Most dealer groups don’t have a marketing problem. They have an ownership problem. Five vendors, each optimizing a different number — the agency optimizes cost per lead, the website vendor optimizes sessions, the SEO shop optimizes rankings, the CRM (customer relationship management) rep optimizes logins, and the OEM (original equipment manufacturer) portal optimizes compliance. Every one of them can show you a dashboard where their number is up. And you still can’t answer the only question that matters: which of this spend sold a car?
That gap isn’t a channel problem. It’s not the agency’s fault, and firing the website vendor won’t fix it. The problem is a missing seat — nobody at the group owns the chain from budget to sold VIN (vehicle identification number), end to end, with decision ownership over every link. This post is the map of that chain. Five links, each one a complete answer on its own, each one covered in more depth elsewhere on this site. Read this page and you’ll understand how the whole dealership marketing system fits together. Then go deep where your group is weakest.
How should a dealer group set its marketing budget?
Work backward from unit goals. Not forward from a percentage of revenue.
The industry habit is to peg dealer group marketing spend to a percentage of gross or to whatever NADA (National Automobile Dealers Association) peers report spending. That number tells you what other stores spend. It tells you nothing about what your stores need. Two rooftops with identical revenue can need wildly different budgets — one sits on a highway with forty years of name recognition, the other opened eighteen months ago in a market where three competitors outspend it.
First-principles thinking runs the other direction. Start with the units each store needs to move this year, by department — new, used, service, F&I (finance and insurance) attach. Then ask: how many sales conversations does that require, how many leads produce a conversation, and what does it cost to produce a lead in this market at this moment? Now you have a budget that’s an output of your goals instead of a ritual you repeat every January. The full method, with the arithmetic, is here.
This changes the conversation with your vendors, too. A percentage-of-revenue budget is a pot to be divided. A unit-goal budget is a contract: this money exists to produce these units, and every dollar has to explain its contribution. Vendors hate that framing. That’s how you know it’s the right one.
In short: set the dealership marketing budget by working backward from unit goals per store — units needed, conversations per unit, leads per conversation, cost per lead — not by copying an industry percentage of revenue.
How do you allocate budget across marketing channels?
Give every channel exactly one job and one leading indicator. A channel without a defined job is a donation.
Paid search exists to capture in-market demand — its leading indicator is qualified lead volume at a target cost. Third-party listings exist to put inventory in front of comparison shoppers — their indicator is VDP (vehicle detail page) engagement that converts to contacts. Social and video exist to build name recognition before the shopper is in market — their indicator moves slowly, and that’s fine, as long as everyone agrees up front what it is. Email and CRM campaigns exist to reactivate the database you already paid to build.
The failure mode is letting every channel claim every outcome. When the agency’s deck shows paid social “influencing” sales that paid search also claims and the website vendor also claims, you’re triple-counting the same customer. One job, one number, per channel. When a channel misses its number for two consecutive quarters, the money moves. No appeals to “brand halo.” How to assign those jobs channel by channel is its own post.
Notice what this requires: someone above the vendors who composes the whole palette — who decides which job each channel gets and arbitrates when two vendors claim the same sale. That’s the missing seat again. No individual vendor can hold this role, because every vendor is a defendant in the case.
In short: allocate dealership marketing budget by assigning each channel one defined job and one leading indicator, then move money away from channels that miss their number — with a single owner arbitrating overlapping claims.
How do you measure marketing from click to sold VIN?
Accept that you’re reconciling four data layers that were never designed to agree, and build the bridge anyway.
Layer one is the ad platforms — Google, Meta, third-party listings — which report clicks and their own version of conversions. Layer two is the website, which reports sessions, form fills, and calls. Layer three is the CRM, which reports leads, appointments, and shows. Layer four is the DMS (dealer management system), which is the only layer that knows the truth: the sold VIN, the gross, the date.
These four layers will never line up cleanly. The ad platform claims a conversion the website never saw. The CRM has a lead the website can’t explain. The DMS shows a sale the CRM lost track of at the appointment stage. Vendors exploit these gaps — each one reports from the layer where their number looks best. When your analytics and your CRM disagree, this is why.
