Note: This is a public demo version, published for showcasing only.
The GL Reviewer is a free web tool that scans a general ledger and flags entries that look out of pattern. It’s aimed at small businesses and accounting teams that don’t have enterprise audit software but still want a sanity check on their ledger before close, review, or audit prep.
You upload your ledger, pick the tests you want to run, and the tool returns a ranked list of the lines that stand out, with a clear breakdown of which tests caught each one.
Example file
How to use it
- Upload your ledger. CSV or Excel, up to 5 MB. The data needs to be provided in the same data format as is the example file.
- Confirm the columns. The tool auto-detects which column is the document date, posting date, account number, debit, credit, and so on. You can adjust the mapping if anything is off, and map optional columns like Account Name, Journal Entry ID, Posting Time, User ID, and Description. Tests whose required columns are not mapped are automatically disabled.
- Pick the tests. Fifteen tests are available across five groups: time-based (weekend postings, after-hours postings, posting-vs-document date delta, period-end clustering), amount-based (top X largest, top X repetitive, round amounts, just-below-threshold), user and journal structure (user activity summary, reversed journals, risky account combinations, unbalanced journals), text-based (suspicious keywords, blank or short descriptions), and a multivariate Isolation Forest anomaly detector.
- Set parameters. Each test has its own knobs — weekend days, time window, threshold values, keyword list, and so on. Sensible defaults are pre-filled.
- Run.
What you get back
A multi-tab Excel workbook. The Info tab records the run parameters. The Convergence tab is the headline view: every flagged line ranked by Convergence Score, the number of independent tests that caught it. An entry caught by four tests is far more worth investigating than one caught by a single test. Each enabled test then has its own tab, with the criteria you set and the full underlying rows. Every row carries a Source_Row column so you can map any flagged line back to its position in your original file.
Risky account combinations
If you map both an Account Number and an optional Account Name column, the Risky combos test gives you a pair picker: dropdowns populated with every distinct account in your file, showing “Number (Name)”, for example, “4000 (Revenue).” Pick the combinations you’d consider unusual together (Cash ↔ Revenue for a B2B business, say), and any journal that touches both is extracted in full.