[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.]
Well, just two months left in 2026 - luckily there’s just no major game releases left, so nothing to see here, folks. *draws picture of GTA 6 box on wall like Wile E Coyote does with a train tunnel and is surprised to get run over by it* But seriously, we do have things to discuss besides the grand thefting of automobiles… see below!
Oh, and heads up - we are 95% sold out of newsletter ads for the rest of 2026. But if you’d like to offer your product or service to GameDiscoverCo readers (>46k of them!) in 2027, please ping us & we can get you our specs, clickthrough #s & rate card. (Ads really took off this year - thanks to all the companies supporting us that way…)
[NEED THE BEST DATA? You can get a free demo of our GameDiscoverCo Pro company-wide ‘Steam deep dive’ & console data, inc. MCP server, by reaching out today - >110 orgs have it. Or, signing up to GDCo Basic gets extra newsletter and Discord access, plus more.]
Gamedisconews: Drawquarium, Hull Rupture arrive
Let’s start everything out with a look at the top game discovery & game platform news since we last checked, as follows:
GDCo Pro's estimates of 'trending' unreleased Steam games by 'new wishlists in the last 7 days', Oct 26th-Nov 2nd sees AI-infused 'too many zombies' survival strategy game Ebbtide (#1, +140k wishlists) again going somewhere. #2 is Football Manager 27 (free demo) & #3 is 2D spaceship management game Hull Rupture.
Elsewhere, there's a lot of former Next Fest titles dotted around the Top 10. Newer, though? There's 'draw your own fish & take care of ‘em' Drawquarium (#5), which has an adorable trailer, and first-person mech repair job sim (clever idea!) Bay 12: Mech Repair, brand new at #12. There's always new games to find...
Physical game percentages? Even for Resident Evil publisher Capcom, they’re a tad desultory at this point: “Its Q2 results for fiscal year 2026 revealed that 92.4% of [unit] sales were digital copies”, split 57% PC and 35% console. (Remember this includes discounted back catalog which moves in higher numbers - nonetheless.)
Are there more PC games coming out right now? ICO’s Thomas Bidaux points out 1,656 releases (!) last week on Steam, and “The number almost doubled from the previous record a few weeks ago (757).” Related: Itch.io says “The number of game pages being created per-day has doubled in the past 2 months [to ~6k/day].”
PlayStation things: the latest PS+ Monthly games for Nov. 2026 include Dead Island 2, Tomb Raider IV – VI Remastered, and the entertaining Monster Train 2; it’s the 20th anniversary of the online PlayStation Store in a few days, so devs got asked their favorite PS Store games: Balatro, God Of War, Inscryption mentioned.
Valve’s Tom G. had detailed comments at a Korean event re: Steam’s current clone/game flood worries & whether platform restrictions make any sense. Well, no: “Making Steam a smarter, more dynamic platform rather than restricting developer access is the starting point of our philosophy.” (Disclosure is their preferred approach.)
In the same Steam Indie Connect meetup, Giardino suggested - in a kinda fair, but likely to be controversial point: “Rather than relying on a single mechanism that someone can easily copy… you [can] grow your game by adding features, personality, and creativity that stem from your own perspective as a developer.”
What’s an alternative way to ‘AI disclosing’ games? That’d be VRChat’s decision to have no AI labels: “95% of survey respondents want some kind of label for AI usage, but there’s very little agreement beyond that… there isn’t one simple answer here.” (They’re thinking about having a “self-declared ‘Human-authored’ mark”, tho.)
You have to be scrappy in today’s funding environment, with the Trash Goblin devs revealing that only about one-third of their game’s funding up to the 1.0 release of the game was from Kickstarter/game sales. (The rest was UK government funds & tax relief, at ~35%, platform support at 5% and own funding.)
Meta’s newest financials? “Within our Reality Labs segment, Q3 revenue was $443 million, down 6% due to lower Quest headset sales, which were partially offset by continued strong growth in AI glasses revenue.” Also: Meta will ship new hardware iterations only when “we have major breakthroughs.” (No new Quests soon?)
The latest Steam docs change? If you want to change your game name after store page review, you need to answer these Qs: “Why is this name change necessary?… Is there anything about the existing game name that you need to change for legal reasons?… Is the product that was advertised to customers changing along with the name?” (Aha!)
How Probably Stolen traded up to 750k copies sold in 5 days…
Recently, we were poking at the success of the demo of Questing Goose & Tinybuild’s “a cyberpunk shopkeeping simulator” Probably Stolen ($20) in the GDCo Pro/Basic Discord, just before the game shipped on Oct. 28th. Could the constant 10-20k concurrent users on Steam for it - spiking to 40k - really be reflective of interest?
Those numbers seemed preposterously high, leading us to start looking under the couch cushions for reasons. But then the full game launched - and immediately hit 85k CCU, peaking at 165k CCU earlier today. That’s for a single-player only, menu heavy inventory management game. It’s not fake. What on Earth is going on here?
Well, the CCU is definitely enhanced by long play times: Tinybuild told us that of their 180,000 playtest players, 45,000 spent more than 20 hours playing. (GDCo is estimating average playtimes for the full game zooming past 10 hours already.) But the game has sold 750k copies, as of yesterday, and is clearly going to do multi-millions.
