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

Exception Management Overview

Important:

Exception Management (EM) 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 Exception Management, contact your NetSuite Account Manager for guidance and next steps.

The Exception Management (EM) feature helps you identify transaction outliers, unusual activity, or other changes to your normal transaction patterns. EM highlights potential incorrect transactions, missing transactions, and suspicious changes to vendor data so issues can be quickly addressed before period close.

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

Exceptions appear on the Exception Management page, organized into two tabs: Transaction Errors (Incorrect Amount, Incorrect Account, and Vendor Information Change) 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:

EM 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, EM 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 Exception Management AI Models

The Exception Management AI models improve over time by learning from your historical data and your actions. When you mark an exception as No Action Needed, you indicate to 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.

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.

Similar exceptions may reappear until the model has enough consistent signals. The models and feature experience will continue to evolve based on user feedback.

Related Topics:

General Notices