Foundation AI helps Sweet James scale by capturing and processing all of its incoming documents with a fraction of the labor

With over $2B in settlements and verdicts, Sweet James Accident Attorneys is one of the fastest growing and most successful law firms in history. The firm receives and processes over 80,000 individual PDF documents totaling nearly 400,000 pages per month, including insurance correspondence, court notices, and medical bills and reports. These documents flood in through snail mail, centralized email boxes set up to capture faxes and service documents, and email attachments received in the individual email boxes of its 450+ attorneys and staff members. Each has to be matched to the right client file, named properly, saved to the correct folder, and tasked to the right person. For many documents, like medical bills and reports, staff then enter specific information, like damages, into the Litify practice management system. As the firm scaled, its manual processes were proving slow, costly, and error-prone.

The Challenge

With over 450 lawyers, paralegals, and pre-suit case managers each receiving important documents via email, it was nearly impossible to ensure that everything was properly entered into Litify. “We found that every time we had some turnover, there would be documents in someone’s email box or saved to their computer that had never gotten entered into the system,” Steve admitted. One day, Mehr found out that an attorney nearly resolved a case off-policy until the firm located an otherwise missing $30,000 medical bill at the eleventh hour. Mehr knew that he needed a scalable solution to capture all of the firm’s incoming documents and process them consistently.

The Approach

By implementing Foundation AI, Sweet James has completely freed its US teams from inbound document processing. The FAI Platform syncs with Sweet James’ Microsoft systems and automatically pulls in all scanned mail, and any emails with attachments from across the firm’s mailboxes. The AI first matches each document to the correct matter. Next, it classifies each document to one of over 100 trained Document Categories. Based on the Category, the FAI Platform automates document-driven workflows like naming and saving the document to the correct subfolder in Docrio, extracting specific data and entering it into Litify, summarizing information, and dynamic, role-based tasking.

Roughly 70% of the documents are processed end-to-end with no human intervention. For the remaining 30%, Sweet James’ document validators use the Foundation AI user-interface to review any decisions or extractions where the AI was uncertain. Mehr explains, “they’ll review when the system can’t narrow down to the right matter, or make changes to the name or verify an amount if necessary. Then, once they validate, the document goes straight to the right place and it kicks off all of our alerts, workflows, and automations.”

The results speak for themselves. Sweet James is processing incoming documents same day, which it was never able to achieve, and it is doing so with 80% less labor hours. As Mehr puts it, “Foundation AI has completely transformed the way that we deal with incoming documents. Every single document is captured and handled the same way, including the attachments sent to the staff. We are down from 28 people doing this, including several well-paid team members here in California who can now focus on our clients, to only 6 document validators. We’ve removed human error from the equation and we have clean data in all of our files.” Mehr continued, “A lot of scaling has to do with scaling technology and really understanding your data. If you’re not capturing 100% of your documents and your data, I don’t see how you survive against firms that do.”