There is one number that tells you whether your marketing is working, and most dealers never calculate it. Not cost per lead. Not cost per click. Cost per car sold. It is the only figure that connects the money leaving your account to the vehicles leaving your lot. Everything else is a proxy, and proxies are where vendors hide.
What Cost Per Car Sold Actually Means
The math is simple. Take every dollar you spent on marketing in a period and divide it by the number of vehicles that spending helped sell. That includes your digital agency, your third-party listing sites, your SEM budget, your social spend, your direct mail, and anything else with a marketing invoice attached. The result is what it truly costs you to move one unit.
If you spent 40,000 dollars across all vendors in a month and sold 100 cars that can be tied back to those efforts, your cost per car sold is 400 dollars. That is a number a dealer principal can actually make decisions with. Think about how you already run the rest of the store. You know your holdback, your pack, your average front and back gross. Marketing is the one line where most stores still accept a vague answer, and that is the line where the leaks are biggest.
Why Leads and Clicks Mislead You
Vendors love clicks and leads because they are cheap to produce and easy to inflate. A 12 dollar lead sounds like a bargain until you learn that 200 of them produced three sales. Clicks are worse. They measure activity, not outcomes. A campaign can generate thousands of clicks and a beautiful dashboard while your actual delivery numbers stay flat.
Picture a store that runs two sources side by side. One sends 300 leads a month and the sales team quietly complains that half the phone numbers are dead. The other sends 90 leads a month and the desk fights over them because they close. On a lead report, the first source looks like the hero. On the delivery board, the second source is the one paying the bills. The gap between a lead and a sold car is enormous, and that gap is exactly where wasted spend hides. When you only measure the top of the funnel, you are grading vendors on effort instead of results.
Why Vendors Avoid This Number
Cost per car sold is uncomfortable for a vendor because it holds them to the same standard you hold yourself. A vendor can defend a cost per lead in isolation. They cannot defend it once you show that their leads close at half the rate of another source at the same price. The number strips away the flattering metrics and leaves only the outcome.
- It exposes sources that generate volume but never convert.
- It makes two vendors directly comparable, no matter how different their reports look.
- It shifts the conversation from activity to accountability.
Watch what happens the first time you ask a rep for cost per car sold instead of cost per lead. A good partner leans in and helps you trace it. A weak one changes the subject back to impressions, video views, or engagement. The reaction alone tells you which vendors are confident their work moves metal.
How to Start Even With Imperfect Data
You do not need perfect attribution to begin. Start with what you have. Pull your total marketing spend for last month from your invoices. Pull your total sales from your DMS. Divide one by the other. That single blended number is already more honest than any vendor dashboard, and it gives you a baseline.
From there, work toward attributing sales to sources. Ask every customer how they found you and log it. Match phone numbers and email addresses from your CRM back to campaigns. It will be messy at first. It gets sharper every month you keep at it, and even a rough version beats flying blind. One store might start with nothing but a whiteboard tally of blended cost per car sold, and within a quarter be splitting it out by vendor well enough to renegotiate two contracts. You do not need a data science team. You need the discipline to run the same simple division every month.
Make It the Number You Manage To
Once you know your cost per car sold, review it every month and break it down by vendor. Cut what runs expensive, feed what runs efficient, and make every vendor defend their line against this one figure. Treat it the way you treat gross. When a source drives units at a cost that makes sense, give it more room. When a source runs expensive month after month with excuses attached, stop funding it. That is how you stop paying for activity and start paying for results.
Frequently asked questions
What is a good cost per car sold?
There is no universal number, because it depends on your average gross, your market, and your mix of new and used. The right benchmark is your own trend and the spread between your vendors. A source delivering cars at half the cost of another is the story, not a figure from an industry average that does not know your store.
How is cost per car sold different from cost per acquisition?
Cost per acquisition often stops at a lead or a form fill, depending on how the vendor defines it. Cost per car sold goes all the way to a delivered, signed unit in your DMS. It is the stricter, more honest version, and it is the one that ties directly to the money in your account.
Do I need special software to calculate it?
No. You can start with invoices and a DMS export in a spreadsheet. Software helps you attribute sales to sources more precisely as you grow, but the first and most valuable version of this number is one you can build by hand this week.
These are the numbers you should own, not rent from a vendor. See how VCTRS helps dealers hold every vendor accountable on our vendor accountability page.