By applying machine learning to analyze journal entries that need approval, Sage Intacct is helping finance teams save time while improving trust in their organizations’ data.

Sage’s new Outlier Detection Feature for General Ledger, now available for early adopters, uses machine learning to learn your company’s usual transaction patterns, evaluate transactions in the current approval cycle, and flag transactions that don’t match your historical patterns.

The outlier detection feature notices any unusual dimensions or combination of dimensions. Say, for instance, that transactions for your accounting department are always tied to your headquarters location (with “department” and “location” both being dimensions) but you’re asked to approve an accounting department transaction that’s tied to a branch office.

The algorithm will highlight the transaction and provide an explanation why it was flagged (in this case, because the department and location dimensions varied from historic patterns). The person reviewing the transaction will have the option to approve it, make a correction, or decline the transaction so it can be returned to the submitter for correction.

This automated review of outliers provides an added control measure to help ensure accuracy and consistency in transaction approvals.

And as the algorithm analyzes more transactions, its machine learning capabilities provide a deeper understanding of your historic patterns. This, in essence, makes the software smarter over time.

The automatic review helps the organization increase trust in its financial data by reducing errors, and frees up time that would otherwise be spent in identifying, correcting and documenting entries manually.

Contact us to learn more about how Sage Intacct helps you save valuable time and improve the reliability of your financial data by identifying outlier journal entries automatically.