Google Ads for car dealerships

Some campaign types are, and the question cannot be answered about Google Ads as a whole. Branded search converts extraordinarily well and often buys clicks you would have had free; non-brand search reaches people who do not know you; Vehicle Listing Ads depend on your feed. A single yes or no describes an average of five different things.

There is no comparable benchmark, because any cost-per-lead figure is a blend of five campaign types in one store's proportions. A store running mostly branded search reports a fraction of one running mostly non-brand, and neither number says which is better run. Build cost per booked appointment by campaign type instead.

Ads assembled from your inventory feed that show a specific unit to somebody searching for that kind of vehicle. Because the ad is built from the feed, performance is decided by feed accuracy (price, trim, photos, availability) far more than by bidding.

Because your feed is syndicated to destinations that have the right to advertise the inventory they receive. It is a syndication and contract question instead of an ad account one. Establish which of your feed destinations may advertise your units, and on what terms.

Some defensive spend where competitors bid on it, yes. Not the share most stores give it. Test it directly: reduce the branded budget and watch total inquiries instead of paid inquiries. The gap between the two is what you were buying.

Search engine marketing is the broader activity of buying visibility on search engines; Google Ads is the platform most dealership SEM runs on. In practice, when a vendor says SEM they usually mean Google Ads plus whatever else they place, so ask which platforms are included.

It can be, and it reliably absorbs branded search and reports it as its own performance. Ask for brand exclusions and for the branded share of its conversions before judging it. Without those two numbers you are comparing it against campaigns it is quietly borrowing from.

They serve different terms well, so the useful version is a deliberate split: paid search on the competitive terms you will not rank for, organic on the ones you can own. What happens at most stores is paid bidding on terms already ranked first organically, because the two vendors report separately.

Ask what they cut last quarter, what share of conversions are branded across every campaign including Performance Max, and who owns the Google Ads account. The third is the most consequential contract term in paid search and is almost never discussed while it can still be changed.

The store. If the account belongs to the agency you cannot see history, cannot audit spend, and leave with nothing when the relationship ends. Agency-owned accounts are common and the fix is trivial before signing and awkward afterwards.

Accuracy before anything else: the advertised price matching the site price, correct trim and drivetrain, real photos of that unit, and sold vehicles leaving the feed promptly. Most Vehicle Listing Ads underperformance is a feed defect that no amount of bid work can reach.

Many are spending too much on one part of it, branded terms and terms they already rank first for organically, while underfunding non-brand search that reaches people who do not know them. The total is usually less wrong than the distribution.

Ads and AI answers are separate systems: paid placement does not make a store more likely to be named in a generated answer. Being nameable there depends on pages a machine can read and business facts it can confirm, which is car dealer GEO work instead of paid media.

They are modelled estimates instead of counted visits, which makes them useful for direction and unsuitable as the number a budget decision rests on. Treat them as one signal alongside booked appointments and answered calls, never as the headline.

Days: that is what you are buying, and it is the honest advantage of paid over organic. What it does not do is compound, so the comparison worth making is not how fast each moves but what remains six months after each one stops.