In the face of processing millions of vital documents amidst workforce limitations and increased demand due to the pandemic, one U.S. state sought a transformative solution for its SNAP and MAGI review documents.
States typically process millions of pages of paper applications, Renewals, and/or Periodic Reporting forms every year. This has become even more challenging for States due to a combination of limited workforces and an increased population of benefit recipients due to the pandemic.
One state in particular was processing millions of SNAP and MAGI review documents. These documents needed to be indexed and identified, tagged with the right case numbers, and updated when new information was available. As a result, the State was leveraging Eligibility Technicians, a valuable resource, to do cumbersome data entry which in turn was leading to slow turnaround times and incorrect decisions due to human error. With the upcoming end to PHE and the redetermination deadlines fast approaching, the State was anticipating hiring 35 additional employees to help with the influx in volume, so was looking for fast, reliable, and accurate ways to speed up their processes.
The State implemented Hyperscience’s Hyperautomation solution to automate document identification, data extraction, data validation, and data entry into the State’s existing Eligibility and Enrollment system.
- Highly accurate automation of data identification + extraction – With Hyperscience the State was able to set a required accuracy level and achieve significant automation of time-consuming data entry tasks
- Performance on handwriting – Hyperscience was able to accurately and effectively extract from low quality documents with messy handwriting, common attributes for the types of documents the State was receiving
- Easy to configure validation rules – Hyperscience provided tools for the easy setup and deployment of validation rules within Hyperscience to validate critical data points
When public health emergency (PHE) provisions began to wind down in spring 2023, the state was prepared. One PHE provision had waived the requirement for benefit recipients to annually update evidence of SNAP eligibility. When the waiver expired and recipients were required to recertify, the state’s monthly volume of images that needed processing soared by 700% to 1,000%.
With Hyperscience, the state was able to automate 99.2% of the fields that previously required manual keying, the accuracy rate registering at 99.4% for handwritten information. As a result of the efficiencies gained through Hyperscience, the State no longer anticipates hiring the additional 35 employees to meet the upcoming demands and can reallocate the budget. Lastly, the success of the project was recently recognized at the ACT-IAC Annual conference as a recipient of the “Innovation Impact Award” and the State continues to expand to additional use cases such as MAGI and SNAP application and verification documents.
Watch this short platform demo to see how Hyperscience leverages proprietary Machine Learning to extract data from complex documents.