What a comp is actually for
A comparable title is a released game you use as a reference point for what your game could do. Comps get used for three different jobs, and the right list is different for each:
| Job | What you need from the comp set | Common mistake |
|---|---|---|
| Estimating demand | Games at your scale, released recently | Using famous games, which estimates the ceiling |
| Pitching a publisher | Games the publisher already understands | Naming a hit, which invites a comparison the pitch cannot support |
| Choosing tags and positioning | Games competing for the same shelf | Picking by art style rather than audience |
The one thing all three need: the comp's buyers should overlap with the buyers you want. A game that looks like yours but sells to a different audience is not a comp, it is a mood board.
Where to start, and what this page assumes
If you have never built a comp list, start with how to market an indie game.
That guide's first step, genre-market fit, already covers the basics in its third checkpoint: finding your five to ten closest titles, starting from Steam's "More Like This" carousel, and reading total players for released games or follower counts for unreleased ones.
This page covers the two things that decide whether the list you end up with is any use: how to filter it, and which member of it to take your numbers from.
Two filters most comp lists are missing
A list built purely from "More Like This" will be full of games that resemble yours and compete on a different budget. Two filters fix that, and both take minutes.
Filter to your scale. Drop anything with a publisher-sized production behind it unless you have one.
Three visible proxies for budget: the number of supported languages, whether the trailer is cut like an ad, and whether the store page has bespoke artwork for every section rather than reused screenshots. A game localised into twelve languages had a budget line you probably do not.
Filter to recent releases. Prefer the last two years. Steam's catalogue size, its discovery surfaces and its price expectations all move, so a comp from five years ago is describing a market that no longer exists.
If your genre had a breakout hit that reset expectations, comps from before it are especially misleading.
After both filters you should still have five to ten games. If you have two, your filters were too tight or your tag is genuinely tiny, which is itself worth knowing.
What to record for each comp
Use the same fields for every game so the set is comparable. All of these are publicly visible without any special access.
| Field | Where to find it | Why it earns its place |
|---|---|---|
| Release date | Store page | Anchors everything else to a market period |
| Price | Store page | Revenue per unit, and a positioning signal |
| Review count | Store page | The best public proxy for units sold |
| Review score | Store page | Separates "sold well" from "was received well" |
| Top 5 tags | Store page | Tells you the shelf it actually competes on |
| Follower count | The game's Steam community hub | A public demand signal, unlike wishlists |
| Supported languages | Store page | A direct read on the budget behind it |
| Has a demo | Store page | Whether the tag's audience expects to try first |
Wishlist counts are not public. No tool returns another game's wishlist count, and any figure presented as one is derived from follower counts with a multiplier. Record followers, which are real and public, and treat any wishlist number as an estimate with a wide range.
One caveat on followers, since every comp is a released game: the follower-to-wishlist relationship only holds pre-launch. After release, followers keep accruing from people who already own and play the game, so on a comp they measure accumulated interest rather than pre-launch demand.
Review count is a proxy for units, not a measure of them. Only a fraction of buyers leave a review.
The fraction varies a lot by genre: roughly 2 to 3 percent for mainstream casual audiences, 5 to 7 percent for CRPGs, and 8 to 11 percent for Souls-likes, on Immutable's calibrated Steam genre multiplier table.
Take the two ends of that range: a game that sold 10,000 copies would carry around 250 reviews as a casual title and around 1,000 as a Souls-like. So review counts compare well within a tag and badly across tags.
The trap: picking comps you would like to be
Left to instinct, a studio picks the games that made them want to build theirs. Those are, definitionally, the games that succeeded enough to be memorable.
Building a forecast on those games is the same error as reading a genre's average revenue instead of its median: it describes the top of a power law, not the distribution you are entering.
The correction is mechanical. Once you have your five to ten comps, sort them by review count and use the middle one, not the best one, as your reference.
Then sanity-check the middle comp against the median for your tag: what the median game in your genre earns on Steam has the figures, and in most sub-genres that median is under $10,000 gross.
If your middle comp is far above your tag's median, your set is still aspirational and you have not finished filtering.
A worksheet you can copy
One row per comp, one column per field. Fill it in a spreadsheet and keep it with the pitch deck.
| Game | Released | Price | Reviews | Score | Top 5 tags | Followers | Languages | Demo |
|---|---|---|---|---|---|---|---|---|
Then answer three questions underneath it:
- What is the median review count across these games? That is your reference point, not the maximum.
- Which of them are in your tag's middle rather than its top? Name them explicitly.
- What do the ones that did best have that you do not? Usually a budget line, a publisher or an existing audience. Write it down, because it is the honest gap between your plan and the outcome you are quoting.
What this does not cover
Finding comps in the first place is covered in how to market an indie game, including the More Like This carousel and the follower-to-wishlist rule of thumb for unreleased titles. This page starts after you have candidates.
Where comp selection sits in the wider sequence is how to market a Steam game, whose first stage is genre-market fit: pick a genre with enough demand, then read it through comparable titles. A comp set is an input to that decision, not a substitute for it.
Turning a comp set into a wishlist target is how many Steam wishlists you need before launch.
Choosing the tags themselves, as opposed to reading them off a comp, is a separate decision that this page does not make for you: it assumes your tags are already set.
And a comp set describes a market, not your game. Use it to bound expectations, then test the actual proposition with a demo, a festival or a live page.
