How Many Photos Does a Marketplace Car Listing Need?
Facebook Marketplace caps a vehicle listing at 20 photos. Across 29,969 vehicles imported through autobook.io, the median unit carried exactly 20, and 55.9% arrived at the cap. Post four photos and you are not competing with a private seller who shot six — you are competing with a dealer listing that filled all twenty.
Nobody has published a number for this. Search it and you get generic Marketplace advice written for people selling furniture, plus photography listicles telling you to use good lighting. Neither answers the only question a rep actually has at 4pm with eleven units to get up: is four enough, and if not, what is?
We can answer it from our own data, so here is the data first and the argument second.
What does the photo count actually look like?
Every vehicle imported through autobook.io in a trailing 90-day window, all accounts, n = 29,969.
| Measure | Result |
|---|---|
| Units with at least one photo | 94.6% |
| Median photos per unit | 20 |
| Units arriving at Facebook's 20-photo cap | 55.9% |
| Sample | 29,969 vehicles |
Read the median line carefully, because it is the whole article. The median is not "about twenty." Twenty is the ceiling. More than half of all units land exactly on it, which drags the median onto the maximum value the scale allows, and the distribution does not continue past it. There is no long tail of thirty-photo listings to aim at.
The other number worth sitting with is 94.6%. That means 5.4% of imported units land with no photo at all — roughly one in nineteen. We publish that miss because a listing with zero photos is a listing that will not sell, and knowing which of your units are in that 5.4% is worth more than a rounder-looking statistic.
Where the missing photos actually go
They are almost never missing from the dealership. They are missing from the part of the website a crawler can reach.
The usual causes, in rough order of how often they bite: a gallery that only loads when a human clicks a thumbnail, so the page's own HTML contains one image and a script; a photo host that blocks datacentre traffic while serving the same file happily to a browser; an image format the destination won't take; and a unit that genuinely went live on the site before the photographer got to it.
That last one is worth checking before blaming software. On most lots a meaningful share of "no photos on Marketplace" traces back to no photos on the website either, usually on the newest arrivals — the ones a rep most wants up.
So how many photos should a listing have?
Twenty, or as close as the unit gives you.
That is a less interesting answer than a threshold would be, and we would rather give you the threshold. So we went looking for one, in our own sold data, and did not find it.
Splitting 802 sold units by how many photos the listing carried when it went up:
| Photos on the listing | Units | Median days to sell | Sold within 30 days |
|---|---|---|---|
| 1–5 | 61 | 19.9 | 67.2% |
| 6–10 | 22 | 19.0 | 63.6% |
| 11–19 | 374 | 18.9 | 64.4% |
| 20 (the cap) | 345 | 15.9 | 65.8% |
The 30-day column is flat. Every bucket lands between 63.6% and 67.2%, including the one with five photos or fewer. The median moves about three days in favour of listings at the cap, and the 6–10 bucket has only 22 units in it, which is not enough to lean on.
We looked for "more photos sell cars faster" and did not find it. That is the honest result and we would rather print it than a number we like better. Anyone publishing "listings with 15+ photos sell 40% faster" is either sitting on a controlled study they have not shown you or making it up, and in this niche it is almost always the second one.
Which leaves the real argument, and it is not about speed.
What we can say without stretching:
- The cap is 20. Filling it costs nothing.
- Over half the dealer inventory on this platform is already filling it, so a four-photo listing is below the field, not average.
- Every photo you leave off is a question a buyer has to ask in a message, and messages are the expensive part of this channel.
That last point is the practical one, and it is the argument that actually changes behaviour. Photos on a vehicle listing are not decoration. They are pre-answered questions. The buyer who can see the rear bumper does not message you about the rear bumper, and on a channel where you are already spending about forty minutes a day answering people, every question you retire in advance is time back.
Run it the other way and it gets concrete. Five photos on a unit means the buyer cannot see the back seat, the cargo area, the tyres, the engine bay or the odometer. That is five conversations, times however many buyers, all to deliver information a phone could have delivered in ninety seconds.
Which twenty, though?
Ten shots do most of the work, and the other ten are the specifics of the unit in front of you. The full shot list, in the order to take them, is in the 10-photo Marketplace shot list — front three-quarter, full side, rear three-quarter, driver's seat, dash with the odometer readable, center stack, back seat or cargo area, engine bay, a wheel showing tread, and the flaw you are disclosing.
