Here’s a number worth taping to the wall of the used car office: −$542.
That’s the gross gap, per unit, between a used car you bought from your own customer and one you bought at auction. Same lot, same market. The auction car retails for $542 less. And that’s before you count the buy fee and the transport bill it dragged along.
We found that number by looking at where retailed used cars actually come from. Across roughly 100 rooftops and 371,227 used retail deals over two full years, 57.3% of the cars sold were sourced from customers: trades, lease returns, cars the store already knew. The other 43% came from somewhere else. Mostly a lane with a bid badge and an auctioneer.
This article is about what that 43% really costs.
The scoreboard
Pull the two groups apart and the P&L splits with them.
The customer-sourced car averaged $2,237 in total gross (front plus back). The car with no customer history behind it averaged $1,695. That’s the $542 difference.
It gets worse at the exits. When a customer-sourced car doesn’t work out and gets wholesaled, the store loses about $50 on the deal. When an auction-lane car fails, the wholesale exit runs $290 to $494 in the red. Six to ten times the loss, on the cars the store hand-picked in the first place.
And the clock is different too. The customer car turns in 21 days. The auction purchase: 33 to 39 days. We’ll do the full working-capital math on that gap later in this series, but the short version is that 12 to 17 extra days of floorplan, insurance, and depreciation costs roughly $400 to $550 a unit. Silently. On every car.
The costs that never make the appraisal sheet
The hammer price is the number everyone argues about. It’s also the smallest part of the problem because every dealer in the lane is bidding on the same car at the same market value. The structural costs are what separate the lanes:
- Buy fee: $300 to $1,500, depending on the sale and the price tier.
- Transport: $200 to $1,000 or more, depending on distance.
- The waiting: those 12 to 17 extra days, at real holding cost.
- The miss: a 6–10x deeper loss when the car has to go back out the wholesale door.
Stack it up, and the auction car starts life roughly $2,500 behind a car sourced from your own customer base. Not because your buyer picked wrong. Because the lane itself charges admission.
One thing we expected to find and didn’t: a recon excuse. The data says the recon bill is nearly identical either way. That finding deserves its own article, and it gets one, because it changes the argument completely. If recon isn’t the difference, better bidding can’t fix this.
What this doesn’t mean
Nobody runs a used department with zero auction cars. Inventory holes are real. Aged-unit replacement happens. A hot segment your customers don’t drive is real.
The point is proportion. The auction should be the backfill, not the strategy. Right now, at many stores, 43% of retail inventory is paying admission fees to a lane while the cheaper, faster, richer source of cars drives past the service drive every morning.
Where the good cars actually are
Here’s the stat that sets up everything else in this series: of every six trades a store takes today, five arrive from customers who had no service relationship with that store. The drive isn’t feeding the used department. The database isn’t either. The most profitable acquisition channel for most dealers is at 1.56% utilization.
That’s the next article: the well under the service drive, how big it actually is, and the one customer type everyone targets who almost never trades.
This series is built on real DMS data: 250,677 trade-ins, 371,227 used retail sales, and 86,349 lease maturities across roughly 100 rooftops, 2024–2025. Every “keep” rate we publish is matched on a same-store basis only, which means the true numbers are higher than what we print.
