Making sense of complex data in seconds.
One of our clients, a leading accountant company was spending too much time matching invoices with bank statements.
They needed to check whether they had been fully or partially paid – before updating their accounting software accordingly. Also, many of their customers used handwriting with their payments, the matching was made even more difficult due to different writing styles and cases.
There had to be a better way, and we knew just what to do.
The project only took two months before being integrated in a fully operational client environment.
Since it was certain that there would be some tricky cases with the need for human support, the goal was to try to reach 100% production accuracy. We built a bespoke software that worked with 3 different sets of data that could be dynamically configured.
- The accounting software database,
- The invoicing system database and its own database for saving configurations and
- The rules.
Using these sets of data, the software was able to intelligently assign different values for each matched attribute, which were configured based on the business case of the customer. If a predefined threshold was met when adding up the points, the operation was marked as successful. Unsuccessful matches were automatically sent to a human operator for manual processing.
Following the implementation of the new system, it is now used on a daily basis to match hundreds of invoices with payments, every day.
Production accuracy has also increased from 95% to 100% depending on the business case.
Our client also managed to save 35% of overall employee time, thus handling more new customers with the existing staff. Also, since all datasources are dynamically configured, more accounting systems and invoicing software can be integrated with the software in the future.