ChallengeAs in any financial organization, Pochta Bank's business processes are closely tied to paper correspondence. Every day the bank sends thousands of emails to its clients. Before sending these emails, an employee needs to process them correctly and register them in the internal document management system. This task includes:
• Recording the recipient, addressee, performers and the date of when the message was created.
• Putting the registration number and date of dispatch on the document.
• Scanning the document and entering the data in the digital card.
Departments of four employees performed this routine work. To relieve people of this routine activity, speed up the process and reduce the number of errors, the management decided to automate the process of registering emails. There were no RPA developers in the bank's IT department, so the customer turned to Sibedge experts.
ApproachThe automation system was developed in several iterations. Initially, it was planned to robotize the processing of correspondence by configuring Microsoft Word templates and collecting the necessary data before signing and printing documents. However, this method turned out to be extremely inconvenient because the emails were constantly edited, which made the data no longer relevant. The idea had to be abandoned.
The second attempt was made after conducting research in the field of OCR. Settings were carefully set, and a robot was trained to scan documents of different quality. As a result, the system could successfully recognize 100% of the test documents, and extra data was discarded. At the implementation stage, it turned out that there were many more document templates than originally expected. The robot could not be fully deployed. A fundamentally different approach was needed.
The third iteration was the creation of a web application for outgoing documents processing by the bank employees according to the specified rules. Each email had a unique QR code assigned to it and a link to a card that was automatically generated in the document management system. In the new system, after making changes to the document, one only needed to scan their QR code and the necessary data would be added to the digital card automatically.
ResultThe RPA system was developed using the bank's equipment. The chosen solution was experimentally implemented in one of the bank's local offices, where the RPA experts equipped a workplace for the robot. Based on the results of the automation system implementation, the following conclusions can be drawn:
• The robot processes emails at least four times faster than a human.
• You only need one employee to support the robot.
• The speed of internal business processes has increased significantly.
• The number of errors in document processing has decreased.
Three Sibedge specialists worked on the project for nine months. The customer was satisfied with the results because robotization reduced costs and freed many employees from routine work. Now the system is being implemented in other offices of Pochta Bank.
- 1 - Project Manager
- 1 - RPA expert
- 1 - Analyst