RPA case – Automation of reimbursement approval for an insurance company


A nation-wide health insurance provider had a partially automated process for reviewing its member’s reimbursement claims. Analysts would perform complex operations involving file import, protocol verification, analysis of 40 to 50 cases in order to decide whether to approve or refuse the reimbursement claim.

They have on average 500 reimbursement claims per day that generate a serious bottleneck for the accounts payable dept.

The customer was using robots created with a commercial RPA tool, but their performance was slow. Part of the process was still done manually. All this with a very high cost of licenses and maintenance.

Actions and Results

We created a robot that does the same but on an end-to-end basis by using the Robot Framework and Python, a license-free and very flexible solution that runs well on different environments and technologies.

We established a benchmarking between the new robot and the old one. A process that before would take more than 4 minutes without going through all the stages, now began running everything in a little more than 1 minute.