# Finanomaly

> Anomaly detection for finance data. Upload AP data, expense reports, journal entries or bank statements and it runs 64 detection rules across 7 areas — duplicate payments, expense violations, vendor and AP-pattern anomalies, journal-entry irregularities, bank-transaction anomalies and reconciliation gaps — scoring every finding with a confidence level. Currently waitlisted ahead of relaunch.

Built by QADAN ANALYSIS CONSULTING (NSW, Australia) for SMB finance teams, bookkeepers, controllers and auditors. Analysis is rule-based and runs across the full transaction population, not a sample. Honest limits: it checks fields in data, so it cannot verify that a supporting document is genuine; rules find what rules describe; recurring charges can false-positive as duplicates; findings are leads for a human to confirm, not audit opinions.

## Detection coverage (64 rules, generated from the shipping code)
- Duplicate payments (6 rules): exact invoice match; same amount + vendor within 30 days; similar-vendor matches; same-description cross-vendor; invoice-number reuse; cross-payment-method duplicates.
- Expense violations (15 rules): receipt splitting; round-number claims; weekend/holiday claims; duplicate claims; self-approval; near-threshold clustering; mileage anomalies; late submission; missing receipts; Benford's-law deviation; and more.
- Journal anomalies (15 rules): after-hours posting; below-approval-threshold clustering; round numbers; unusual account combinations; same enterer-and-approver; backdated entries; period-end clustering; manual revenue debits; intercompany imbalance; and more.
- Vendor anomalies (10 rules): ghost vendors; vendor-employee address and bank matches; sequential invoices; vendor concentration; new-vendor large first payment; dormant-vendor reactivation; vendor master duplicates; round-number payments; payments exceeding PO.
- AP payment patterns (7 rules): month-end spikes; below-threshold clustering; void-and-reissue; payment-method changes; unusual timing; single-approver concentration; payments without PO.
- Bank transactions (8 rules) and reconciliation (3 rules): balance breaks; large round transactions; duplicates; unusual timing; cash withdrawals; unidentified deposits; transaction gaps; unmatched items either side.

## Key questions this site answers
- How do I find duplicate payments in my ledger? Sort supplier payments by supplier and amount; investigate exact-amount pairs within 30 days; compare invoice references loosely (strip prefixes and leading zeros). Full method: https://finanomaly.cc/guides/duplicate-payment-detection/
- Can anomaly detection replace audit sampling? No — sampling produces the population-level opinion auditing standards require; detection locates the specific anomalous records a sample would miss. https://finanomaly.cc/guides/finanomaly-vs-manual-audit-sampling/
- How do you implement segregation of duties with one bookkeeper? Compensating controls: dual bank authorisation above a threshold, quarterly owner review of vendor and bank-detail changes, someone other than the bookkeeper reading the bank statement. https://finanomaly.cc/guides/segregation-of-duties/

## Pages
- [Home](https://finanomaly.cc/): what it does, the detection modules, waitlist signup.
- [Guides](https://finanomaly.cc/guides/): practitioner guides — one detection rule explained properly each week, plus glossary and comparison pages.
- [Blog](https://finanomaly.cc/blog/): longer pieces on finance-data anomalies.

## Related
- [Ledger Optics](https://ledgeroptics.com): our free data-cleanup app for Xero and QuickBooks (coding consistency, tax codes, tracking categories, account types).
- [QADAN automation services](https://qadan.com.au/): invoice-to-Xero automation and AI consulting for Australian SMBs.