Product Information

Invoice Keying

Problem Statement

The PI team spend a lot of time performing the mundane task of entering each line of an Invoice into a slow and temperamental system.

Process Overview

An invoice sent to the PI team needs information within extracting and uploading to the temperamental system. This was time consuming as it could have up to 200 lines that need processing individually in the system, taking up to 15 minutes per line due to the unresponsive system. Two extra streams of trackers were added 6 months after the project was live


It was possible to drastically reduce efforts and cycle times to extract essential information from an invoice and perform a match by using a combination of RPA and AI (Optical Character Recognition + Natural Language Processing). The data collection from the invoices was streamlined using RPA and entered into the system in the same way as a human at a much quicker rate.

Business Benefits

12 FTE

released for additional work