// Dealer operations ·
Your cars aren't waiting on photos. They're waiting on Thursday.
A full photo set now finishes in three and a half minutes. Cars still take days to go live, because merchandising happens in one burst a week.
Daniel Melo · 7 min read
Over 53 calendar days this summer, 15 lots sent us 223 photo batches. The median batch went from upload to finished set in 3 minutes and 25 seconds.
Those same 223 batches landed on only 37 of the 53 days. Zero of them arrived on a Sunday. Fifty-five of them — a quarter of everything — arrived on a Thursday.
The processing bottleneck is gone. Nobody moved the calendar.
The constraint moved and the workflow didn't
For fifteen years the honest answer to "why isn't that car listed yet" was logistics. The photographer came Tuesdays. The editor took three days. The listing went up when the folder came back. Everything about how a lot schedules merchandising was built around a turnaround measured in days, and that was the correct design for the constraint that existed.
That constraint is now three and a half minutes.
What we see in the data is a workflow that hasn't noticed. Of 210 consecutive batches from the same lot, 96 arrived within half an hour of the one before it — someone sitting down, working through a stack, and clearing it. The median gap between two batches from the same lot is 56 minutes. Then it goes quiet, sometimes for days: eight gaps ran three days or longer.
That's not a throughput pattern. It's a batch-processing pattern, and it's a habit left over from when batching was the only sane way to work.
The week has a shape, and it isn't flat
| Day | Batches | Share |
|---|---|---|
| Monday | 31 | 13.9% |
| Tuesday | 39 | 17.5% |
| Wednesday | 26 | 11.7% |
| Thursday | 55 | 24.7% |
| Friday | 40 | 17.9% |
| Saturday | 32 | 14.3% |
| Sunday | 0 | 0% |
The hours tell the same story. Ninety-two percent of all batches land between 9 a.m. and 7 p.m. local, with a hard spike at 10 a.m. and a second one at 4 p.m. Merchandising is being done by whoever has a free hour in the middle of a selling day, which is exactly why it clusters on the day that hour exists.
Across individual lots the same shape holds: an average of 6.2 active days inside a 10.5-day window, and 17.6 photos on each of those days. Not two cars a day, every day. Six or eight cars in one sitting, then nothing.
What batching costs, in the only unit that matters
Take a unit that lands on your lot Friday afternoon. Under a Thursday cadence it goes live the following Thursday. That's six days where the car is on your floorplan, insured, depreciating, and invisible to the market.
Six days at a typical floorplan and carrying cost is real money on its own. But that's the smaller half. The larger half is that those six days come off the front of the listing's life — the window when the vehicle is newest to every shopper watching that price band, and when search and marketplace surfaces give a fresh listing its best treatment. You don't get those days back by leaving the car up longer at the end. The tail of a listing's life is not worth what the front is.
Run it across a 40-unit lot turning inventory monthly and you're paying for roughly a week of dead time on every car, permanently, because merchandising happens on Thursday. That's a line that never appears on any invoice, which is precisely why it survives — and it's one of the costs we argued belongs inside cost per car sold rather than in a separate report nobody opens.
Daily merchandising is smaller than it sounds
The objection is always staffing, and it's the wrong shape. Nobody is proposing a full-time photographer. The arithmetic:
- A lot taking in 40 units a month receives 1 to 2 cars on a typical day.
- A proper set — walkaround plus interior plus condition — runs about 12 minutes at the car.
- Processing finishes in under 4 minutes, unattended.
So daily merchandising for most independent lots is a 15-to-30-minute task, done by whoever took the car in, on the day they took it in. It is smaller than the Thursday session it replaces, because the Thursday session carries the overhead of setting up, remembering which cars are pending, and shooting eight vehicles while losing patience by the sixth. Our numbers show what that impatience produces: the sets that come in during the big batches are the short ones, and short sets are the VDPs that convert at 1%.
There's a second-order effect worth naming. When merchandising is daily, the question "is this car live?" has one answer — yes, since the day it arrived. When it's weekly, somebody has to track a pending list, and pending lists are where units get lost for a month. Two of the fifteen lots in this dataset signed up, took a median of under three minutes to run their first batch, and then never ran a second one. Not a tooling problem. A nobody-owns-this problem.
What to do Monday morning
- Timestamp your last ten takes. For each one, write the date the car hit the lot and the date the listing went live. The gap is your real merchandising lag, and it's usually two to three times what the team estimates out loud.
- Give the day one owner. Whoever books the car in shoots the car in, same day, before the keys go on the board. Not a role, not a hire — a rule that the unit isn't received until it's shot.
- Kill the pending list. If a car is on the lot and not live, that's an exception to chase today, not a row in a spreadsheet to work through Thursday. The point of measuring this is that lag is the one inventory number you control without spending anything.
Three and a half minutes of processing. Six days of waiting. Only one of those is a technology problem, and it's already solved.
// Straight answers
Same day, or next morning if it arrives after close. That's achievable now because the processing side is measured in minutes, not days. If your current lag is five or six days, don't aim at same-day immediately — get it to 48 hours first and hold it there for a month, because a cadence you can actually sustain beats a standard you abandon in week three.
It feels more efficient and measures worse. Batching saves a few minutes of setup and costs several days of market exposure per unit, and it reliably produces shorter photo sets, because the eighth car of a session gets less attention than the first. The efficiency gain is real but it's an order of magnitude smaller than what it costs.
More, not less. A low-volume lot has fewer units carrying the month, so a week of invisibility on any one of them is a bigger share of the total. It's also easier to fix: at two takes a week, daily merchandising is twenty minutes, twice a week, done by the person who already handled the car.
Shoot it anyway and reshoot after detail. A listed car with honest, unpolished photos outperforms an unlisted car every single time, and the interim set costs you nothing to replace. The alternative — waiting on detail, which waits on the detailer's queue — is how a five-day lag becomes a twelve-day one.
They're the same discipline applied at two points, and both are about not making the market wait. If you have to pick one to fix first, fix the one with the worse gap — for most lots that's [lead response](/blog/dealer-lead-response-time), where the delay is measured in hours against a window measured in minutes.