Outlier Assistant

After GL Outlier Detection is configured and running, journal entries routed through the approval process go through GL Outlier Detection for evaluation.

When you enable the Outlier Assistant, you tell Intacct to change the workflow by returning flagged entries to the submitter instead of routing them to the approver. This lessens the workload for approvers by ensuring that the submitter has addressed any adjustments needed from the outlier entries before sending them onto the approver.

Outlier Assistant was designed to be the only step in the approval process where it makes sense to do so. Therefore, even the smallest organization can benefit from Outlier Detection insights. Smaller organizations do not always have a large approval process—for example, everyone approves their own entries. These organizations can still take advantage of the machine learning features in Intacct.

With the Outlier Assistant, the evaluation process runs in the background. If a transaction is flagged as an outlier, the transaction is sent back to the submitter for re-evaluation.

From there, the submitter can adjust the entry, and it is re-evaluated. Or the submitted can resubmit the transaction without any change, indicating that the outlier entries are intended.

After a transaction is submitted, the Outlier Assistant performs the first evaluation. Entries without any outliers detected are routed as normal to the approver or posted for those submitters who can self-approve.

Enable Outlier Assistant

You enable Outlier Assistant for each journal that’s been configured for Approvals.

1. Go to General Ledger > Setup > Configuration.
2. Scroll down to Approval options.
3. In the Approvers list, for each journal to apply Outlier Assistant, place a checkmark in the Outlier Assistant column.
4. Save your selections.

The Outlier Assistant workflow begins immediately, as long as the Outlier Detection is in the Ready status.

What to look for when a transaction is flagged

Outlier detection uses historical patterns to flag unusual account-dimension combinations, irregular amounts, or both. Hovering over the outlier will explain why a particular line has been flagged.

If a submission is flagged as containing an outlier, the entry is returned to the submitter for next action, even if the submitter can self-approve or is the Admin approver.

If it is submitted again without change, it will be routed for approval or posted as appropriate based on the approval configuration settings.

Review the submission

GL Outlier Detection provides an additional data point that helps ensure accuracy in journal entries.

When a transaction is returned to the submitter for re-evaluation, the submitter can either:

• Make a correction and resubmit the entry, which will then be re-evaluated as a new entry by the Outlier Detection service.
• Determine that the entry is correct and submit again. This routes the transaction to the approver or posts it, based on the approval configuration settings.

The submitter's action becomes part of the learning for future GL Outlier Detection evaluations done on general ledger transactions. This enables Outlier Detection to learn continuously.