Trust Account Anomaly Scanner

Paste three months of trust account transactions and your active matter list. The scanner surfaces patterns a qualified bookkeeper or trust accounting specialist should review.

Anomaly detection only. Not legal or accounting advice. Every flag must be reviewed with a qualified professional before action.

Trust ledgers are one of the most sensitive document types in a firm — they pair client names with dollar flows. The scanner doesn’t need real names to detect anomalies. Label transaction descriptions and matters as [MATTER A], [MATTER B], [CLIENT 1], etc. The pattern analysis (commingling signals, missed reconciliations, unusual amounts) all works on structure, not identity.

Date, description, amount, balance — one per line. Use matter placeholders in descriptions.

Matter placeholder, practice area, opening date — one per line

Excel, CSV, PDF export from your trust accounting software. Up to 3 files, 10 MB each.

Output will appear here once the tool runs.

Questions people ask first

What does the Trust Account Scanner flag?

Patterns across three months of trust account activity — commingling indicators, deposits without a matching matter, withdrawals without a contemporaneous invoice, reconciliation gaps between bank, firm ledger, and client subledgers. Each flag has a confidence level and a recommended next action.

Is this a substitute for monthly reconciliation?

No. Monthly three-way reconciliation (bank statement, firm trust ledger, client subledgers) is what your state bar expects, full stop. The scanner is a separate set of eyes on patterns the reconciliation may not surface — a deposit that matches the bank but is mis-attributed in the client subledger reconciles cleanly and still represents a problem.

What format should my transactions be in?

Plain text or CSV-style. Date, description, amount, matter (if your accounting software exports a matter column). Most case management systems — Clio, MyCase, PracticePanther, LeanLaw — export this directly. Don't paste client SSNs or bank account numbers; the scanner uses transaction-level data, not banking metadata.

Will it tell me my state's specific trust rules?

It is calibrated against ABA Model Rule 1.15 and the major variations across high-rule-density states (CA, NY, TX, FL especially). Findings cite the rule and recommend the next action. Where your state has a less-common variant, the flag still surfaces — you'll need to check your specific rule before acting.

What if I find a real problem?

Talk to your bar's ethics line or an ethics counsel before remediation. Self-reporting requirements vary by state. The scanner names the pattern; the response — whether and how to disclose — is a decision the scanner can't make for you.