Case study · Large U.S. health insurer
Cutting manual claims review by 95%
Clinical staff reviewed hundreds of millions of claims a year by hand. We automated the workflow, routed it through an AI engine, and took it from interviews to production in about six months.
- people interviewed who touch claims every day
- 50+
- claims in the proofs of concept
- 10–20M
- less manual processing
- ~95%
- from interviews to production
- ~6 months
The engagement
Decide, build, and stay.
01
Decide
We interviewed 50+ people who touch claims every day. We mapped where cost and member experience broke down, which steps an engine could own, and which needed a clinician. The client funded the build on that case.
02
Build
We ran proofs of concept across lines of business on 10–20 million claims. Manual processing dropped by nearly 95%. The engine sends exceptions to a person rather than deciding them alone.
03
Stay
Once the business case was proven, we drove the POCs into production and handed the system to the client’s team for full deployment. They own the code and the records.
Also for this client
- A new member-facing product for medication adherence.
- Member self-service run by an AI agent.
Contact