How many sales do Gumroad products actually get?
The same 226 listings, read twice
On 2026-08-08 this project recorded the lifetime unit counter on every product page in its crawl that published one. On 2026-08-26 — 18 days later — it read the same pages again. 229 of them were still live; 226 were still publishing a counter at both readings, and those are the only rows compared below. Nothing here is an estimate: it is one number minus another number, and both CSVs are in the repository if you would rather do the subtraction yourself.
Which categories this covers. The listings behind these 226 product pages come from 15 of Gumroad's 15 top-level categories (3D, Audio, Business & Money, Comics & Graphic Novels, Design, Drawing & Painting, Education, Fiction Books, Films, Fitness & Health, Gaming, Music & Sound Design, Other, Photography, Recorded Music), so they are not an even draw across the platform. They are also unevenly weighted: 57% of them sit under 3D alone, against the 7% an even draw would give it. Weight them accordingly.
What actually moved in 18 days
| Units sold in the window | Listings | Share |
|---|---|---|
| Nothing at all | 60 | 26.5% |
| 1–4 units | 48 | 21.2% |
| 5–19 | 54 | 23.9% |
| 20–99 | 39 | 17.3% |
| 100 or more | 15 | 6.6% |
So 73.5% of these listings sold something in 18 days, which is the cheerful way to say it. The useful way is the first row: 26.5% sold nothing at all, and another 21.2% sold four units or fewer. Roughly a quarter of the set — 54 listings — moved twenty units or more.
The median listing that moved at all moved 10 units. Against its own lifetime total that is a change of 1.9%, so the median mover is a product that is ticking over rather than taking off. The largest single move in the window was 2,797 units.
Three things in this data that argue against it
10 counters went down. A lifetime unit total should not fall. Refunds and sellers resetting a counter both produce this and they cannot be told apart from outside, so 4.4% of the set is doing something the measurement cannot explain. It is reported rather than filtered out.
3 sellers switched the counter off between the two readings. That is the one change no single snapshot of this marketplace can show you, and it is the reason this page exists.
Only listings that publish a counter are in here at all. Publishing one is voluntary, and a seller who turns it on is likelier to be happy with the number. Read every figure above as the optimistic end of the marketplace, not the middle of it.
What did not move
Prices barely did: of the 177 listings priced in US dollars at both readings, 2 changed price in 18 days. (49 listings quoted in another currency are excluded — the earlier reading carries no currency column, and comparing across two would manufacture movement out of a unit mismatch.) Ratings moved on 68 of the 226, which is well under half the 166 that moved units — a reminder that ratings are a lagging and lossy proxy for sales, which is the subject of its own page here.
What this means if you are deciding whether to list something
The honest read is that an established listing on this platform is more likely to be quiet than busy, and that the difference between the quiet quarter and the busy quarter is not price — the correlation between a category's price and the share of it selling is close to zero. It is which category you are in and whether that category's demand is already held by its incumbents. The demand ranking is here and the case against bothering is here. Both are free.
The report is the version of this with the answer in it: every one of 255 categories classified by whether its demand is reachable or already held, and what changes about your approach in each case. You are paying for the interpretation, not for the rows. The rows are free, above and in the repository. If the data is all you wanted, take it and skip this.
Read the report — $249 Read three sections free first Or take all four CSVs free Selling here yourself? Read a worked audit of one storefront
The sample is the method sections lifted unedited out of the report — what was measured, the background rate, and the limits in full. Every section that says what to do is in the paid one. The free mirror is $0 with a $0 minimum and asks for an email at checkout. The same files without one: the data folder. The report carries a money-back guarantee, stated on the checkout page: full refund, no reason needed, and you keep the files. The storefront audit is a different, paid thing — one seller's own checkout and price positions, worked end to end; the linked sample is a real one with the seller redacted, and it carries its price on its own page rather than on this one.