GL Outlier Detection overview

Sage Intacct GL Outlier Detection is a Machine Learning (ML) service that uses your historical transaction patterns, evaluates current transactions in the approval cycle, and flags transactions that do not match.

Working with General Ledger (GL) journal entry approvals, GL Outlier Detection shows an indicator when something looks out of the ordinary. This provides an added control measure to help you ensure accuracy in transaction approvals.

Transaction patterns are different for every company, and GL Outlier Detection learns from your specific data. As more data accumulates and you respond to outlier notifications, GL Outlier Detection evolves and adjusts the evaluation when detecting outliers in your transactions.

What's GL Outlier Detection?

When a transaction is sent for approval, GL Outlier Detection uses ML to help approvers identify transactions that are outside the organization’s historical pattern. Journal entry Approvers see an Outlier column in when approving entries: 

• Outliers are indicated with an exclamation mark (Outlier).
• Non-outlier transactions have a dash (Nonoutlier).
• For not-yet evaluated transactions, the column appears empty.

For example, transactions for the Accounting department are always tied to the HQ location. But a transaction is sent for approval with the Accounting department and the Alaska location. (Both Department and Location are dimensions.) Outlier Detection notes the discrepancy and flags the transaction as an outlier, giving you, as an approver, another data point for evaluation of the approval.

Use Outlier Detection as an assistive technology to take a proactive role in the approval process. With Outlier Detection, it’s easier to identify transactions that are outside the normal pattern of transactions. Outlier Detection does much of the work for you. Quickly see if a submission needs to be examined more closely before being approved.  To get details about why it’s an outlier, hover over the Outlier. You’ll see the reason that the transaction is flagged—for example, Department A is not typically paired with Location B.

As always, it’s up to the approver to decide the correct action: 

• Approve the transaction as-is
• Make a correction directly
• Decline the transaction to send the entry back to the submitter for correction

Commonly used terminology

There are some terms used with AI/ML and Sage Intacct GL Outlier Detection that you might not be familiar with. It’s helpful to understand these terms as you begin to use Outlier Detection.

Term	Description
Model	Creating a model (or modeling) is the process of using historical data from your company to build ML algorithms. The algorithms can then provide you with the ability to approve or not approve a transaction based on known historical data trends.
Evaluation	To detect outliers, each transaction is evaluated. Outlier Detection scans the data and notes any item that’s outside the historical norm.
Detection	When a transaction is evaluated and lies outside the historical norm, GL Outlier Detection can determine that the transaction is an outlier. You're then notified that this transaction needs further review by a human approver.
Outlier	An outlier in this case means a transaction submitted for approval that does not match the general pattern seen historically. Basically, it’s an exception or combination used outside the norm. Outliers usually require a closer look to ensure correctness.
Predictive	GL Outlier Detection evaluates data and considers whether a human approver found a transaction to be an outlier. It uses this information to predict whether other transactions are outliers.

Outlier Detection workflow

The following shows how GL Outlier Detection works, from start to finish.

Key features

Outlier Assistant

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

The Outlier Assistant enables a change to the workflow by sending flagged entries back to the submitter for re-evaluation instead of notifying the approver of the flag. This significantly lessens the workload for approvers and ensures that the submitter is aware of the possible issue.

More about GL Outlier Detection

Can I rely only on GL Outlier Detection for approvals?

Not entirely. GL Outlier Detection is a helpful tool when approving journal entries, but no tool—not even AI—is foolproof. Use the same vigilance that you always have when reviewing and deciding how to handle a submission.

Even if a submission was not flagged as an outlier, review that transaction before approving or declining it. The ultimate decision on which action to take is yours. GL Outlier Detection is just another tool in your toolbox.

How does GL Outlier Detection flag entries?

GL Outlier Detection uses your historical data to identify patterns in your General Ledger journal entries. It flags incoming entries that do not conform to those patterns. Patterns are typically established on a per-journal basis and largely based on standard fields of an entry, such as account, key dimensions, and amount.

GL Outlier Detection is dynamic, meaning it continually learns as new data becomes available and new patterns emerge. For example, a transaction is flagged at first, but then a new pattern is recognized. Entries conforming to the newly identified pattern are no longer flagged as outliers.

What does it mean when an entry is flagged?

If a General Ledger journal entry is flagged as an outlier, the algorithm has determined that the entry is unusual based on the available data and the currently identified patterns. The reason that the entry is flagged can be viewed by hovering over the outlier flag within the entry. The reason it's flagged could be:

• It uses a value for a dimension that’s not typically used in this journal.
• It uses a new type of entry that’s never been seen in the data before.
• It does not conform to other patterns identified by the algorithm.

GL Outlier Detection algorithms are based on common transaction practices and consider relationship patterns in historical data. In some cases, your business practices can evolve or be different for a legitimate reason that GL Outlier Detection has not identified as a pattern. Consequently, GL Outlier Detection flags some entries that you consider normal. The flag is an additional tool to guide you to transactions that require extra scrutiny.

What does it mean when an entry is not flagged?

There could be many reasons why an entry is not flagged or needs extra attention. This can be due to a policy change within your business, for example, or another type of issue that the algorithm has not yet detected.

GL Outlier Detection is another data point in your review process, which can help when you review journal entries, flagged and unflagged, for approval.