Historic Green Matching Enhancements

1. Improved Green Matching Accuracy:

The success rate of Green Matching for bank feeds has historically been lower than targeted. To address this, the process has been updated to include the following steps:

• Bank Transaction Navigation: Before initiating reconciliation, the robot now navigates to the Bank Transactions page for each account.

• Transaction History Extraction: The automation extracts up to 2,000 previous reconciled transactions (approximately 10 pages of data, typically covering around a year’s worth of transactions, or more for accounts with low transaction volumes), to find matches for the new transactions.

2. Advanced Name Matching and Cleanup:

If the initial name matching fails due to low confidence, the following steps are performed:

• The robot extracts the Bank Transaction name and applies multiple cleanup routines.

• If the remaining name contains six or more characters, all characters following the first number are removed.

• Special characters and the term “INV” (from the end of the name) are stripped, leaving only the company name.

Once cleaned, the bot performs a lookup of the company name in the extracted transactions, opening the most recent occurrence in a new tab. The invoice line is extracted and compared with the corresponding invoice reconciliation line. If a match is confirmed, the transaction is classified as a Green Match, and the system proceeds with the reconciliation.

Fix to Cash Coding

3. Error Handling in Cash Coding:

• An increase in errors was observed when the automation could not confirm whether any cash coding actions had been performed, particularly when there were no applicable transactions.

• To resolve this, selector logic has been enhanced to eliminate such errors and improve the accuracy of cash coding operations.