Which campaigns produced units that actually delivered?

Automotive retail is measured in leads because leads are countable the same day. A third-party listing site, a paid search campaign and a social campaign all report lead volume, and the cost per lead comparison follows automatically.

The problem is the numerator. A lead that buys a used compact at modest gross and a lead that buys a new truck with a full F&I package are one lead each and multiples apart in what the store earns. Cost per lead cannot see that and never will.

Ranking sources on delivered gross changes the ordering in most stores, and it is the ordering the general manager actually needs, because the budget is defended on gross rather than on traffic.

Listing marketplaces are a large line item in most dealer budgets and they report their own performance: views, leads, sometimes influenced sales on their own definition. There is no neutral ground on which to compare them with the store's own paid search.

Uploading the marketplace's lead export as a source file creates that ground. Both are ranked on delivered gross from the same deal export, which is the only comparison that settles the argument.

A buyer who returns for service is worth materially more than the deal alone, and fixed ops is where much of a store's profit actually lives. The advertisement that produced them is scored against the front-end gross only.

Because the match is on the customer, later service revenue exported as further rows credits the channel that produced them originally. Sources that bring customers who come back stop looking equivalent to sources that bring one-time price shoppers.

A group running several rooftops across brands and markets has no useful average. Demand, competition and inventory differ, and so does what works.

A store column splits the report so each rooftop's spend follows its own evidence, which is also how the accountability usually runs.

Every store knows response time matters and most measure it against lead-to-appointment. That is the right instinct and the wrong denominator, because an appointment is not a delivered unit and the relationship between speed and gross is not the same across sources.

Exporting the lead timestamp alongside delivered deals lets the report show gross against time to first contact, per source. A marketplace lead that decays in fifteen minutes and a branded search lead that does not are managed differently once that is visible.

This is usually the finding that changes staffing in the BDC rather than spend in the ad account, and it is not available from any single system the store already runs.