The numbers, by sub-genre
Our guide to how to market a Steam game tells you to judge target-market demand on three things before committing to a genre, and revenue by tag is the first of them. This page is that number, computed across 107 sub-genres, with 20 of them shown below.
"Games on Steam" is how many titles carry that tag in this dataset. That is the population each median describes, not the number measured to produce it.
Every median is read from a sample of roughly twenty titles taken at the tag's middle rank, which is how you estimate a midpoint without pulling an entire catalogue. Figures are rounded, because they are estimates.
| Sub-genre | Games on Steam | Median copies sold | Median gross revenue |
|---|---|---|---|
| Open World Survival Craft | 684 | 7,290 | $102,000 |
| 4X | 648 | 3,900 | $50,000 |
| JRPG | 4,359 | 1,490 | $21,000 |
| Colony Sim | 1,167 | 2,280 | $17,300 |
| RTS | 2,894 | 1,390 | $16,700 |
| Grand Strategy | 1,298 | 1,550 | $14,700 |
| Dating Sim | 5,541 | 1,780 | $14,400 |
| Visual Novel | 12,646 | 1,140 | $10,300 |
| RPG | 24,666 | 979 | $9,770 |
| Farming Sim | 1,494 | 1,080 | $8,690 |
| Souls-like | 2,044 | 1,040 | $8,460 |
| Adventure | 52,404 | 726 | $6,520 |
| Strategy | 27,804 | 578 | $4,040 |
| Idler | 4,358 | 574 | $2,870 |
| Horror | 15,142 | 867 | $2,590 |
| Action | 53,193 | 508 | $2,530 |
| Puzzle | 26,452 | 477 | $2,390 |
| Survival Horror | 5,430 | 683 | $2,050 |
| Match 3 | 1,673 | 298 | $1,830 |
| 2D Platformer | 11,366 | 209 | $1,040 |
Read the bottom of that table as carefully as the top. A 2D Platformer is one of the most common things an indie studio builds, and the median one sells an estimated 209 copies.
The median is the number that matters, and it is not the average
Take Colony Sim. The median game in that tag earns an estimated $17,300. The highest revenue estimate among that tag's best-selling titles is $255 million. That is roughly a 14,700 times gap inside one genre.
An average across Colony Sims is dominated by the handful of titles at the top, so it describes a level of performance almost no game in the tag reaches. The median describes the game in the middle, which is much closer to what a new entrant should plan for.
So whenever you see a revenue figure for a genre, ask which statistic it is. An average, or a figure drawn from a list of successful games, describes the top of a power law rather than the distribution you are entering.
Most genres do not clear $10,000 gross
Across the 107 sub-genres in this dataset, counting gross revenue before Steam's cut:
| Median gross revenue | Sub-genres |
|---|---|
| Under $5,000 | 56 of 107 |
| Under $10,000 | 81 of 107 |
| $10,000 or more | 26 of 107 |
The median across all of them is $4,837.
That number is gross: Steam's 30 percent, VAT, refunds and discounting all come out of it.
Our guide to how much Steam takes models net revenue at roughly 35 to 57 percent of list price, the low end for globally distributed games and the high end for US and EU concentrated ones.
The 35 to 57 percent band is a model rather than a measured rate, and the gross figure it applies to is an estimate, so read the result as an order of magnitude: a $4,837 gross estimate comes out at roughly $1,693 to $2,757 in the developer's hands.
How to use this without drawing the wrong conclusion
Three ways this table gets misread.
It is not a ceiling. The median describes the middle of a tag, not its best case: what the tag looks like by default, not what your game is capable of.
A low median is not automatically a bad genre. A tag that is cheap to enter fills up faster, and a bigger, more varied field pulls the middle down without saying anything about what a well-made entry can do.
A high median is not automatically a good bet. Open World Survival Craft has the highest median here at $102,000, and it is also one of the most expensive and technically demanding things a small team can attempt.
A high median in a tag full of large productions is telling you about the budgets, not the opportunity.
Compare the median for your tag against what your game actually needs to earn, then pick reference titles from the middle of the distribution rather than the top. How to pick comparable titles for your Steam game is the method.
Where these figures come from, and what they cannot tell you
They are estimates, not sales reports. Steam does not publish per-game sales. Every number on this page is derived from a licensed third-party sales-estimation dataset, pulled 2026-09-10 across 108 tags, of which 107 returned data. Treat them as well-modelled estimates with real error bars, not as measured revenue.
Tags are not genres. Steam has broad store genres and granular community tags; these figures key off the tags, and a game can carry several, so the same title appears in more than one row.
One tag is excluded, and here is why. Median copies is read across every title in a tag, free and paid, while median revenue is read across the paid titles alone, because a free game has no unit price to multiply by.
In most tags that gap is small enough to ignore. In Battle Royale it is not: its leaders are free, so the row divides to roughly $190 a copy, thirteen times any other tag in the dataset.
The $190 figure is real but it describes a handful of paid games rather than the tag, so it is left out rather than printed with a caveat.
Revenue is copies multiplied by the current list price. It takes no account of discounts, regional pricing, bundles, refunds or any price the game charged in the past, and most of those push the real figure below the estimate rather than above it.
This is a snapshot. Figures were pulled in September 2026 and the catalogue moves. Re-pull before making a decision that depends on the exact number rather than the shape.
