Expense payment automation

Case study

Expense payment automation using UiPath & ABBYY

Our client is in insurance claims management, specializing in commercial property and casualty insurance. In this business, a centralized team of well-trained resources were handling Expense Payments, which is an integral part of their claims processing. A centralized team was involved in manual processing of around 100,000 invoices per month, using various tools and applications. In an effort to reduce manual errors and cut down the time taken for this process, our client was considering to implement robotic process automation.


  • Invoice received from various sources comes in different formats
  • Manual data entry and verification process – highly time-consuming
  • Use of multiple disconnected automation processes

Tech stack


Imaginea engineered an RPA solution to run the expense payment process automatically, with minimal downtime. Easy to deploy reusable components within the solution ensured that the process was quickly implemented wherever required.

RPA implementation was done in phases for easier adoption. The initial phase was focused on data extraction from expense invoices and the second phase was Payment automation.

How Our Solution Helped

Deployment of RPA reduced the users’ average handling time and resulted in efficiency gains

Overall approach

We decided to use reusable worklets for automation. These worklets are tiny independent testable workflows that can be easily integrated in any process. The worklets were used to replace manual intervention in the identified processor lines of the expense payments process flow. Our overall approach was to:

  • Define the Why – The first step was to identify the process that needs to be automated. Once that was done, the RPA Goal for the process was determined. In this particular process the accuracy was already high and hence the goal was to improve efficiency and reduce cost.
  • Assess the What – After determining the RPA fitment, the steps in the process were identified for automation. There were two steps where the data was being entered manually from a source to the destination. Apart from just copying of data, there was no value add in those steps. Hence they were identified to be automated. Also the payment was rule based and didn’t need human intervention and hence was a good candidate for RPA.
  • Define the How – The identified automation opportunities were classified based on the below criteria and were prioritized so that they can be handled in phases.


Our reusable worklets approach led to

  • Faster delivery
  • Make localized changes, which gives better maintainability of process code
  • Better process readability

In summary, RPA deployment resulted in

  • Increased process efficiencies
  • Improved outcome quality
  • Reduced transaction time and costs

Talk to us