What Dome Keeper's creator 'gets' about shipping the right game
Also: the latest Steam charts for Pro/Plus users, and some Gamescom news.
[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.]
We’re back, and congrats to the Gamescom leavers who have abandoned the event to ‘the public’, lol! We’ll analyze titles with biggest player interest next week. In the meantime, we have a Q: which of the 50 Indie Games That Changed The World (it’s a new book - here’s a review) were you most happy - or surprised - to see on there?
Also before we start: we’re being generous with free newsletter content nowadays. So if you like what we’re doing, please support us with a GDCo Plus subscription - if an indie - or via a GameDiscoverCo Pro sub for your company. You get awesome data (LOTS of it, inc. more Steam & console, w/Pro), Discord access & more newsletter. Ta.
Game discovery news: lotsa games at Gamescom!
Let’s kick off with a catch-up on the discovery and platform news that engrossed us since earlier this week:
The full Opening Night Live streaming event at Gamescom is viewable here (or above!), but we found this RockPaperShotgun analysis and a TWIG rundown of the 40+ trailers both good overviews. Notable reveals included a new Lego Batman game from TT Games and an early teaser on the next Black Myth game.
From our platform perspective, the Xbox-owned MachineGames & Bethesda’s Indiana Jones & The Great Circle had two interesting elements - confirmation of a Switch 2 release in 2026, and the trailer end slate again showcasing the now default triple Xbox platform branding: ‘Xbox Series X|S - Xbox on PC - Xbox Cloud’.
PlayStation’s pre-tariff stockpiled PS5s ran out, so the hardware’s going up in price in the U.S. New prices as follows: “PlayStation 5 – $549.99 (from $499.99); PlayStation 5 Digital Edition – $499.99 (from $449.99); PlayStation 5 Pro – $749.99 (from $699.99)” (And Sony’s moving U.S. PS5 production outside of China…)
The Future Games Show showcase also went down, and the Epic Games Store editorially rounded up highlights, including Hell Let Loose: Vietnam & more - many trailers are on the FGS YouTube channel, with Euro/American Truck Sim for console and a new Halloween game reveal some of the most-watched.
The ‘Xbox handheld’ ROG Ally devices announced an October 16th release date, and are playable at Gamescom, with Xbox confirming a Handheld Compatibility Program, rating games ‘Handheld Optimized’ or ‘Mostly Compatible’, similar to Steam Deck ratings. (No final pricing yet due to “macro-economic” reasons, tho!)
Microlinks: the latest Xbox Game Pass additions include Gears of War: Reloaded, Dragon Age: The Veilguard, Void/Breaker & more; the EVO fighting game champs, owned by Sony since 2021, got sold to India’s NODWIN Gaming; Matej Lancaric’s longform piece on softlaunching mobile games is very instructive.
Kate Koroleva’s been testing Steam trailers on social media to get metrics on watch time/interest, noting: “Raw gameplay from second one outperformed everything: no cinematic intros, just gameplay.” (And a narrative voicover gave them a 2x boost in views and watch time.)
For folks with a game out soon & worried about that Hollow Knight: Silksong release date (Sept. 4th!), my comment in the GDCo Plus/Pro Discord: “I think disruption effects are… overstated in general. I would be worried if i had a Metroidvania launching in Sept.” Looks like devs disagree, tho, and discourse is hot, but I get it.
Valve’s rumored console-like ‘Fremont’ TV-focused SteamOS device popped up on a PC benchmarking site. If it ends up being the spec of the Geekbench benchmark, commenters suggest it’s “comparable to the base PS5/Series X, but with better power efficiency and probably slightly better ray tracing performance.”
My latest column for SkillUp & friends’ This Week In Videogames site ($) - for a B2C audience - is about whether Steam wishlists matter: “Getting more wishlists for your game should always be the goal of the developer. Just don’t do it by paying people, hacking Steam, or summoning the undead to mash the wishlist button.”
Esoteric microlinks: the gamer motivation gods at Quantic Foundry look into whether young people are getting less conscientious; film nerds, look at what The Quorum says about more accurate box office forecasts and rejoice; the biggest U.S. PC game publishers in June 2000? EA, Havas (!), Hasbro & Infogrames…
What Dome Keeper creator 'gets' about shipping the right game
We don’t often feature a talk as our lead story for the newsletter - although I (Simon) did help originate the GDC YouTube channel, which has >70 million views nowadays. But we spotted René Habermann’s talk (above) that was part of the French Game Camp event in June (here’s the other talks in FR/EN), and just had to feature it.
