Spray and pray went upstream

The industry spent a decade lecturing dealers about spray and pray. Blanket the market, measure nothing, argue about the invoice. Then the accountability moved one tier up, and something awkward happened: tier one didn't follow it.

A dealer today cannot draw a match-fund dollar without an approved vendor, compliant creative and sales attribution reporting built into the campaign. GM's dealer-side program is explicit about this: in-Market Retail funds reimburse digital advertising only when the work fits the program's rules, and GM's vendor requirements include sales attribution reporting. Prove it or fund it yourself.

Meanwhile the national tier runs the same model it always has. Brand spots, sponsorships, model-launch flights, a lot of reach. And when the flight ends, the dealer who is supposed to convert that demand receives a deck of impressions.

What follows is the pattern, not a documented program rule: the national flight lands on the brand site, not the store's. No landing page on the store's own domain, no data capture, not even a form, no delivery survey asking the buyer what actually brought them in, and no market-level file the dealer can join against its own sold units. Where a specific program does hand over a file or a survey, this argument does not apply to it; the criticism is that such programs are the exception.

That is the asymmetry worth naming. The tier that spends the dealer's money demands attribution. The tier that spends the brand's money hands over exposure counts.

Exposure is not nothing. It is just unfalsifiable

Brand media does work. Model launches move consideration, sponsorships build familiarity, a well-bought national flight lifts branded search in every market it touches. Nobody credible argues otherwise.

The problem is that the lift arrives at the store as an assertion, not a dataset. The dealer sees television weight and a branded search curve, and is asked to infer a causal link. When the quarter misses, nobody can prove the exposure helped, and nobody can prove it didn't. That is not a measurement problem. It is a budget argument waiting to happen, and it usually gets settled by whoever has the better slide deck.

Three things that would cost almost nothing

A dealer-specific landing URL in every flight. One path per market, per model, with tracking parameters attached. If the campaign can buy the impression, it can name the destination. Right now the handoff is a logo and a hope.

Shared first-party capture. The OEM already pays for the traffic. Let the store keep the identifier when the shopper raises their hand on an OEM-hosted page, within consent rules and franchise law. A model page that collects nothing is a dead end by design.

A delivery survey. One question in the CRM at delivery: what made you choose this store? One identity-resolution vendor puts the share of dealership site visitors who never submit a form at roughly 90%. That is a vendor's number, not an audited one, though it is directionally consistent with what any dealer's own analytics show about form fills against sessions. The survey is often the only first-party signal left. It is cheap, imperfect and vastly better than the annual argument about brand health.

The data exists, it just doesn't flow down

OEMs know which markets got weight, when, and at what frequency. They have brand-tracking studies, matched-market tests and media-mix models. Some of that is genuinely proprietary. Very little of it is so sensitive that a dealer cannot see a market-level read.

What dealers typically get is the conclusion, not the file. Which means every conversation about next year's allocation starts from the OEM's framing, and the dealer's only counter-evidence is their own sold units, which the OEM already has too.

What to put on the table

The ask is not transparency as a virtue. It is four specific artifacts:

1. Market-level branded search and consideration lift, by flight, by metro.
2. A dealer URL and tracking parameter convention for every national campaign.
3. A shared definition of what counts as an attributed sale, written down before the campaign runs.
4. Retention of the delivery survey question, reported back in aggregate with the flight data.

None of that requires the OEM to hand over its media strategy. All of it lets the dealer reconcile exposure with revenue, which is the only test that survives a bad quarter.

Why this gets sharper in 2027

Assistants and AI Overviews quote what is verifiable: hours, prices, inventory, reviews, structured product data. Impressions are not citable, and neither is a sponsorship. The stores that will be named in an answer are the ones holding first-party data and clean structure, and the OEMs that will be credited are the ones whose campaigns actually handed that data to a local point of sale instead of a billboard.

The short version

Demanding attribution from the dealer tier while supplying none to it is not a measurement standard. It is a procurement habit. The fix is not a better deck. It is a landing page, a capture field, a survey question and a data file, all four of which cost less than the media they are supposed to justify.

Our own report page is built to be read that way: every rate has a stated denominator, and the campaign ladder is ordered by booked appointments rather than impressions.