1. Accounting
  2. General Accounting
  3. Financial Exception Management
  4. Financial Exception Management Overview

Financial Exception Management Overview

Important:

Financial Exception Management (FEM) is currently a limited release feature that’s not available to all customers. Eligibility depends on data volume and model training readiness; accounts with a large amount or insufficient historical data may be restricted. If you can’t access FEM, contact your NetSuite Account Manager for guidance and next steps.

Additionally, FEM is a production‑only feature and can't be enabled in Sandbox Accounts because those environments typically lack the required historical transaction data (approximately 18 months) needed for model training and inference.

The Financial Exception Management (FEM) feature helps you identify transaction outliers, unusual activity, or other changes to your normal transaction patterns. FEM highlights potentially incorrect transactions and missing transactions for your review, so issues can be quickly addressed before period close.

FEM uses customer-specific machine learning models trained on your organization's historical data to understand your transaction patterns. For optimal results, FEM learns from your last 18 months transaction activities. When the feature is enabled, every hour, FEM reviews transactions created or edited in the past hour and creates exceptions with supporting context when activity appears out of pattern, and an alert is sent to the controller or a person who created the transaction.

Exceptions appear on the Financial Exception Management page, organized into two tabs: Transaction Errors (incorrect amount, incorrect account) and Expected Transactions (missing transactions). The list prioritizes likely issues and provides details explaining why an item was flagged and a short list of similar, error-free transactions for comparison. From there, users can open the underlying transaction, resolve the exception after correction, mark No Action Needed, or reopen as needed. For missing transactions, users can optionally create the expected transaction to fill the gap.

Important:

FEM works best when you have consistent historical data that reflects normal business activity. Performance may be limited for test accounts, new accounts, or accounts with insufficient historical data. Over time, as you resolve or dismiss exceptions, FEM improves its understanding of your patterns and more accurately surfaces items worth investigating.

Privacy and data isolation: Each customer’s models are trained only on their own data and are not shared with other customers or third parties

The Financial Exception Management AI Model

FEM improves its suggestions over time by learning from your historical data and your actions. When you mark an exception as No Action Needed, you teach the system that the behavior is expected, and when you resolve an exception after correcting a transaction, you confirm it was a true issue. Both negative and positive feedback help refine future suggestions.

At times, FEM may surface low-confidence suggestions to gather feedback. Similar exceptions may reappear until the model has enough consistent signals, therefore, consistent user feedback helps the system recognize these patterns and reduce resurfacing over time.

Models retrain regularly to incorporate recent activity and feedback - weekly for Incorrect Amounts and Incorrect Accounts, and at the start of each accounting period for Missing Transactions.

Model and feature experience will be evolving over time based on user feedback. Therefore, consistent feedback is valuable to the system.

Related Topics:

General Notices