United StatesFinance and insurance2023 – ongoingTrupanion
Machine-learning models that emulate a claims adjuster for direct vet payments and claims
Trupanion's claims models pay vet invoices at checkout in five seconds and automate nearly 40% of traditional claims
Problem
Pet owners usually pay the vet bill themselves and wait for reimbursement. Trupanion wanted to pay veterinary hospitals directly, which required deciding on invoices while the owner was still at the counter.
Approach
Trupanion built machine-learning models trained on millions of claims processed by its own staff. The models mirror real-world policy decisions and run inside its patented vet portal. Staff review a pet's first claim for accuracy, and later claims are handled by automation. The same approach is applied to traditional medical claims.
Results
Takeaway
Training models on years of adjuster decisions let the insurer move payment to the point of care. Manual review stayed in place for first-time claims.
Categories
- Industry
- Finance and insurance
- Business function
- Customer serviceFinance, risk and legal
- Innovation type
- Process innovationProduct and service innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
Trupanion's press release describes machine-learning models trained on millions of claims that mirror policy decisions in its vet portal. It states that over 60% of portal invoices are automated, an invoice can be paid at checkout in five seconds and nearly 40% of traditional medical claims are automated.
Full details are in the original sources.