Every dealership knows its cost per acquisition. Most know its cost per lead. Almost nobody knows its cost per photographed vehicle, which is strange, because every car on the lot goes through it and nothing goes live without it.
The reason is that the cost is split across four budget lines that never get added together. A salary sits in payroll. A booth sits in capex. A vendor invoice sits in marketing. A software subscription sits in IT. Nobody owns the total, so nobody calculates it.
Here is the total.
The four ways to get a car shot
In-house photographer. You hire someone. They shoot, edit and upload.
Outsourced lot visits. A vendor comes to the lot on a schedule and shoots whatever is ready.
Photo booth. You build a dedicated space with fixed lighting and a turntable or a fixed camera rig.
Phone plus software. Whoever is free walks out with a phone, and software handles the editing, backgrounds and output.
The inputs
Everything below is published, and I have named the source for each so you can argue with it.
An in-house photographer runs $3,000 to $5,000 a month plus gear, according to Photoroom’s 2026 breakdown. Payscale puts vehicle photographer pay at $15.00 an hour on average, in a range of $10 to $21, updated September 2026. Outsourced lot visits run $40 to $150 per vehicle. A dedicated photo booth costs $30,000 to $150,000 to build, both figures again from Photoroom. Software sits at $120 to $499 a month depending on volume tier, per HelloRamp’s published pricing.
Two caveats before anyone quotes this at a vendor.
The Payscale figure rests on eleven salary profiles. That is a thin sample and I would not build a budget on it alone. It is useful as a sanity check on the monthly range, not as a precise number.
The Photoroom figures come from a vendor in the adjacent space, so treat them as directional. They are the only published ranges I could find, which is itself the point of this article.
The assumptions
Stated plainly so you can change them.
Photographer at the $4,000 midpoint, plus roughly $83 a month for gear amortised over three years. Booth at $60,000 amortised over five years, which is $1,000 a month, plus an operator at $3,500. Outsourced at the $95 midpoint of the published range. Software at published tiers, plus five minutes of a porter’s time per car at $15 an hour, which is $1.25 a vehicle.
That last line matters. Software does not shoot the car. Somebody still walks out with a phone, which is why the capture step in tools like AI-guided photography is aimed at untrained staff rather than photographers. Any model that leaves out that labour is selling you something.
At 50 cars a month
In-house lands at $81.66 a car. A booth is $90.00. Outsourcing is $95.00. Software is $6.83.
At this volume there is no argument to have. Outsourcing costs roughly fourteen times the software route. That is not a close call, it is a different category of decision. Fixed costs spread over fifty cars are brutal, and a booth that cost sixty thousand dollars is being asked to justify itself fifty times a month.
At 150 cars a month
In-house drops to $27.22. A booth drops to $30.00. Outsourcing stays at $95.00, because per-car pricing never scales. Software is $6.58.
This is where outsourcing stops making sense for most stores. The two fixed-cost options have each fallen by about two thirds while the variable one has not moved at all.
At 400 cars a month
In-house is $10.21. A booth is $11.25. Outsourcing is still $95.00. Software is $6.25.
And here is the thing worth noticing. In-house now costs about one and a half times the software route. You are comparing $10.21 against $6.25, and a four dollar difference per car sits comfortably inside the range where other factors should decide it.
So the honest version is this. At low volume the answer is obvious. At high volume it is a judgement call. Most of the content written about this pretends it is obvious at every volume, which is how you can tell who wrote it.
What this model does not capture
Three things, and all three can outweigh everything above.
Time to live. A car photographed on day one and a car photographed on day four cost the same to shoot and do not cost the same to own. Three extra days is three days of floorplan interest and three days of depreciation on a depreciating asset. Plug your own flooring rate in. Stores that have compressed this gap, like Diamond II Auto Sales, measure the improvement in arrival-to-live rather than in cost per photo, and that is the right frame. For most stores the carrying cost swamps the per-car photography cost entirely, which means speed matters more than price and the cheapest option is not automatically the right one.
Consistency. A booth produces the same output every time. A rotating cast of porters with phones does not, unless the software enforces it. If your group runs multiple rooftops, variance across stores is a real cost even though it never appears on an invoice.
Rework. Nobody models the cars that get shot twice. Bad weather, missed angles, a sold unit that comes back. Add ten to fifteen percent to any in-house or outsourced figure and you will be closer to reality.
Run it on your own numbers
Four lines and you are done.
Total monthly cost of your current method, including the labour nobody counts
Cars photographed last month, actual not planned
Divide
Multiply your average days-to-live by your daily flooring cost, and add that
Most dealers who do this find the per-car number is lower than they feared and the days-to-live number is worse than they thought. The money is rarely the problem. The delay usually is.
Which is the uncomfortable conclusion. If you are choosing a photography method on price alone, you are optimising the smaller of the two numbers.
Written for HelloRamp.ai, AI visual merchandising software for car dealerships.
Top comments (0)