One dataset. Five ways to use it.
The same per-vehicle evidence base serves very different jobs. Each of these is a real workflow tied to a concrete number the console already produces.
Check the 5-year outlook before you negotiate.
Before buying a used 2019 Corolla, you pull its 5-year component outlook and see suspension issues typically appear around 11 years / 130,000 km — well after your intended ownership — while the transmission risk is low. You walk into the negotiation knowing the real forward repair exposure, not a gut feel.
Design a term that excludes the known early-failure part.
Pricing an extended warranty on a Ford Focus, the cross-market signal flags the PowerShift transmission at ~9× market. Instead of declining the whole book or over-charging, you use the coverage builder to design a term that excludes that one component — and price the rest on its real burn cost.
Reliability badges that lift warranty attach.
Every listing shows an evidence-based grade and an expected repair cost. Buyers trust a badge that traces to real inspection data; you attach an embedded warranty priced per vehicle at checkout, and durable models stop subsidising the trouble ones.
Forecast next year's maintenance budget across a mixed fleet.
You run a mixed ICE/EV fleet through the portfolio endpoint and get cost-per-1,000 km and a five-year maintenance forecast, with the parts-cost-inflation trend layered on top so the budget reflects where repair costs are heading, not just where they are.
Portfolio-level visibility across cedants.
Score an in-force book to see its risk distribution, the twenty vehicles driving the loss ratio, and the exclusions that would have paid for themselves — a shared, evidence-based view across cedants. (Aspirational at pilot scale; the portfolio scoring that underpins it is live.)