceleration or reversal of an existing pattern
• Any data point sourced from outside the GL (headcount, volume, pricing) is cross-referenced or marked [VERIFY]