Make and model search behaviour, high-volume low-margin enquiries, and a category where programmatic page architecture beats hand-built pages.
Vehicle searches are structured in a way almost no other category is. People search by make, model, year, part, condition and location, in combinations — and each combination is a real query with real intent. That structure is a genuine opportunity: it can be covered systematically rather than guessed at.
It's also a category where enquiry volume is high and margin per job is often thin, so cost per lead discipline matters more than in most industries. A $90 lead is fine for a workshop retaining a customer for years and fatal for a business paying for scrap vehicle collection.
Build the page architecture to match how people search — make, model and location combinations generated from a data model rather than written one at a time, with the discipline to skip combinations that can't support genuinely distinct content. This is exactly the programmatic page generation work we do, and automotive is the category it suits best.
Alongside that: qualification built into the enquiry form so vehicle details arrive with the lead, tight geographic targeting matched to your actual collection radius, and offline conversion imports so the platforms learn which enquiries became completed jobs rather than which became form fills. In a thin-margin, high-volume category that last step is often the difference between profitable and not.
Cost per completed job, segmented by vehicle type where the economics differ. Some vehicles are worth chasing and some aren't, and once the measurement can tell them apart, the budget can follow.
It is if you generate everything. Thin, near-duplicate pages across hundreds of make-model combinations are a fast route to being classified as doorway spam. The discipline is refusing to generate combinations that can't carry genuinely distinct, useful content — which is a data-model decision made before anything is built.
Usually the enquiry is too easy and too vague, and the ad platform is optimising toward volume because that's the only signal it has. Adding qualification fields and feeding completed jobs back as offline conversions changes what the algorithm is chasing. Enquiry count typically drops and booked jobs rise.
Whatever matches your actual operating radius. Paying for enquiries you have to turn down is worse than useless — it trains the platform toward the wrong audience while costing you money.
Very, in the main centres, and much less so in specific make-model-location queries. That gap is the whole strategy.
Book a free discovery session. We'll audit what you're running, tell you what's leaking budget, and show you what we'd do differently — no obligation.