Revenue audit
How To Plan A Revenue Recognition Audit That Tests Evidence, Judgment, And Control Ownership
A sound audit plan follows the obligation before it follows the invoice.
Lead audit analyst
October 28, 2026
We admit the common mistake plainly: revenue is often treated as settled once an invoice exists. A sound audit goes further, testing contract terms, performance evidence, estimates, cut-off, and reporting controls. This guide explains how to plan that work, distinguish fact from judgment, and document findings that remain useful across future reporting cycles.
Begin with obligations
Read The Contract Before The Ledger
We start with a common mistake: treating an invoice as the main evidence that revenue belongs in the reporting period. It is not. An invoice confirms a billing event, while the accounting conclusion depends on what the organisation promised, when the obligation was satisfied, and which evidence existed at close. We compare contract wording with operational reality, including delivery conditions, customer acceptance, cancellation rights, service periods, amendments, and continuing obligations. This step often exposes a gap between commercial language and accounting policy. A sales team may regard an arrangement as complete while installation, support, approval, or another defined deliverable remains outstanding. We document the relevant facts before reviewing the ledger, then separate confirmed evidence from unresolved judgment. Historical adjustments and earlier exceptions help us identify wording that has produced inconsistent decisions. The result is a review that follows the transaction’s substance rather than accepting the most convenient administrative record.
Test both sides
Test Cut-Off Against Transaction Evidence
The usual cut-off mistake is to select invoices issued near year end and check their dates. That is only a starting point. We trace selected transactions through order records, dispatch or completion evidence, customer acceptance, service logs, billing, ledger posting, credit notes, and later amendments. Each source answers a different question, so one matching date cannot settle the conclusion. A service may have started without being complete, or goods may have been delivered while acceptance remained outstanding. We focus on unusual timing, manual entries, large contract changes, and transactions immediately before and after the reporting boundary. None of these features proves an error, but each needs an explanation linked to the contract and evidence available at close. We compare current findings with previous adjustments because repeated small exceptions can reveal a process weakness more clearly than one isolated correction. The conclusion should identify the relevant evidence, the reporting implication, and any limitation that management must resolve.
Expose the judgment
Make Estimates Traceable
Estimates require discipline because uncertainty cannot simply be removed from a reporting file. We examine returns, rebates, concessions, milestones, and other variable amounts by tracing assumptions to source data, calculations, approvals, and later developments. Later outcomes can inform the review, but they do not automatically prove that an earlier estimate was unreasonable. We ask whether the judgment reflected information available at the reporting date and whether comparable arrangements received consistent treatment. We record who approved the assumption, when it should be revisited, and what evidence would change the conclusion. Unsupported precision is not stronger analysis; it is uncertainty presented in a more confident format. Where evidence is incomplete, we state the limitation rather than filling the gap with an unsupported assumption. This makes the review more useful to management and gives future reviewers a clear basis for understanding how the amount was determined.
Assign durable ownership
Turn Findings Into Durable Controls
A written policy can be sound and still fail in operation. We test the handoffs connecting contract review, delivery evidence, billing, amendments, credit notes, and period-end close. The practical question is whether the assigned person performed the control, retained evidence, resolved exceptions, and understood when escalation was required. Manual overrides receive attention because they can bypass a sensible workflow without producing an obvious system warning. We compare system fields with policy requirements and identify where information is re-entered, delayed, or left to informal messages. Findings are classified as confirmed errors, missing evidence, estimation uncertainty, policy gaps, or control failures, because each category needs a different response. Over a three-to-five-year horizon, a process dependent on one experienced employee remains fragile. Remediation should define the decision point, owner, supporting record, and review method so the change remains workable as staff and arrangements change.