Interested in working together?
Please contact us if you have any questions or want to learn more about how we can work together. You can download the PDF fact sheet and share it with your colleagues or leadership team.
A global insurance broker sought to automate processing of unstructured insurance policy quotation documents to extract key data from documents faster and more accurately than by current human operators. The company also wanted to mine insurance quote and policy data for trends and data monetization through value-added client-facing analytics.
SSA & Company ingested a sample of each contract type and used Natural Language Processing (NLP) to understand patterns and structure of variability. The team then created a predictive model to parse specific syntax, which was tested and used on live documents. The NLP fed a workflow tool for human users to review the machine’s outputs and predictions. Humans corrected any errors and the algorithm learned the corrections to better its accuracy. Once the human approved, the app automatically generated various output documents in excel and PDF.
This automated process led to 75% reduction in document processing time, resulting in ~10 month ‘break even’ and 6x ROI over two years. Additionally, it increased accuracy and form data completeness by 17% increase, resulting in ~$8M reduction in financial exposure.