Begin by writing the review question in one sentence. “Did the bank balance reconcile?” is narrower than “Is cash complete and appropriately presented?” Define the population, period, entity, system, and exclusions. Then select a route through the evidence: source, transformation or calculation, ledger or schedule, review note, and owner decision where needed. A route that stops at a screenshot is a gap, even if the screenshot looks persuasive.
Test ordinary and unusual work separately. Ordinary items show whether the routine path is reproducible. Reopened items, corrections, late sources, estimates, and manual journals reveal whether the pack preserves changes and escalation. The reviewer should record gaps by dimension rather than adding a vague “insufficient support” comment. That diagnosis helps distinguish document retrieval from a missing population, broken linkage, weak review note, or unresolved judgement.
For remote teams, the index should include source name and version, reporting period, owner, preparer, reviewer, status, and next action. If a source changes, preserve the old state or a change reference. If an owner accepts an exception, retain that acceptance with the item it resolves. If no response has arrived, state that fact. A close pack becomes more useful when it shows what is not known instead of presenting an empty exception column as completion.
The index should also explain its stopping rule. If the pack is complete for a narrow reconciliation but not for a broader reporting conclusion, say so in the heading and handoff. That small statement prevents later readers from treating a workpaper as evidence for a question it was never designed to answer. It also makes future comparisons fair because the next close can preserve the same scope.
