JournalscopeGuide
ISA 240 journal entry testing: criteria and documentation
Short answer. ISA 240.32(a) requires you to test the appropriateness of journal entries recorded in the general ledger and other adjustments made in preparing the financial statements, including entries at the period end, whatever your assessment of the risk of management override. Show that the journal population is complete (roll it forward to the trial balance), choose criteria that target the characteristics of inappropriate entries, rank journals by risk, test a selection that includes an unpredictable element, and document the source data, the criteria, the parameters, the selection and your conclusion.
1. Prove the population is complete first
A test on an incomplete extract proves little. ISA 500.9 asks you to evaluate whether information produced by the entity is reliable, including its accuracy and completeness. For a ledger extract the usual evidence is a roll-forward: for every account, opening balance plus the movement in the extract equals the closing balance in a trial balance you have agreed to the financial statements. Add a check that debits equal credits and look at gaps in journal numbering. ICAEW's guidance on automated journals testing makes the same point: say how you confirmed that all the data was included.
2. Criteria that target inappropriate entries
The application material to ISA 240 lists characteristics of inappropriate journal entries: entries to unrelated, unusual or seldom-used accounts; entries by individuals who typically do not make journal entries; entries recorded at the end of the period or as post-closing entries with little or no explanation or description; entries made before or during the preparation of the financial statements that have no account numbers; and entries containing round numbers or consistent ending numbers. Firms add criteria of their own, such as weekend postings or self-approval. The criteria Journalscope applies, with their default weights:
- Posted at a weekend (weight 1): Entries made outside normal working patterns, when review is less likely.
- Posted on a public holiday (weight 1): Entries made outside normal working patterns, when review is less likely.
- Round amount (weight 1): Round numbers are a recognised characteristic of estimates and manipulated entries.
- Amount ending in 999 (weight 1): Consistent ending numbers just below an authorisation limit.
- Large journal (at or above your threshold) (weight 1): Entries large enough to be material on their own.
- Entered after period end, dated in the period (weight 3): Post-closing entries made in preparing the financial statements.
- Entered long after its effective date (weight 2): Back-dated entries that change a period already reported internally.
- Posted by a user who seldom posts journals (weight 2): Entries made by individuals who typically do not make journal entries.
- No user recorded (weight 2): Entries that cannot be traced to a person.
- Hits a seldom-used account (weight 2): Entries to unrelated, unusual or seldom-used accounts.
- Keyword in the narrative (weight 1): Narratives that point to adjustments, corrections or instructions from management.
- No narrative on any line (weight 1): Entries with little or no explanation or description.
- Debits and credits do not balance (weight 3): One-sided entries: the ledger or the extract is incomplete or altered.
- Suspense or clearing account (weight 1): Accounts used to park amounts without a clear counter-entry.
- Posted outside working hours (weight 1): Entries made late at night or early in the morning.
- Same person posted and approved (weight 2): No segregation of duties between preparer and approver.
- Manual journal (weight 1): Manual entries bypass the controls built into the accounting system.
- Possible duplicate of another journal (weight 1): Duplicated entries that inflate balances.
- Near period end (weight 1): Entries recorded at the end of the reporting period (ISA 240.32(a)(ii)).
3. Rank, then select
- Give each journal a risk score, for example the sum of the weights of the criteria it meets, and test the highest-scoring journals as key items.
- Add an unpredictable element: a few journals chosen at random from the rest. Use a seed so the selection can be re-performed.
- Keep the full list of flagged journals in the file, not only the ones you tested, so a reviewer can see what you did not select and why.
4. Document it so a reviewer can re-perform it
- Source data: who extracted the ledger and trial balance, from which system, entity and period; file names and hashes.
- Completeness: the roll-forward result and how you resolved exceptions.
- Criteria and parameters: which criteria you applied, which you could not (missing fields) and why each threshold is what it is.
- Selection: the rule for key items and the random seed.
- Testing: for each selected journal, the evidence inspected, the authorisation and the business rationale, and any exception.
- Conclusion on management override (ISA 230: what was done, by whom, the evidence and the conclusion).
"We tested the journals the software selected" is not enough on its own: the file should explain the selection criteria and the risk scoring. Journalscope's method note is written to be saved in the file for that purpose.
Three ways to do it, compared
| Excel filters and pivots | Audit data analytics platform | Journalscope | |
|---|---|---|---|
| Cost | Included with Office | Annual subscription, priced by the vendor | $790 a year for up to 5 auditors ($474 founding price until 30 Nov 2026) |
| Completeness roll-forward | Built by hand per client | Usually included | Every account, with exceptions listed |
| Criteria and risk score | Filters you write each time | Configured in the platform | 19 documented criteria with weights; thresholds you set |
| Reproducible selection | Hard to re-perform | Usually | Key items plus a seeded random pick |
| Workpaper | You build it | Platform reports | 9-sheet Excel workpaper with evidence columns |
| Where the client's ledger goes | Your computer | Usually the vendor's cloud | Stays in your browser |
Excel works for a small ledger if you have time to build and document the tests. A platform suits firms that want one tool for many analytics. Journalscope does one job: journals testing for ISA 240, with the completeness check and the workpaper.
Try it free
Run the free sample or load your own ledger: completeness, flagged journals and the selection show on screen without sign-up. Exporting the workpaper and CSVs needs a licence: Firm $790 (up to 5 auditors), Practice $1,490 (up to 25), Network $2,900, each one payment for 12 months; $474, $894 and $1,740 with FOUNDING40 until 30 Nov 2026.