The average car salesperson sells 10 vehicles a month. That number has not moved in 75 years. Not when computers arrived. Not when CRMs promised to organize every follow-up. Not when digital retailing promised to start the deal online. Three generations of technology, billions in dealer spend, and the core productivity number of the entire retail model sits exactly where it sat when the salesman kept his book in a shirt pocket.
Any other industry would treat that as a five-alarm finding. Automotive treats it as wallpaper.
The number that ate every tool
Walk through what 10-a-month survived.
The mainframe era computerized accounting and inventory. Ten a month. Desktop computing put a screen on every desk. Ten. The CRM wave promised no lead left behind, with task lists, templates, and manager dashboards. Ten. The internet delivered pre-informed shoppers and the appointment culture. Ten. Digital retailing moved half the transaction online and promised the showroom would just be delivery. Ten.
Each wave produced real change at the edges: better record keeping, faster paperwork, more measurable marketing. None of it changed throughput per salesperson, which means none of it changed the underlying model. It just gave the existing model better office supplies.
Why the tools bounced off
Technology layered on top of an unchanged model does not change outcomes. That's the whole answer, but it's worth seeing the mechanism.
The traditional sales model is a serial, single-threaded craft. One salesperson owns one customer through the entire road-to-sale: meet, qualify, land on a car, walk, drive, pencil, close, deliver, follow up. Every step lives in one person's hands and head. The constraint was never information or record keeping. The constraint is that a single human can only run so many serial relationships at once, and most of each relationship is administrative, not persuasive.
Hand that model a CRM and you've digitized the shirt-pocket book. The salesperson still personally executes every step, plus now the data entry. Some stores measured productivity going down after CRM adoption, which surprised nobody on the floor.
The model was never redesigned around the tool. The tool was bolted onto the model. Bolt-on is the industry's default relationship with technology, and 75 years of flat throughput is the bolt-on model's report card.
The first technology that touches the constraint
What makes the current wave different is not that it's smarter software. It's that agents attack the actual constraint: the serial, administrative load on each human.
A digital employee can run the follow-up sequences, the appointment confirmations, the lead qualification, the status updates, the post-sale touches, for every customer simultaneously, around the clock. The salesperson stops being a one-person assembly line and becomes the human at the moments humans win: the greeting, the walk, the negotiation, the trust, the close.
That is a different model, not a better tool. And the distinction is exactly why most stores will fumble it. Buy agents and bolt them onto the unchanged road-to-sale, and they'll join the CRM in the graveyard of things that didn't move the number. Redesign the workflow first, then deploy agents into the documented steps, and throughput per human finally has room to move.
The sequence is unforgiving: documented workflows before automation, clean data before intelligence. Skip it and you've bought wave six of better office supplies.
The question for your store
Pull your own number. Units divided by sales headcount, trailing 12 months. Most stores land at 10, give or take, exactly where the industry has lived since the Truman administration.
Then ask the harder question: if you deployed agents tomorrow, would they be running a redesigned process, or would they be bolted onto the same model that already absorbed five technology waves without flinching?
The number has been waiting 75 years for someone to change the model instead of the tools. The stores that finally do will not look marginally better on this metric. They will look like a different business.
FAQ
How many cars does the average salesperson sell per month? Ten vehicles a month, a figure essentially unchanged for 75 years.
Why didn't CRMs improve salesperson productivity? CRMs digitized record keeping but left the serial, single-threaded sales model intact. The salesperson still personally executed every step, plus the data entry.
What actually constrains salesperson productivity? One human running every step of every customer relationship serially. Most of that load is administrative, not persuasive.
Can AI improve cars sold per salesperson? Only if the sales model is redesigned around it. Agents handling follow-up, qualification, and administration in parallel frees humans for the moments that close deals. Bolted onto the unchanged model, AI joins the list of tools that didn't move the number.
What should a dealership do before deploying sales AI? Document the sales workflows end-to-end and clean the data layer. Automation multiplies defined processes, not habits.
Todd Smith is the Founder and CEO of QoreAI and the author of The Intelligent Dealership: How AI and Data Transform Automotive Retail.
Frequently asked questions
How many cars does the average salesperson sell per month?
Ten vehicles a month, a figure essentially unchanged for 75 years.
Why didn't CRMs improve salesperson productivity?
CRMs digitized record keeping but left the serial, single-threaded sales model intact. The salesperson still personally executed every step, plus the data entry.
What actually constrains salesperson productivity?
One human running every step of every customer relationship serially. Most of that load is administrative, not persuasive.
Can AI improve cars sold per salesperson?
Only if the sales model is redesigned around it. Agents handling follow-up, qualification, and administration in parallel frees humans for the moments that close deals. Bolted onto the unchanged model, AI joins the list of tools that didn't move the number.
What should a dealership do before deploying sales AI?
Document the sales workflows end-to-end and clean the data layer. Automation multiplies defined processes, not habits.