So the Tinybuild team - publishing producer Artyom Safarov, global marketing lead Arnaud Richard, and Chinese marketing lead Dandan Ye - were kind enough to answer all our questions in a giant document, which we highly recommend you read. But if you don’t, here’s the gist on takeaways:
Probably Stolen is a major rebuild of a game Tinybuild was pitched: you can still see the original title, Deepspace Emporium, on Steam. The publisher ‘couldn’t get enough’ of the demo’s gameplay, but existing art (from programmer/designer Bei Ning Pan) was ‘programmer art’ adjacent & the game had already been in Next Fest.
The graphical rehaul was twinned with interesting new gameplay ideas: After coming on as publisher, Tinybuild got “a talented pixel artist who took charge and created a whole new visual style for the characters, environment and items.” Along the way, existing store management & counterfeit detection gameplay got deeper, with the idea of “sentient rogue AI” now controlling some of your customers.
These new ideas pushed the game into the hot cyberpunk game space: look, people can’t get enough of cyberpunk games (see: Nivalis Nights and, uh, Cyberpunk 2077.) So Artyom noted: “Do we keep it a generic sci-fi or move to the cyberpunk which (arguably) is a more catchy setting?” The latter won out!
So what we have here is an already deep trading & inventory/item management game which also had counterfeit detection a la Papers Please, and now also No I’m Not A Human-style fake detection, and had a hot setting & developed a lot more depth (class systems, special ‘runs’ with weird goals, a ton of diverse customers & Easter eggs.)
This was all done with one solo Canadian dev, plus, per Artyom, “artists and contractors for music, sounds and UI. Depending on how you count, it’s either a solo project or a game made by a team of under 5 people, I guess. The team is remote, spread across Northern America, Europe and the rest of the world. They have never met each other in person.”
Just finishing off on what makes this game special, in our view: the busy inventory screens make it feel deep, there’s a lot of the classic ‘grinding and collecting’ in there, but there’s also always new things to see. Don’t underestimate how much a tiny team can add depth to a cleverly structured roguelite. There’s just an ocean of stuff to do…
Dandan Ye notes that “several interconnected systems, including production and crafting, inventory management, faction reputation, and narrative progression” are key: “Even a single item can present several choices.. sell it for profit, use it yourself, turn it into something more valuable, trade it for faction reputation, or even give it away just to feel better about yourself.”
Artyom adds, re: depth: “Do you know that you can genetically modify the plants? And that the water purity is actually simulated when you drop a chemical tablet in it? You can also breed rats by combining the specimens with particular features, and creating pure-grade drugs [using chemistry].” (There are detailed rat ranching guides on YouTube already.)
We have room for one more set of three bullet points, and the Tinybuild team wrote another 12 pages (with making-of art! go read it all!) But here’s what we found notable in the pre-launch & launch phases for the game:
Despite being a late-announcing game, a lot of marketing poured in: this game was unveiled in late August, and still did this well? Blimey. Arnaud Richard notes: “There was organic momentum, but also a lot of work involved in understanding it and adapting the experience. By the day before launch, we had generated 578,528 wishlist additions, with strong growth both in China and elsewhere.”
Non-obviously, China is a particularly strong market for the game: early momentum via Bilibili was key, and 51% of the 246k playtest activations were from China, with 17% from the U.S and no more than 3% from any other country. So working with Chinese publisher XD and marketing there has been vital.
Pricing was carefully done, manual, and seems to have worked well: full write-up in doc, but a “mostly manual” approach to pricing has the Chinese price at just 55 yuan ($8.25), a 59% discount on the U.S.. Despite being more expensive in the U.S., “CIS, MENA, SASIA, and LATAM are priced around $7-8.” This approach semi-ignores Steam recommendations in favor of sentiment & seems to have worked.
It’s interesting - in digesting this last point, you can argue that yet again, an inexpensive, longer-playing game that’s $5-$15 in most territories is breaking out. (Despite it being a tad more, $17.99, at U.S. launch.) We’re getting to a new normal with this ‘cheaper but deep’ trend - and it’s not what I would’ve expected 3 years ago.
Finally: in a world where we thought 30-minute ‘runs’ are the normal for roguelites, one of the reasons Probably Stolen is being played for so long is that even speedrun-style runs are 2-5 hours, and full runs can be 10-15 hours. But the game hook pushes you through that like a knife through randomized butter - and everyone’s happy.
[Again, congrats to Bei Ning Pan and the Tinybuild crew for making a narrow-spec but super deep game & thx for sharing it with us.]
GameDisco Show: is Discord the new SEO Reddit?
Before we quit, here’s the latest The GameDiscover Show, ft. myself (Simon) and Post Games’ Chris Plante talking trends - sub to the podcast via Apple Podcasts, Spotify & elsewhere (audio), with a video version on YouTube (above).
The latest GDCo Show talks about an important discoverability change for Discord servers, why average wishlist count increases in October's streaming showcases were slower than earlier in 2026, the significance of a game citing dev costs in increasing its post-launch default price, and what Roblox's latest results say about the platform.
[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, devs, and other smart game industry folks.]