Insurance Companies Use AI to Make Payables Touchless

To take on the challenges of accounts payable (AP) invoice automation and the insurance submission intake, Kanverse has deployed a product that uses several technologies: artificial intelligence (AI), machine learning (ML), optical character recognition (OCR) and natural language processing (NLP).

“It’s actually a hyper-automation platform,” Kanverse CEO and Founder Karan Yaramada told PYMNTS. “AP invoice automation is only one use case for the platform.”

Compared to conventional systems, this hyper-automation platform reduces both the time to implement and the total cost of operations. While making changes to a conventional system requires bringing in an IT team or even the vendor, the Kanverse platform can change by learning on the job.

“In our case, users can turn on the learning mode and do the steps that they would do, and then the system will learn automatically,” Yaramada said. “So, you can pretty much avoid going to the IT team, bringing in a vendor, making the change, testing it and going through the whole nine yards.”

Eliminating Manual Work and Errors

This is particularly useful in insurance, a regulated industry that relies on complex paperwork. Using OCR alone simply takes the paper and converts it into a digital form, which is only about 20% of the solution.

A platform that also uses AI and ML, on the other hand, does much more. “Using AI and ML, you can teach the system, ‘okay, if the user checks this box, make sure these other 10 things are done,’” Yaramada explained.

Catching incorrect data before it goes downstream can save the expense of processing it later. Yaramada noted that it also frees up the people who are now processing this data to instead do things like analyze the business, sell or improve processes and efficiencies. It will also enable companies to reduce costs and eliminate paper.

Making 90% of the Process Touchless

Yaramada said the company is confident that the platform can make 90% of the process touchless. In the remaining 10%, where there may be anomalies or extreme cases, people can still get involved.

“Whether it’s AP invoice automation or the insurance industry submission process, there’s a lot of tribal knowledge that makes people critical in these processes,” Yaramada said. “I think it is unnecessary. People don’t have to be critical in the process; let the process run by itself.”

Although the product is still in its early stages, the company has already received “great feedback” from customers. Yaramada reported that an AP invoice automation customer said tasks that used to take days to process now take seconds, and that an insurance industry customer said the platform is going to differentiate them from their competitors because they can move fast now.

“They can deploy all these people that are doing these jobs to do the productive tasks and the more exciting, value-added kind of work,” Yaramada said.