I built MotorMetrics because I needed it myself

Years later, in 2024, I did a bit of driveway dealing: buying cars at auction, tidying them up, selling them on. Pricing each one properly meant hours of looking things up, paying for lookups and valuations, and cross-checking listings by hand. Hours of it, to land on a number, then miss the car at the fall of the hammer anyway.

When I went looking for data to build on instead, I found the same thing everywhere: a number, no evidence, and licensing that stopped you doing anything useful with it. Nothing was transparent. So I started collecting the market directly.

Every day, every dealer site, every car. What began as a workaround is now 235,000 cars advertised for sale right now, more than 485,000 recorded dealer sales since March 2026, and more than 4 million price observations. All of it resolved to one catalogue and valued on one engine. The dealer platform, the valuation API and the market reports don’t come from three datasets. They come from that one.

I price from dealer cars only, because dealers are who I build for. Private and auction listings are useful to buy from, so I show them. But a private seller’s asking price never sets what a dealer’s car is worth.

MotorMetrics is small, and that’s deliberate. It’s me, in Perth, a plumber turned project planner, building with AI tooling because the alternative was licensing someone else’s opinion. Built carefully, one piece at a time, with every number traced back to the evidence that made it.

That shows up in how I work. I onboard dealers by hand. I publish how the numbers are made and how accurate they are. When the data can’t support a number, I say so rather than guess. And I show my working, always, because I remember what it was like being on the other side of a black box.

If you’re a dealer, the platform is built for your yard. If you’re building something on vehicle data, our API options are built for you. Either way, the number comes with the market behind it.