The fix is not a magic attribution tool. The fix is a monthly discipline: match sold VINs from the DMS back to CRM records, match CRM leads back to website sources, and treat the ad platforms’ self-reported conversions as advertising for the advertising. The metric that ends arguments is cost per sold unit by source, computed from the DMS backward. It won’t be perfect. It will be honest, and honest beats precise. The reconciliation, step by step, is here.
In short: measure dealership marketing by reconciling four layers — ad platforms, website, CRM, and DMS — monthly, working backward from sold VINs in the DMS to compute cost per sold unit by source, and treating platform-reported conversions as unverified.
What matters more than generating additional leads?
Handling the leads you already have. Speed and process beat more media, and it isn’t close.
At most groups the marketing budget goes up every year while the lead-handling process hasn’t been audited since the CRM was installed. Mystery-shop your own stores this week. Submit a lead on a used truck at 7 p.m. on a Saturday and time the response. Read the first email your CRM sends — is it a templated wall of text with the wrong salesperson’s name, or does it answer the question the shopper actually asked? Call the store and count the transfers.
The math is unforgiving. If lead handling leaks, every marketing dollar upstream leaks at the same rate. A group that spends more on media while its response process bleeds is buying a bigger bucket with the same hole. Fixing response time, follow-up cadence, and appointment-setting discipline is the least expensive gross on the table, because the leads are already paid for. What happens to a lead after it arrives, and where it leaks, is here.
This is also where the missing seat is most visible. The agency doesn’t own lead handling — that’s “the store’s problem.” The desk manager doesn’t own marketing — that’s “corporate’s problem.” The lead falls into the gap between them, and nobody’s comp plan notices. Someone has to own the whole chain, or the chain owns you.
In short: before adding media budget, fix lead handling — response speed, follow-up cadence, and appointment discipline — because a leaking process discounts every marketing dollar spent upstream, and no vendor owns this handoff by default.
How does a dealership get recommended by AI assistants?
By becoming the answer, not just the ad. A growing share of shoppers now ask an AI assistant — ChatGPT, Gemini, the AI summary at the top of a search page — “who’s the best place to buy a used Tacoma near me?” instead of scrolling ten blue links. When that happens, the assistant names a handful of dealers. Either you’re named or you’re invisible. There’s no position four.
AI assistants assemble answers from what’s publicly verifiable about you: consistent business information everywhere it appears, reviews with real substance and real responses, inventory and pricing data that’s structured so machines can read it, and pages on your site that answer buyer questions in plain, complete sentences rather than burying them in a slider. The sections of this post are written as standalone answers for exactly this reason — extractable, quotable, complete. What to fix first, and in what order, is here.
The uncomfortable part for dealer group marketing: none of your current vendors owns this either. The SEO vendor optimizes for rankings that matter less each year. The agency buys ads the assistant never shows. Getting named is a property of the whole system — your data, your reviews, your content, your reputation — which means it belongs to whoever owns the whole chain. Same missing seat. Fifth link, same argument.
In short: to get named by AI assistants, make your dealership machine-legible: consistent public business data, substantive reviews with responses, structured inventory information, and pages that answer buyer questions completely — a whole-system property no single channel vendor controls.
Who should own the dealership marketing chain end to end?
One person. With decision ownership, a mandate that crosses vendor boundaries, and a seat that reports on sold units, not channel metrics.
At most groups doing $10M and up, that seat is empty. The vendors each own a link, the general managers own their stores, and the whole chain — budget from unit goals, channels with defined jobs, measurement to sold VIN, lead handling, AI visibility — belongs to nobody. Every section above ends at the same conclusion because it’s the same problem five times.
I’ll be direct about my own background, because I expect you to check: I haven’t worked inside an auto dealership. I spent four years at La Mesa RV, an 11-location RV dealer group, and I’ve spent a career across creative, development, and management — enough range to look at a marketing operation holistically and pressure-test what every vendor tells you. The techniques in this post are industry agnostic; the chain is the chain whether the unit is a Tacoma or a travel trailer. What I bring is the seat itself: someone whose only number is the whole system working, sitting alongside the business instead of inside any vendor’s profit and loss.
If the argument on this page matches what you see in your own stores — dashboards up, certainty down — the next conversation is about your chain, your stores, your numbers. I want to understand where your specific links leak before anyone touches a budget line. That conversation costs you an hour and usually pays for itself in the first vendor meeting after it.