When the Paycheck Protection Program (“PPP”) was announced there were few details but there was one certainty – nearly every single business that qualified would be in line to sign up. As the launch date approached banks scrambled to get processes in place, small business CFOs got their paperwork in order, and the Small Business Administration (“SBA”) exclaimed their application and management system ‘E-TRAN’ was ready to roll. Then, E-TRAN crashed.
As soon as the PPP hit the news, banks immediately started to contact technology companies. The volume that would normally take place over several years would need to be accomplished in a span of weeks. Some banking platform providers had built bolt-on modules that could help streamline the application to approval to E-TRAN submittal to funding steps. Others turned to RPA. The bots were built up to follow the same steps: grab applicant data, log in to E-TRAN (a web-based system), submit the loan to the SBA, get an approval, and move on. Leading up to the PPP start-date, the new modules and the bots were performing their tests as expected. Loans were submitted and test-borrowers were getting their SBA loan numbers within minutes. However, when E-TRAN crashed all of those supporting technologies came to halt as well.
One of the tremendous features of an RPA platform is the ease in which bots can be programmed and deployed. When the normal E-TRAN process stopped those modules and bots were of no use. Fortunately for the RPA client, new bots were designed and put into production the same day to come up with a work-around. For this specific client, bots were built to take in the same exact data files that were designed for the original process and turn those into XML files (a text format ingested directly by databases). The loan originators could save the application data files to a folder, a bot would pick them up, convert those into an XML file that E-TRAN could read, and then submit those to the SBA through a different method. The bank went from submitting 5 applications an hour to 50.
While we always recommend a robust development lifecycle, there are crucial circumstances that call for extraordinary measures. The lesson remains the same – RPA can adapt to the environment and processes at-hand. The business users and developers were able to think through clever solutions in the moment and build from there. Unfortunately for those that used the banking module (through no real fault of their own), they were not able to pivot as quickly.
As we continue to work with clients, this flexibility proves valuable time and time again. Every company has their own systems, configurations, and processes and RPA is a tool that can accommodate and improve upon that foundation.
