Should you list a Gumroad product in more than one category?

The short answer

No — not as something you can act on. Sellers whose products turn up under more category searches do have far more demand: the median goes from 1 rating for the 3,446 sellers found under one category to 79 for the 74 found under 6 or more. That table is on this page and it is real. It is also, on the evidence below, mostly demand causing breadth rather than the other way round, and the single fact that settles it is that 17 of the 45 most in-demand sellers in this sample appear under exactly one category.

What “categories” means here, before any number

This is not how many categories a seller ticked a box for. It is how many of the walk's category searches returned their listings. Gumroad orders those results by demand, so a product that sells surfaces under more terms by construction. Every figure below is an association, and the direction is exactly what is in question — so the page is organised around testing it rather than assuming it.

Every seller, banded by breadth

The association is strong to the eye and modest in the data: across all 4,543 sellers the rank correlation between breadth and demand is 0.342. For comparison, the same correlation for catalogue size — the thing the previous guide measured — is 0.283. Breadth beats catalogue size. Both are weak, and 75.9% of sellers sit in the first row, so the last row is 74 storefronts and should be read as such.

Holding catalogue size roughly fixed

The obvious objection is that broad sellers are just sellers with more products. They are not: the pattern survives inside each catalogue-size group, including among sellers with exactly one product, where the number of products cannot vary at all.

Sellers with 1 product (3,262 sellers)

Sellers with 2–4 products (1,007 sellers)

Sellers with 5 or more products (274 sellers)

That is the strongest form of the case for breadth, and it is worth stating plainly because the rest of this page takes it apart. A single product appearing under four or more searches has a median of 35 ratings against 1 for a single product appearing under one. Same catalogue size, 35× the demand.

Why that is the ranking measuring demand, not breadth creating it

If breadth were something a seller supplies, you would expect to find sellers who supplied it and got nothing back — broad listings with no demand. They are almost absent. 88.0% of the 1,710 sellers with no ratings at all appear under exactly one category, against 68.5% of sellers with any demand, and the broadest unrated seller in the entire sample reaches 6 categories — in a sample whose maximum is 21. Breadth without demand barely exists here.

The correlation behaves the same way when the confound is squeezed. Among the 3,262 single-product sellers it falls to 0.169, roughly half the sample-wide 0.342. That is the direction you would expect if much of the sample-wide figure were breadth and catalogue size both tracking the same underlying thing.

The fact that settles it

The top 45 sellers — the 1% holding 52.5% of all 204,971 ratings in this sample — have a median breadth of 2 categories, and 17 of them (37.8%) appear under exactly one. Whatever produced the demand at the top of this marketplace, it was not spread. It is the same shape as the catalogue-size finding: the top of this market is one thing that works, not several things placed widely.

What this does license you to do

Two things, both diagnostic rather than strategic.

Ranking under exactly one search term is where the dead listings are. 43.7% of one-category sellers have no ratings at all, against 1.4% of the broadest. If your product surfaces under one term and nothing else, you are in the group where most of this platform's silence lives. That is a signal to re-examine the product, not a signal to add categories.

The lever adjacent to this one is real. Which searches return your listing is decided by its title and description text, and that is the part you control. This data cannot tell you it will work — nothing here isolates a seller who rewrote a title — but it is the only intervention in the neighbourhood that is not simply relabelling.

What it does not license: listing the same product under more categories to buy demand. Nothing here measures that, the correlation is compatible with it doing nothing, and 17 of the 45 biggest sellers did not do it.

What breadth is a lower bound on. The walk went three pages deep per category, so a listing that ranks deeply under a term was not recorded as appearing there. Breadth is therefore undercounted, and undercounted hardest for low-demand listings — which inflates the gap this page reports between narrow and broad sellers. The finding that breadth is not a lever is conservative: correcting the bias would weaken the association further, not strengthen it. Prices here are the seller's own median asking price, and demand is measured in ratings, which are a floor on buyers rather than a sales count — what one rating is worth in units is measured separately.

Which sample this is, and why it is not the other one

This page and the catalogue-size guide are the only two derived from the category walk — 8,322 products from 4,543 sellers, 204,971 ratings — because it is the only sample that records who sells what and under which category. Every other guide here is measured on a separate sample of 1,344 products drawn from 42 category searches. The two are never averaged; each is published as it was measured.

Check it yourself

Every figure above comes from the seller CSV (4,543 rows, one per seller, with a categories column) and the listing table it derives from, with the derivation alongside them. Both are in the DOI-archived record, CC BY 4.0, no signup.