What 'long tail' should you expect for your PC game in 2026?

We crunched 'too much data'. Also: new GDCo show, news and lots more...

[The GameDiscoverCo game discovery newsletter is written by ‘how people find your game’ expert & company founder Simon Carless, and is a regular look at how people discover and buy video games in the 2020s.]

Happy, uh, Tuesday, is it? And almost August already? This year is flying by, and ‘too much news’ certainly seems to be part of it. (Look, we’re trying to catch it all for you, like Tobey Maguire/Peter Parker caught that tray. Hopefully not with 156 takes.)

Before we start, congrats to the winners of GameDiscoverCo’s ‘Scrappy Saves The Animals’ game jam with the Oakland Ballers on Itch.io! Check out video of the entrants, and you can play the top games - including the winner, Scrappy’s Rescue Run. Games will be showcased at the video game-themed Ballers game this Friday.

[THE DEEPEST PC/CONSOLE DATA? You can get a free demo of our GameDiscoverCo Pro company-wide ‘Steam deep dive’ & console data by reaching out today - >110 orgs have it. Or, signing up to GDCo Plus gets the rest of this newsletter and Discord access, plus more.]

Game discovery news: Paper Shredder, Guildrun?

Let’s start things out with a look at the various game platform & discovery news - stacking up like this:

• GDCO Pro's estimates of 'trending' unreleased Steam games by 'new wishlists in the last 7 days', July 13th to 20th, is, duh, headed by Dear Passengers. But all of the Top 3 are new, with first-person cleaning game Paper Shredder (#2) also getting a ton of short-form video love, and adding ~140k wishlists quick.

• The final new entry is super-slick, great looking autobattler roguelike Guildrun (#3), whose wishlists are arguably more 'solid' than the other top games, in that it has a playable demo which already hit 11k CCU on Steam. We've talked about other top titles before, inc. Soulframe (#4) and Dinoblade (#5, out this week!)

• Mainframe’s Sulka Haro did a big data dive on publicly accessible ‘Steam x AI’ data, revealing: “An AI-flagged game reaches a real modest-success tier (≥100 reviews ≈ ~3,000 sales) at only about 55% of the non-AI rate - and that ratio hasn’t improved in two years.” A second part adds: “AI games rate lower - but not catastrophically.”

• Steven Lake’s interview with Edmund McMillen reveals that after “Steam recommended we do this extreme 90% off sale” for The Binding Of Isaac Complete Bundle in the Steam Summer Sale, it had “close to 4 million” extra units sold, many at ~$5. (Catalog titles - you really want to be in that ‘90% off’ feature spot!)

• Academic and game maker Paolo Pedercini - ever provocative - has article-ized a talk on ‘A History Of Games Through Mass Extinctions’, and he notes: “Industry commentators* have been talking about a deprofessionalization of the Western game industry” (*me), adding: “Science Fiction author and futurologist Bruce Sterling has this maxim: ‘Whatever happens to musicians happens to everybody’. He sees musicians as the patient zero for the precarization of all cultural industries.” Spicy stuff.

• One of the folks hacking devs to inject malware into their Steam games & grab player crypto got caught. Here’s the full legal filing against him - he “infected the devices of some 8,000 people, the FBI said, gaining access to about 80 crypto wallets and stealing at least $220,000.” A 1 in 100 wallet success rate still pays off, eventually…

• Roblox things: they’ve announced the UGC-centric Build - “a new mobile-first creation tab within the Roblox app, and a new suite of AI-powered tools within Studio for creators of every level”; one investor claims successful Roblox creators get “terrible, predatory offers from deeply unethical hypercapitalistic companies on the platform who intentionally take advantage of stark deal asymmetries with 15-20+ year old devs.”

• The latest Xbox Game Pass titles to debut on the subscription platform include big titles (the awaited Halo: Campaign Evolved, Game Freak’s intriguing Beast of Reincarnation) and some trendier breakouts (Mistfall Hunter, Corsair Cove, Shift At Midnight.)

• Valve just dropped a decent-sized update to SteamVR ahead of the (presumed) Steam Frame VR hardware release. The updates “brings a good slate of standard Steam features to SteamVR [and] a more VR-native user experience, in part thanks to its new UI for SteamVR’s Dashboard”, according to RoadToVR.

• A mobile platform ICYMI: “Epic Games and Google… jointly withdrew their attempt to retroactively settle the lawsuit that’s changing how Android app stores work in the United States - and that means Google will be forced to carry rival app stores inside of its own.” (As soon as Wednesday, apparently?)

Before we move on, we wanted to show off a new feature in GameDiscoverCo Pro. We added automagic graph annotations for Steam demo launches, Next Fest and 190+ (!) streaming showcase events. Here’s a particularly good example, tracking the rise of co-op Scottish weapon brawler Tears Of Metal - out this week:

Our Sponsor: Player.gg, where playtesters find your game

Here’s a thing nobody tells you about playtesting: your first testers aren’t just QA. They’re your first wishlisters, your first Discord members, your first “hey have you seen this game” moment. But only if they actually wanted to play it in the first place.

Player.gg by FirstLook is a community of thousands of players actively hunting for their next game. When they find yours, you’re not just getting bug reports. You’re building an audience before launch.

