You fund the advertising. They close the sale.

A manufacturer funds demand generation; an independent dealer captures the revenue. The two live in different systems owned by different companies, and no amount of tag management crosses that boundary.

So the national marketing budget gets justified with lead counts and dealer anecdote, and every budget review turns into an argument about whether the leads were any good. Nobody can settle it, because nobody has the sales data and the campaign data in the same place.

The only thing a dealer has to do is send a periodic sales export — a contact detail, an amount, a date. No software in their environment, no access to their CRM, no IT project, which is what makes this achievable across a network that will not agree on anything.

Head office combines those exports into one sales file per period, adds a dealer column mapped as Location, and sees the network in one report: revenue by dealer, with how much of each traces to a channel. Seats in that workspace see every dealer, so they stay with head office, which passes each dealer its own figures. Where dealers must each log in and see only their own book, each can be its own workspace on the Agency plan — and then there is no combined network view.

A national lead has to go somewhere, and the rule that sends it — nearest ZIP code, round robin, territory map, whoever responded last time — quietly decides what the campaign's results will look like. Send a lead to a dealer who does not follow it up and the campaign produced nothing. Send the same lead to one who calls within the hour and it produced a sale. The media performed identically in both cases and the report will not say so, because routing is a marketing variable disguised as an operations decision.

Running every dealer's closed sales against the same central campaign data separates part of the two effects. With a dealer column mapped as Location, each dealer shows its revenue and how much of it traces to the central campaigns. Where most dealers show a healthy attributed share and three show almost none, the campaigns are reaching customers and those three dealers are a conversation for somebody. Where the attributed share is thin everywhere, the media is the problem and head office owns it. That distinction is currently unavailable, which is why co-op disputes go round in circles: both parties have a plausible story and neither has a per-territory revenue figure to test it against.

It also makes a routing change measurable, which matters because these changes are usually made on conviction and never evaluated afterwards. Switching from a ZIP code rule to a response-time rule, or reassigning an underperforming territory, can be read on closed revenue before and after across the same campaigns. There is no window and nothing had to be instrumented at the time, so the earlier period can be constructed retrospectively from data the dealers already hold. That removes the usual objection, which is that nobody measured the old arrangement while it ran.

A dealer network's revenue from a customer does not end at the first sale, and in many categories the first sale is the least profitable part of it. Parts, consumables, servicing, warranty work, accessories and the replacement purchase four years later all flow through the same dealer to the same customer. Head office funds the campaign that produced that customer and is credited with one transaction's worth of value. Every argument about whether central marketing pays for itself has been conducted on that understated figure.

Where dealers can export the whole customer relationship rather than the initial sale, the picture changes substantially. Each subsequent transaction credits the channel that produced the customer, and a campaign's return is measured across the life of the relationship it opened rather than across the month it ran. That is a larger ask of a dealer than a sales export, and it is worth making of the dealers who dispute the budget most vigorously, because it is the analysis most likely to change their minds. Starting with initial sales and adding aftersales later is a reasonable sequence.

A transaction-type column mapped as Product / service keeps the two readings visible, because first sales and aftersales then show as separate lines, each with how much of it traces to a channel. Head office usually wants to know which campaigns produce customers; a dealer usually wants to know which produce customers who come back. The channel ranking includes the aftersales, so a channel producing many first sales to customers who never return is visibly a worse investment than its headline suggests, and one producing fewer customers with long relationships is visibly a better one. Neither fact is visible without the aftersales rows.

The first finding is usually about participation rather than performance. Some dealers send a clean export within a week, some send a spreadsheet with four columns and no header, and some do not respond at all, and the pattern can say more about how they feel about the central marketing program than about their size. That is itself information, and it is better held explicitly than inferred from the tone of the next regional meeting. The dealers who participate first are often the ones who already believe the fund works.

The second is variation. A national campaign rarely performs uniformly across a network, and the spread between territories can be wide enough to make the network average useless as a planning figure. Head office has historically had no way to see this, because lead counts per territory reflect population rather than performance and the revenue stayed inside dealer systems. With the dealer column in place, the attributed share by dealer shows the spread on one page, and turns a single budget decision into a set of territory-level ones. Where a few territories stand out, what they do differently — response time, follow-up, opening hours — may be something the rest could copy.

The third is the size of the untraced share. Dealer networks tend to have large Direct / Unknown buckets, because walk-in trade, existing relationships and each dealer's own local advertising all produce revenue central campaigns did not. That number is not a failure of the exercise and should be presented as a finding: it is the share of network revenue head office marketing was never going to influence. Knowing its size is what makes the attributed portion defensible in a room full of dealers who suspect the fund of claiming credit for their work.