A "favorable" cost variance (actual below standard) looks like good news in the leadership review. In reality, it can signal a cheaper, lower-quality ingredient quietly substituted in, or a standard cost that's become fiction. Here's the real formula, the three variants that actually separate accountability, and the four root causes to check before celebrating a favorable number.
(Actual COGS - Theoretical COGS) / Theoretical COGS x 100One row per production order batch produced
Actual_COGS = total cost of goods sold per the management accounts (RM, packaging, direct labour, overheads); Standard_COGS = theoretical cost based on approved BOM and standard cost rates.
depreciation variances from revaluation; one-off write-offs declared separately; currency revaluation on RM stocks.
in commodities-exposed businesses (dairy, cocoa, wheat, edible oil), food cost variance is not a production efficiency signal — it is a commodity price signal; standard costs locked at budget become meaningless within 6 months of a commodity spike.
Splits the total variance into its core components: Purchase Price Variance (PPV) and Material Usage/Yield Variance (MUV). When to use it: Always. A single, blended variance figure is useless for accountability. This split separates procurement's impact (price) from operations' impact (yield).
Isolates the variance component caused by selling a different mix of products than was budgeted. When to use it: When the sales team has significantly shifted the product mix towards lower-margin or higher-cost SKUs, explaining a variance that neither procurement nor operations can.
Calculates variance on a weekly, not monthly, basis for key commodities like dairy, cocoa, or edible oils. When to use it: In commodity-exposed categories. A monthly average hides weekly price spikes that should trigger immediate commercial or hedging action.
Outdated Bill of Materials (BOM) or routings. When the standard BOM or labor routing in the ERP doesn't match the current production process, every batch produced will automatically generate a variance, even with perfect execution.
In the ERP, compare the standard material consumption and labor hours for a top SKU against an engineering audit of the actual process on the shop floor. If the BOM has not been updated in >12 months, it is likely wrong.
'Production efficiency is down' — this looks like an operational failure, but the problem is that the baseline for 'standard' is a work of fiction.
Sourcing non-standard materials to hit PPV targets. Procurement substitutes a cheaper raw material that meets basic spec but has lower yield or requires more processing time, directly causing an adverse usage or labor variance.
Cross-reference the Production Order Variance Report in the ERP with the Purchase Order history. Flag any batch with high material usage variance where the input material PO was from a non-primary supplier or had a significant favorable PPV.
'Purchase Price Variance: +4% savings' in the procurement dashboard — this looks green because it measures the cost of buying, not the total cost of using.
Poor overhead cost absorption from unplanned short runs. When the schedule is broken to run small, urgent batches to meet a customer service commitment, fixed overhead costs are spread over fewer units, inflating the actual cost per unit.
In the MES or production scheduling system, calculate the ratio of actual run length to standard run length. If >10% of production orders are less than 50% of the standard batch size, cost absorption is being destroyed.
'Customer service level is 99%' — this looks green because the cost of achieving it through inefficient production runs is hidden in an aggregated FCV number.
Inaccurate yield or scrap factors in standard cost. The standard cost build-up assumes a certain level of unavoidable waste or yield loss. If the actual process consistently generates more scrap than the standard allows, a negative variance is guaranteed.
Pull the scrap transaction data from the WMS or MES for the last 3 months. Compare the actual scrap rate for a high-volume SKU against the scrap allowance in its standard cost card in the ERP. A persistent gap >1% means the standard is wrong.
'Material Usage Variance is consistently -1.5%' — this is presented as a recurring operational failure, when in fact it's a planning failure to set a realistic standard.
Decompose Variance Report by Driver. Aggregated variance reports hide accountability → Modify the standard monthly FCV report to split total variance into Price, Usage, and Mix components → Each department sees their direct impact, ending the blame game.
Mandate Sign-Off for Non-Standard Materials. Procurement makes cost-saving substitutions without operational input → Institute a formal, mandatory sign-off where the Plant Manager accepts potential yield/efficiency loss before a non-standard material is used → Usage variance from substitutions becomes a documented, accepted risk.
Weekly Variance 'Flash' for Key SKUs. Monthly variance reporting is too slow to be a control tool → Create a simple weekly report tracking actual vs. standard material and labor for the top 5 highest-volume SKUs → Operations gets near-real-time feedback on cost performance, enabling correction within the month.