Journal Entry Testing with Caseware IDEA

Add this to your PBC request: "Complete journal entry listing for the [period] including: journal entry number, posting date, effective date, created by (user ID), account number, description, debit amount, credit amount, and approval status."

Most ERP systems (SAP, Oracle, NetSuite, QuickBooks) can export this. If the client uses QuickBooks, have them export the Journal Entry report as an Excel file.

Critical fields: You must have at least: date, user ID, account number, debit/credit amounts, and entry description. Without these, key tests won't run.

What you should see: A flat file with one line per journal entry line item (each debit and credit is a separate row). A mid-size client typically has 1,000–20,000 entries per year.

Launch Caseware IDEA. From the main screen, click New to create a new project. Name it using your standard engagement naming convention (e.g., "ClientName_2025_JE_Testing").

IDEA organizes work into projects — each project can contain multiple imported databases and analysis results.

1. In your project, click File → Open → Database (or drag the file into the project workspace)
2. Select your client's journal entry Excel or CSV file
3. IDEA opens the Field Definition screen — this is where you tell IDEA what each column represents

Field mapping (match these to your client's column names):

• Date field: map to the posting or effective date column
• Amount field: map to the debit/credit column (or total amount if pre-netted)
• Character fields: map user ID, account number, description, journal entry number

1. Click Finish — IDEA imports and indexes the dataset. For 10,000 entries, this takes 30–60 seconds.

What you should see: Your journal entry population displayed in IDEA's grid view, with column counts and field types shown in the status bar. Verify the record count matches what the client provided.

Benford's Law tests whether the first digits of your transaction amounts follow a predictable distribution (1 appears ~30% of the time, 2 appears ~18%, etc.). Unusual distributions may indicate manipulation or data entry issues.

1. In IDEA, go to Analysis → Benford's Law
2. Select the amount field and choose First Digit or First Two Digits
3. Click Apply — IDEA calculates the actual vs. expected distribution and displays a chart
4. Look for digits where actual frequency significantly exceeds expected — these warrant further investigation

What you should see: A Benford's Law chart showing actual vs. expected distribution. Minor deviations are normal; significant overrepresentation of any digit (especially 5, 7, or 9 — common manipulation targets) warrants a follow-up explanation.

Workpaper note: Export the chart (right-click → Save Image) and paste into your workpaper, along with your conclusion on whether the distribution is consistent with expectations.

1. Go to Analysis → Duplicates → Duplicate Fields
2. Select fields to match on: amount, date, vendor/payee, description — exact combination depends on what your audit program specifies
3. IDEA produces a list of entries where these fields match across multiple rows

What you should see: A flagged list of potential duplicates. Review each: some are legitimate recurring entries (monthly rent, depreciation); others may represent duplicate payments or recording errors.

1. Go to Analysis → Summarization or use a Filter → create a filter where the amount ends in 000 (e.g., amount modulo 1000 = 0)
2. Review entries over your materiality threshold that end in round numbers — these may indicate estimates or manual entries that bypass normal controls

Common IDEA filter syntax: In the filter window, enter @ROUND(Amount/1000, 0)*1000 = Amount — this selects all entries where the amount is an exact multiple of $1,000.

1. Unusual accounts: Create a filter to find journal entries that post to unusual account combinations (e.g., debit to an equity account, credit to a revenue account that shouldn't be credited by journal entry). Discuss with your senior which combinations are flagged in the audit program.
2. Late postings: Filter for entries where the posting date is after the period end date. Any entry posted to the prior period after the cutoff date warrants investigation.

IDEA filter for late postings: PostingDate > @CTOD("12/31/2025") — entries with a posting date after the fiscal year end.

Export each analysis result set as a separate tab in Excel (IDEA allows direct export). For each test:

• Record the population size tested
• Record the number of items flagged
• Document which flagged items were investigated and the conclusion
• Note which flagged items are exceptions requiring further response