- Anomaly Detection: How Alerts Are Raised and Resolved: how duplicate, amount, and field anomalies are flagged, and what dismissing an alert changes in later scans.
- How to Set Up Anomaly Detection Before Your First Close: judge whether each default detector earns its place on your ledger, and turn off the ones that don’t.
- How to Create a Custom Detector: write a rule for a threshold the built-in detectors don’t cover, using a worked example of large journal entries posted to cash.
- How to Clear the Anomaly Queue During Close: work the alert list down before you lock the period, and write dismissal reasons that improve how later scans score similar findings.
Anomaly Detection
Anomaly Detection
An hourly scan that flags duplicate, out-of-pattern, and unusual-account transactions, and the routine for working the alert queue before close.
Anomaly detection runs an hourly scan across posted transactions and flags likely duplicates, amounts far outside a vendor or customer’s recent history, and accounts or classifications a vendor rarely uses. Three detectors run, each on by default and each a single on or off switch for your organization, and you can add custom rules for thresholds of your own.
Alerts accumulate whether or not anyone reviews them, and nothing in the product prompts you to look, so this is the one AI feature with a recurring routine attached.
Last modified on September 14, 2026