Why? René is both the lead creator of 2022’s ‘roguelite action miner’ Dome Keeper, which has 1.4m copies sold LTD, and also the upcoming first-person ‘planetary defense sim’ PVKK*, due out in Summer 2026, and already much-loved by streamers. It has 37k+ Steam followers, and we’re guessing it’s at ~700k wishlists. (*Yes, PVKK’s full name is Planetenverteidigungskanonenkommandant, which is a) deliberately the most German thing ever b) extremely funny, right?)
So this is an interesting starting point - somebody making two successful-looking games, but in very different styles/genres. And the talk is called ‘Don’t Ship the Wrong Game: A Framework for Choosing What to Finish’ - an absolute sweet spot for GDCo, since we firmly believe that product-market fit trumps everything else in this space.
The central thesis of the talk is that there are three different ‘types’ of wrong games that devs can make. And the first is simple - it’s the ‘undesired game’:
As René says, this is “basically a game that releases… and no one was really waiting for that game. No one really wants to play it.” And the big issue? The idea was unvalidated: “Maybe you put out a demo, maybe you put out an announcement trailer, but no-one cared… no-one played the demo, but you still kept on developing this game.”
Is this irritatingly obvious? Well, yes and no! The ease of making video games in the 2020s allows everyone to dip in and out of project concepts quickly. And Dome Keeper, like The King Is Watching, was started as a Ludum Dare game jam concept that got a lot of positive traction. And here’s how René thinks of project structure:
Watch the full video for a better explanation, but it’s a systematic process of gradually expanding from prototype to ‘core fancy’ (MVP?) to demo to release, sharing externally, while having real validation points where you might kill the idea or pivot.
There are obviously many games that it’s difficult to ‘partially’ commit to, from a tech or scaling perspective. But I do - like René - feel like the game biz has weird issues with the idea that you can ‘just’ start and stop projects. Is it an ego or flexibility thing? But you totally can! (If your project is the right scale, or you have the right funding.)
The second different type of ‘wrong’ game - perhaps less wrong - is what René calls the ‘misaligned game’. We’ve talked about this before as ‘expectation management’, and it’s that “there was a vision of the game communicated, but… the game that released is something else somehow.” (Sometimes this is ‘cos of implied visuals or graphical themes.)
As a minor example, René notes that Dome Keeper always gets users adding the Steam tag Tower Defense, despite the fact the devs keep removing it, saying: “people want to believe this is a tower defense game, even though it only shares like 5% of the DNA” of that subgenre. (Luckily, that didn’t affect its success too badly, hah.)
One great point he makes is that defining “experience goals”, rather than a raw list of features, is a way better way to align the product. So it’s more - what is the game, and what do people like about it - and less ‘here’s the gameplay mechanisms’?
Finally, the third type of game is what René calls ‘the overdeveloped game’ - the idea that “people are interested… and this game could work out, but we spent 7 years” on it. So the eventual $ return just isn’t going to match the spend on the game.
This is particularly relevant in today’s crowded market. And there are various reasons for this, as listed above. Perhaps the devs are too attached to making the game ‘perfect’, there’s no strict deadlines, or the devs “redo stuff over and over again” without necessarily massively improving things. Though we agree that games take time.
He also singles out art scoping: “For some games, having fantastic art is great - and for other games, it’s nice, but it doesn't change the commercial success of the game.” René cites Schedule I: “the developer could have spent three more years on trying to make photorealistic 3D art… but no one cares about this, and the game would not have sold more if it had nicer looking graphics.”
Finally, René is actually a GDCo Plus/Pro Discord member, so we DM-ed him to ask him a follow-up question: “Of the three types of games you list: undesired, misaligned, and overdeveloped, what do you see the most of? And which one is it MOST difficult to avoid?”
He pondered this and replied: “Most common, I'd say, is undesired. It's so easy to do a passion project and never check if anyone else likes it. Most challenging to avoid… is overdeveloped, because it's so subtle and it depends so much on the game. Sometimes investing tons into art is worth it. Brotato would not have done better with nice graphics. Expedition 33 might have not been this well received with pixel art.”
There’s actually a whole second section of his talk on prototyping that looks great too, but we are so out of room/time, here. So once again, thanks to René - and Game Camp - for their work on fleshing out some important thinking.
Steam debuts this week: a diverse set of winners!
Moving on, let’s take a look at the big Steam debuts this week, for the benefit of our Plus/Pro subs. And there’s an interesting trio (above) atop the new game charts by peak concurrent user, plus a couple of mid-sized disappointments: