The final analytical review before signing
What the final analytical review at audit completion checks under ISA 520, how it differs from fieldwork analytics, and how to record its conclusion.
By Achekam · · 5 min read · For Accountants & audit firms in South Africa, Nigeria, Kenya
Every section of the file is reviewed and cleared. The lead schedules agree, the adjustments are posted, the papers are signed off. And yet, read as a whole, the statements tell a story that does not quite fit: gross margin up sharply in a year the client described as difficult, receivables growing faster than sales, a cash balance that looks healthy in a business that kept asking for payment extensions. Each area was audited separately and each passed. Nobody stood back.
That standing back is the job of the final analytical review. ISA 520 requires the auditor to design and perform analytical procedures near the end of the audit that help form an overall conclusion on whether the statements are consistent with the auditor's understanding of the entity.
The overall conclusion it supports
The final review is not more substantive testing. Its question is broader: now that the figures are final, do they make sense in the light of everything we know about this business?
That question matters for three reasons:
- it can reveal a previously unrecognised risk of material misstatement, which would require the auditor to revise the risk assessment and perhaps do more work;
- it tests the coherence of the statements across areas, which no single working paper does;
- it gives the engagement partner a direct view of the figures before signing, rather than a view mediated by section conclusions.
How it differs from analytics during fieldwork
Analytical procedures appear three times in an audit, with different purposes and different standards of precision.
| Stage | Purpose | What it requires |
|---|---|---|
| Planning | Identify risks of material misstatement | Broad comparisons; enough to point at areas of risk |
| Fieldwork | Obtain substantive evidence for an assertion | A precise expectation, reliable data, a threshold for acceptable differences, and investigation of differences above it |
| Completion | Form an overall conclusion | Final figures compared with expectations and with the auditor's understanding |
The fieldwork version, covered in substantive analytical procedures, is evidence for a specific assertion. The final version is not designed to be. It does not need the same precision, but it does need someone senior enough to know what the figures should look like.
Comparing final figures with expectations and last year
Work from the final statements, after all adjustments. A practical sequence:
- Compare each line with the prior year and with the expectation formed at planning. Note movements that were not explained during fieldwork.
- Recompute key relationships: gross margin, expenses as a proportion of revenue, receivable and payable days, inventory turnover, interest against average borrowings, depreciation against cost.
- Read across statements: does the cash flow statement agree with the movements in the statement of financial position? Do tax charges make sense against profit?
- Test the story: if the client lost a major customer, revenue and receivables should show it. If staff numbers fell, payroll should too.
- Look at the notes: commitments, contingencies and related-party balances should be consistent with what the team learnt.
A worked example in words: the final figures show gross margin rising, while the team's understanding is that the client cut prices to keep its largest customer. The inventory section was cleared on existence and valuation testing. The final review prompts a question nobody asked in fieldwork: could closing inventory be overstated through costs capitalised that should have been expensed? The answer may be benign, a change in product mix, but it must be found and recorded before signing. Reviewing the balance sheet as a whole is the subject of reviewing the statement of financial position.
Unexplained movements
When the final review turns up something unexpected, ISA 520 requires the auditor to investigate: inquire of management, obtain evidence about the response, and perform other procedures where needed. An explanation that is not corroborated is not an explanation.
If the investigation shows that the risk assessment missed something, the right response is to reopen the relevant area, not to write a memo explaining it away. That can be uncomfortable at completion, which is precisely why the review should happen before the signing meeting is booked, not on the morning of it.
Doing the review in Achekam
A final analytical review is only as good as the figures it uses. In Achekam, a working paper can have a spreadsheet workbook whose figures are refreshed from the engagement's trial balance, with any figure it could not read listed. Refresh the review workbook after the last adjustment, and read the list of unread figures; a ratio built on a cell that silently did not update is worse than no ratio. Refreshing is refused on a locked file or a paper under review, so do it before the paper goes to the manager.
Lead schedules are built from the engagement's trial balance with each figure traceable to its lines, and staff can record an audit adjustment on a line with a reason. When the review asks "where does this number come from?", the trace answers it, and an adjustment that changes the picture is recorded where it was made.
On the statements themselves, staff can run an automated review of a draft that lists findings for a person to accept, resolve or dismiss. It is advice only and does not block signing. The Validation & Math Engine separately runs fixed, non-AI checks such as double entry and note ties. Neither replaces the auditor's overall conclusion; both can feed questions into it.
Recording the conclusion
The completion work item should hold the review as a step with a written response and its documents. A step is marked done only that way, or set aside with a reason, and steps are never deleted.
Staff can ask AI to draft a conclusion on the working paper, and the draft goes into the file only when a person saves it. A useful conclusion names the significant movements, the explanations obtained, the evidence that supports them, and states whether the statements are consistent with the team's understanding of the entity. "Final analytics performed, no issues" is not a conclusion; it is a tick.
The ACCA's article on completing the audit covers the rest of the completion stage that this review sits alongside: reviewing files, evaluating misstatements, subsequent events and going concern, and written representations.
Who should do it
Have the final analytical review prepared by the manager and read by the partner before anything else in the completion file. Of all the papers in the file, it is the one most likely to make the partner ask a question that matters.
General guidance only, not advice on your own circumstances. Tax and payroll rules change: check the current position with the authority or your adviser.
