With AI-based document processing, this top-10 commercial insurer drastically reduced manual invoice processing, modernized its tech stack, and re-deployed data-keyers to more valuable roles. The cherry on top? The company unlocked access to data that previously had been trapped in non-structured documents well beyond the extraction capabilities of their previous automation stack.
Addressing Legacy Technology and Disparate Systems
The Fortune 500 insurance company was used to doing things in the same way for decades, and had amassed multiple legacy systems to handle their processing needs. These separate systems didn’t talk to each other, requiring manual data entry to transfer information between them.
Additionally, the company’s existing RPA solution struggled to accurately extract data from invoices. These documents contained handwritten notes, stamps, and fax marks—imperfections that resulted in low processing accuracy, leading to further manual review and correction. And because invoice formats vary widely, they needed a solution that supported the extraction of data from the handwritten invoices from small vendors as well as the multi-page invoices provided by much larger corporations.
With corporate mandates in place and an intelligent automation team to support, the business set its eyes on intelligent document processing.
Seamless Integration with RPA
By integrating Hyperscience with Blue Prism, the organization’s existing robotic process automation (RPA) solution, the company saved time and money. While RPA automates well-defined, rules-based tasks, it is unable to unlock and lift unstructured document data (like the kind contained in PDFs or images). This is where Hyperscience came in.
With Hyperscience’s Blue Prism integration, Blue Prism’s digital workers can send the accounts payable invoices to Hyperscience for automated classification and data extraction. The structured data output files are then picked up and routed downstream.
This flexibility also helped optimize for the future. Instead of being locked into the rigid template design of RPA, they used the application’s built-in machine learning—training digital workers to recognize layout changes. This greatly decreased document processing times, as the solution handles messy handwriting and document imperfections with ease.
Automating with Greater Clarity & Control
With its legacy solution, the company’s reliance on manual document processing meant missing SLAs, which led to frustrated customers. As a result of implementing Hyperscience, the company can meet these SLAs, and employees who were previously restricted to data entry can pivot to more valuable work. As for better visibility into business processes, the agency’s internal audit team can now trace where each data point comes from, and can easily view the original document—something that was previously impossible.
Ease of use and customization options also influenced the decision to adopt Hyperscience. No longer were employees left waiting on IT or development resources to implement changes any time a vendor form was updated. Now, data center employees could collaborate with machine learning to implement the changes themselves. New levels of self-sufficiency led to more efficient processes, allowing the company to get more productivity out of its existing resources.
Watch this short platform demo to see how Hyperscience leverages proprietary Machine Learning to extract data from complex documents.