Cross Insurance uses GenAI-powered document extraction to better serve customers

What was the challenge?

“As an insurance broker, our services team consolidates insurance policy details from multiple providers every time we respond to a customer request. We were doing this manually, inputting benefit attributes and plan rates from Summary of Benefits and Coverage (SBC) documents. Because those documents came from different insurance issuers, they were often set up with different file types and structures. This process was prone to inefficiencies and delays. We wanted to streamline and automate the process to minimize manual work, enhance data accuracy and get information to clients faster – ultimately improving their experience.”

Describe the solution delivered by the PwC community of solvers.

“We worked with PwC to help identify and implement a solution utilizing Document AI technology, powered by Google Cloud. This solution allows us to extract plan attributes from SBC documents automatically and upload SBC files directly from Salesforce to Google Cloud. Generative AI (GenAI) then automatically labels the files and extracts key data, significantly reducing the time it takes to train the document model. GenAI can identify attributes without prior labeling, resulting in enhanced flexibility and accuracy. The extracted information is then promptly returned to Salesforce, either in real time or through batch processing.”

How does the solution blend the strengths of technology and people?

“Automating these repetitive tasks allows our agents to focus on higher-value activities, such as data analysis and decision-making, that can lead to more precise quotes for customers. The data is right at their fingertips in Salesforce and they're confident that it is accurate. They’re now empowered to better serve customers and make better-informed decisions.”

Where or how did innovation and unexpected ways of thinking come into play?

“Advanced GenAI capabilities helps us extract necessary details from insurance documents — it made remarkable improvements in data accuracy and efficiency for our team. The project has also enabled us to lay the groundwork for future GenAI projects at Cross Insurance. Meanwhile, we no longer face the challenge of classifying, labeling and extracting data from complex and varied document structures. As a result, productivity and cost savings have improved and our clients are receiving quotes faster than ever before.”