What a wishlist actually measures#
Adding a game to a Steam wishlist costs a player one click. In exchange they get an email when it launches or goes on sale, and Valve gets a signal of interest. That makes wishlists cheap to collect and noisy to interpret. Some come from people who will buy on day one, some from a trailer someone half-watched two years ago.
Three properties make them useful anyway:
- They're counted the same way for everyone, so benchmarks across games mean something.
- They trigger launch notifications, so they directly cause first-week sales, not just predict them.
- Steam's own systems pay attention to them. Valve doesn't publish its ranking formula, but developers consistently report that wishlist activity and early sales affect how much visibility a launch gets.
Conversion: the benchmark#
The figure to know: GameDiscoverCo found a median first-week sales-to-launch-wishlists ratio of about 0.105 for Steam games released August to October 2024 with at least 5,000 wishlists. Write it as
"Median" is doing a lot of work. Conversion varies by a factor of ten across games: price, genre, review score, launch timing and how much of the wishlist base was earned recently versus long ago all move it. Treat as a distribution, not a constant. In our launch model we use a lognormal with a log-sd of 0.8, which puts the middle 80% of games between about 0.04 and 0.29.
Note that the ratio counts all first-week sales, including people who never wishlisted, divided by wishlists. It's a ratio, not a literal conversion rate, which is why some games exceed 100%.
How many wishlists do you need?#
"Enough" depends on what you need the launch to pay for. Invert the conversion relation with a lifetime multiplier (lifetime units divided by first-week units) and net receipts per unit :
For a $20 game ( after Steam's 30% fee and ~12% refunds and tax), and an assumed :
| Goal | Revenue target | Median-case wishlists |
|---|---|---|
| Pay back a $150k self-funded build | $150k | ~39k |
| Recoup a $600k publishing deal | $600k | ~155k |
| Fund the next game ($1.5M) | $1.5M | ~387k |
These are median cases: half of games with that many wishlists do worse. For 80% confidence on the $600k deal, our Monte Carlo needs about 335k. That gap between median and confident is the most underestimated number in indie planning.
Estimating sales from reviews (the Boxleiter method)#
Steam doesn't publish sales, but it does publish review counts, and a roughly stable fraction of buyers leave a review. Multiplying reviews by a factor gives a sales estimate:
This is usually called the Boxleiter method. Jake Birkett's 2018 analysis of games with known sales found an average multiplier of 82 and a median of 77, with a wide range from 30 to 150. Newer games in his data skewed lower, with medians in the 60s. Use it with care:
- Use a range, not a point. Report to reviews, not one number.
- Genre and audience matter. Players of some genres review more readily than others.
- It counts Steam only, and misses bundles, key resellers and other stores.
The best use isn't estimating a competitor's revenue. It's building your own comparables table: take five similar games, estimate their sales at 1 month and 12 months from review counts, and you have a genre-specific estimate of the lifetime multiplier , the most important assumption in any wishlist forecast.
Velocity beats totals#
A wishlist count is a stock; what predicts a launch is the flow. Two games with 40k wishlists can be in very different positions: one gaining 400 a day in the month before launch, one that peaked at its announcement two years ago and has been flat since. Old wishlists convert worse, so track:
- Daily net additions (adds minus deletions) over trailing 7 and 30 days.
- Spikes per beat: wishlists gained per trailer, festival, demo update or creator video. That's your marketing ROI, measured.
- The share earned in the last 90 days before launch. More recent is better.
Festivals are the biggest single spikes most indie games get. Steam Next Fest in particular lets each game participate once before release, so time it for when your demo is strongest and your launch is close enough to convert the attention.
A worked forecast, start to finish#
Here's the full chain for a hypothetical $18 narrative game, three months from launch with 52k wishlists and gaining about 300 a day.
1. Launch wishlists. 52k + 90 days × 300 ≈ 79k, before the launch-week bump. Call it .
2. First week. At the median ratio, units. Use a range, not the point: at a log-sd of 0.8 the 10th–90th percentile ratios are about 0.04 and 0.29, so roughly 3,000 to 23,000 units.
3. Lifetime multiplier. From five comparables' review counts at one month and twelve months (×30–80 each, ratio of the two), suppose you find . Lifetime median ≈ 21,800 units.
4. Net receipts. , so median lifetime net receipts are about $242k. Regional pricing pulls that down. If a third of sales come from regions priced 40% lower, the blended drops by about 13%, to ~$9.60, and receipts to ~$210k.
5. Decision. A recoupable budget of $200k is roughly a coin flip to earn out. $400k is a long shot. A self-funded team with $150k spent is likely, but not certainly, in profit. This is the conversation to have with a publisher, with the numbers on the table. Our Monte Carlo does steps 2–5 with the uncertainty carried through properly, and the deal calculator turns receipts into what the studio actually takes home.
Wishlists for AI-native games#
AI-native games have a demo advantage and a disclosure question. The advantage: a demo where players can actually talk to a character produces the kind of emergent clip that spreads, and every clip is a wishlist driver. Build clip sharing into the demo. The question: Steam requires disclosure of AI-generated content, and live-generated content needs a description of your guardrails. How you write that disclosure affects how players react to the store page. See AI disclosure and ratings and the disclosure checklist.
One more thing: inference cost scales with players, so a surprise hit launch is also a surprise bill. Plan capacity from your wishlist forecast. Our capacity planning research shows how.
Frequently asked questions#
What percentage of Steam wishlists convert to sales?
GameDiscoverCo reported a median first-week sales-to-launch-wishlists ratio of about 10.5% for Steam games released August–October 2024 with at least 5,000 wishlists. It varies by roughly 10× across games.
How many wishlists should I have before launching on Steam?
Work backwards from the revenue you need: divide it by net revenue per unit × first-week conversion × lifetime multiplier. A $20 game targeting $600k needs around 155k launch wishlists in a median case under our assumptions.
How do you estimate Steam sales from reviews?
Multiply the review count by a factor, commonly between 30 and 80. Jake Birkett's 2018 dataset found an average of 82 and a median of 77, with newer games lower.
Do old wishlists still convert?
They convert less well than recent ones. Track net wishlist additions in the months before launch, not just the total.
Sources#
- Game World Observer, "Wishlist-to-sales ratio on Steam: GameDiscoverCo benchmark", 6 Dec 2024. gameworldobserver.com/2024/12/06/wishlist-to-sales-ratio-steam-gamedis
- Jake Birkett, "Using Steam reviews to estimate sales", Game Developer, 4 May 2018. www.gamedeveloper.com/business/using-steam-reviews-to-estimate-sales
- Neowin, "Valve introduces new revenue share tiers for Steam", 2018. www.neowin.net/news/valve-introduces-new-revenue-share-tiers-for-steam