• Players discover your game because they were looking for it

• Every tester is a potential day-one player

• Free to list, live in 5 minutes

500+ studios are turning playtests into communities. Yours could be next: set up your free account at FirstLook today.

What’s the ‘long tail’ for your PC game in 2026?

So your game launched on Steam. Your first week (or month’s) sales are in. That’s a good starting point. But how much money can you expect at the end of the first 6 months, or 1 year, or 2 years? We’ve covered this before, notably back in 2021 with a reader survey, but a GDCo subscriber pointed out we didn’t benchmark it recently.

Luckily, we have data estimates for every single Steam game. So we looked at a host of recent titles - all games launched from July 2023 to July 2025 (so they can accumulate at least a year of sales) with Month 1 sales of >1,000 units.

And here’s the headline numbers. Voila: the copies sold/revenue multiplier across all 5,000 games in that cohort, with Month 1 being 1.0x revenue and copies sold:

That’s a lot of numbers! But to simplify things: if you had 10,000 copies sold at the end of Month 1, the median result says you’d have 14,600 cumulative sold by Month 6, 17,900 by Month 12, and 22,700 by Month 24. (LTD median gross revenue multiple is about 7% lower after Month 12, and 12% lower after Month 24, due to discounting.)

We also know people like to use other multipliers in their calculations, specifically ‘Week 1 sales’ as a base, so we can easily show multipliers for common milestones:

As you can see, if you sold X copies in Week 1, you will - at median - have sold 1.39x those copies by Month 1, 2.59x the copies by Year 1, and 3.32x copies by Year 2. (And the revenue is slightly different - 1.42x the revenue by Month 1 (more, due to launch discounts expiring!), 2.46x by Year 1, and 2.99x by the end of Year 2.

Was this more than you expected? Less? The only thing we’ll note is that a median hides big swings. For example, Lethal Company has a 4.14x Week 1 to Month 1 sales ratio, Mouthwashing has a 2.54x ratio, but Fuga: Melodies Of Steel 3 had a frontloaded 1.04x, and Wargroove 2 had 1.08x. (Still, it’s a great starting point…)

And the important point is this: if you see ‘hey, X game grossed $1 million in Week 1’, it’s quite likely to gross $3 million in the next couple of years - and eventually stretch out to $4-5 million or more. (Unless horribly frontloaded.) So there’s some significant residual revenue to arrive there, that you should plan for…

Since we know people love using Week 1 as their benchmark, here’s the same graph as above, but with Week 1 as 1.00x instead of Month 1. And perhaps unsurprisingly, long tail is down from the 2021 survey, when we had Year 1 as 3.05x Week 1 revenue. It’s now (well, as of 2023-2025) showing Year 1 as 2.46x Week 1 revenue, about 20% down.

(We’d expect that, because there’s been strong Steam revenue growth since 2021 - as much as +100% in aggregate - but the # of launched games is outpacing that. And that’s likely to affect decisions on increased choice made during the first year or two of release. We hear 2017/2018 was even more halcyon for fatter ‘long tail’. But that’s practically the Jurassic…)

Finally, we decided to look at the data through a few lenses. First, with the rise of cheaper games, we thought we’d separate out <$10 games from >=$10 ones, discovering that more expensive titles have slightly better long tail (2.7x Week 1 copies sold in Year 1, up 4% compared to the 2.59x median.)

Since most <$10 titles are more semi-professional and have less likelihood of appearing very cheap on discount, cos they’re already cheap, they fare worse. (Their Year 1 units sold long tail is 10% less than median, at 2.32x.)

Then there’s multipliers by total Week 1 sales, which show - perhaps unsurprisingly - that games that sell better on Week 1 have a better long tail. In the 1k-10k space, presumably affected by some hyped games that were received poorly, you only get 1.96x Week 1 sales in Year 1.

But if you sold 10k-100k copies in Week 1, you’re hoping for 3.16x as many copies in Year 1 (22% above the median), and if you sold 100k+ copies in Week 1, we show a huge median of 4.64x. (But again, viral outliers are going to be affecting this…)

Finally, we hear co-op games are hot, and therefore may have a more pleasant long tail, due to friends recommending to friends, better streamability, etc? Well, the data bears this out, with 3.53x Year 1 sales compared to Week 1 for all games that had co-op, online co-op, etc in their Steam tags.

That’s a super-impressive 35% up on the median Year 1 to Week 1 multiplier of 2.59x. We would consider designing co-op into your next game - if appropriate. (Seriously, it’s the #1 thing we ask people to be added, after seeing a promising idea.)

GameDiscoverCo Show: AI, refunds, Apple Arcade!

After a brief holiday/sickness hiatus, The GameDiscover Show has returned, featuring both myself (Simon) and Post Games’ Chris Plante. Reminder: you can get it on Apple Music, Spotify & elsewhere (audio), with a video version on YouTube (above).

This time out, we talk about GDCo’s recent AI survey, whether Steam refunds for short games are bad news & how the Apple Arcade catalog has settled in, over time. Plus we discuss where Meccha Chameleon fits in the Steam ecosystem. (And I find out I’ve been pronouncing ‘Meccha’ wrong…) Fun times…

[We’re GameDiscoverCo, an analysis firm based around one simple issue: how do players find, buy and enjoy your PC or console game? We run the newsletter you’re reading, and provide real-time data services for publishers, funds, and other smart game industry folks.]