Manufacturing SOX Compliance Consulting

Most public manufacturers value inventory and cost production using standard costs — a predetermined cost per unit for material, labor, and overhead, set at the start of a fiscal year or updated quarterly, against which actual transactions are recorded throughout the period. The mechanism creates a structural risk that does not exist under actual costing: if the standards themselves are wrong — a purchase price variance that was never rolled into the standard, a labor routing that no longer reflects the actual process after an equipment change, an overhead absorption rate calculated on a production volume assumption that did not hold — every unit produced against that standard carries the error forward into inventory and COGS until someone catches it. The catch mechanism is variance analysis: purchase price variance, material usage variance, labor efficiency variance, and overhead volume/spending variance, each computed as actual cost minus standard cost, reviewed by someone with the authority and cost-accounting background to distinguish an operational issue (a supplier price increase, a scrap event) from a standard-setting error that needs to be corrected at the source.

The SOX-relevant control is not 'variances are calculated' — most ERPs do that automatically as a byproduct of the standard-cost transaction flow. It is whether variances above a defined materiality threshold are investigated, documented, and resolved before they compound across multiple periods, and whether the standard-cost revision process itself (who can change a standard, what approval it requires, how often standards are refreshed against current actual costs) is governed with the same rigor as any other financially relevant system configuration. A control environment where variances are calculated and reported but nobody is accountable for closing the loop — explaining the variance and correcting the standard if the standard was wrong — is one of the most common findings in first-year SOX assessments at manufacturers that recently crossed the accelerated-filer threshold.

The bill of materials — the structured list of components, quantities, and routings that defines how a finished good is built — is not just an engineering or production document. It is the direct input to the standard cost roll-up: change a component, a quantity per, or a labor routing step on the BOM, and the standard cost of every unit built against that BOM changes with it, whether or not anyone remembers to re-run the cost roll-up. Uncontrolled BOM changes are therefore a direct path to a costing misstatement: an engineering change order that substitutes a cheaper component without updating the standard leaves inventory overstated; a routing change that adds a labor step without a corresponding standard-cost update leaves COGS understated the moment production starts using the new routing.

The control that matters is a formal engineering-change-order (ECO) workflow that requires cost-accounting sign-off before a BOM change goes live in the production system — not just engineering and quality approval — paired with a triggered cost roll-up (recalculating standard cost for any finished good whose BOM changed) before the next transactions post against it. Equally important is restricting who can edit a released BOM directly in the ERP or PLM system outside the ECO workflow; an unrestricted or loosely audited BOM-edit permission is functionally equivalent to an unrestricted journal-entry permission, because it can move standard cost without a corresponding accounting review.

External auditors rely heavily on physical inventory observation — attending a cycle count or full physical count, testing the count procedure, and tracing a sample of counted items to the inventory ledger — as a substantive test of inventory existence, but the SOX control question is upstream of the audit: does the company operate a cycle-count program rigorous enough to keep book-to-physical variances small and explainable between full counts. A mature program counts high-value or high-velocity SKUs frequently (often monthly or more), reconciles book quantity to counted quantity, investigates variances above a set threshold, and requires management sign-off on adjustments before they post — rather than treating cycle counts as a warehouse housekeeping task disconnected from the accounting close.

A second, frequently under-controlled risk sits at the interface between the shop-floor manufacturing execution system (MES) — which tracks actual material consumption, labor time, and production completions in real time — and the financial ERP, which posts those transactions to inventory and WIP accounts, often on a batch or near-real-time feed. Interface failures here (a dropped completion transaction, a duplicate material issue, a units-of-measure mismatch between the MES and ERP item masters) directly misstate WIP and finished-goods inventory even when both systems are individually functioning correctly, and because the error originates in a system outside the ERP's own controls, it is easy for it to go undetected until a cycle count or physical inventory surfaces an unexplained variance. A completeness reconciliation — comparing total production and material-issue transactions in the MES to what actually posted in the ERP for the period — closes that gap.