Past those ten, spend the remaining frames on whatever this particular unit will get asked about:
| Unit type | What the extra frames should show |
|---|---|
| High-mileage anything | Tread depth on all four, service records, belts and hoses |
| Truck | Bed floor, hitch, undercarriage, tailgate condition |
| Anything with leather | Driver's bolster, rear seat backs, steering wheel wear |
| Older unit | Every panel gap and every flaw, close |
| Third row | Third row up, third row down, cargo behind it |
| Anything modified | The modification, clearly, plus whatever it replaced if you have it |
The instinct on an older unit is to shoot around the damage. Do the opposite. A visible scuff you photographed is a scuff the buyer priced in before they messaged. A scuff they find in the parking lot is a renegotiation, and often a walk. It also does something to the rest of the listing: a seller who photographed the bad panel is implicitly telling you the other nineteen photos are honest too.
Order matters after the first one as well
Buyers swipe in sequence and most stop early. The order that survives that is: exterior walk-around first, then the driver's seat and dash, then the rest of the interior, then mechanical, then flaws last.
Putting flaws last is not hiding them — they are in the same twenty, and the description names them. It is that a buyer who has not yet decided they like the unit has no context for a close-up of a scratched bumper, and a scratch is a much smaller thing after nine photos of a clean car than it is as photo two.
Does the first photo matter more than the rest?
Considerably. Marketplace is a grid of tiles, each one a cropped photo with a price over it, and the first frame is the entire advertisement at that moment. Nothing you typed is visible. The other nineteen photos only matter to people who already tapped, which makes photo one a different job from photos two through twenty.
Shoot it front three-quarter, with the vehicle pulled out of the inventory row so there isn't another car's bumper in your hero shot. Two things kill more hero shots than bad light: another vehicle in frame, and a crop that cuts the wheels. Marketplace crops to a tile, so anything important near an edge is at risk.
What about AI-edited photos?
They are allowed, and Facebook may label them. Meta applies an "AI Info" label to content its systems detect as AI-generated or AI-edited, which is a disclosure, not a penalty — it does not remove a listing or restrict an account on its own. We went through what triggers it and what it means for a vehicle listing in the AI Info label, explained.
The rule that keeps you out of trouble is simpler than the policy: the photo has to be of the actual unit, in its actual condition. Cleaning up a background is cosmetic. Editing out a dent is a misrepresentation, and the person who finds it will be standing next to the car. Listing accuracy is your responsibility under Facebook's Commerce Policies, and it is also just the difference between a delivery and an argument.
Where autobook.io fits, and where it doesn't
Import pulls whatever the source website published, up to the cap, which is why the median comes out at 20 — the photos are already on your dealership site and the work is moving them, not shooting them. Where the source is thin, the import is thin, and no software fixes a website with four photos per unit.
AI photo work is a separate, premium, opt-in thing at 2 credits per image. It replaces backgrounds with studio scenes, removes licence plates and plate frames, and adds reflections and shadows while preserving the actual vehicle including aftermarket modifications. It also classifies interior versus exterior shots and leaves interiors alone, because buyers are meticulous about the space they will be sitting in. It is never automatic and it does not run on every photo — you pick the images, and each one you pick costs credits.
What it will not do is give you twenty photos when the source had three. There is no invention step, and there shouldn't be. That part is a phone and six minutes.
Frequently asked questions
What is the photo limit on a Facebook Marketplace vehicle listing?
Twenty. It is a hard cap — you cannot add a twenty-first, and there is no setting that raises it.
Is four or five photos enough to sell a car on Marketplace?
It will not stop the listing from going live, but it puts you below the field. More than half of the dealer units we import arrive at the full twenty, so a five-photo listing is competing against listings with four times the information.
Do more photos make a car sell faster?
We won't claim that, because our data shows what listings carry rather than what caused a sale. What is defensible is that unanswered questions turn into messages, and every photo removes one.
What should the first photo be?
A front three-quarter with the vehicle pulled clear of the inventory row. It is the only image most buyers ever see.
Why did my listing import with no photos?
Usually the source gallery only loads for a human clicking through it, or the photo host blocks non-browser traffic. Sometimes the unit genuinely has no photos on the website yet. Check the website first.
How long does a car take to sell once it's up?
Across 802 units sold through autobook.io, the median was 18 days from the most recent post, with 65.2% inside 30 days. The full distribution and its caveats are in how long it takes to sell a car on Facebook Marketplace.