This week’s theme connecting several of the strongest stories is making value easier to see. From Steam matching games with the right players to casual puzzles built around satisfying transformations, clarity is becoming a powerful advantage in discovery and design. But visibility alone won’t pay the bills: Mario Kart Tour and Nerial offer sobering reminders that enormous reach and creative distinction still need a sustainable business underneath. Let’s get into it:
Steam is making one of its most important discovery surfaces fully algorithmic
Valve is replacing the manually scheduled placements in Steam’s Discounts & Events section with personalized recommendations. During testing, Valve says the new system showed customers 10× more games per day and produced meaningful increases in store-page visits, wishlists and cart additions. One publisher, Hooded Horse, reported that a niche title received more than 7,000 visits from the section in one month after previously receiving almost none.
Why it matters: Steam discovery is moving another step away from editorial gatekeeping toward behavioural matching. For smaller games, that could be positive: a niche title no longer needs to win one of a limited number of manually assigned promotional slots to reach its natural audience.
What to learn/watch: this makes conversion signals, tags, audience overlap, wishlists and price elasticity even more strategically important. Store-page optimisation is increasingly less about appealing to everybody and more about giving Steam enough information to identify exactly who should see the game.
Industry implication: Steam is gradually becoming less like a digital shopfront and more like a recommendation engine with a store attached. For indies in particular, understanding the discovery algorithm is becoming part of publishing strategy.
dig deeper
EU regulators have moved from discussing virtual currencies to coordinated enforcement
On September 30, the EU’s Consumer Protection Cooperation Network launched coordinated actions involving nine publishers over virtual-currency practices in games including Candy Crush Saga, Clash of Clans, Gardenscapes, Minecraft, Valorant and Forge of Empires. A parallel action involving Activision Blizzard covers Call of Duty Mobile and Diablo Immortal.
The issues include displaying real-money prices, avoiding currency bundles that force consumers to buy more currency than needed, withdrawal rights, understandable terms and protections for children. Industry groups Video Games Europe and the EGDF dispute parts of the CPC’s legal interpretation, while proposing their own measures including real-money pricing and refunds for unused currency bundles.
Why it matters: this goes directly at some deeply embedded F2P design conventions. Gems and coins don’t merely facilitate transactions; they deliberately separate the emotional experience of spending from actual currency.
What to watch: real-money-equivalent pricing beside virtual-currency prices. If that becomes standard across the EU, teams will finally get meaningful data on whether currency abstraction itself materially affects conversion and ARPPU.
Industry implication: monetisation UX is becoming a regulatory discipline. Casual and mobile teams should increasingly involve legal/compliance expertise during economy design, rather than reviewing systems shortly before launch.
dig deeper
Casual watch: web games are considerably larger than the industry tends to assume
A new Newzoo/Google study estimates that 46% of online consumers aged 13–55 play HTML5/browser games monthly, rising to around 60% among 13–30-year-olds. Eighty-three percent play those games on smartphones, while three-quarters of those players also play conventional mobile apps.
The discovery data is particularly interesting: only about 23% go directly to a game portal, while 45% begin with web search. More than half say they’d simply switch websites if their preferred game disappeared from a portal. Sites containing H5 games also recorded roughly 40% longer visit times.
Why it matters: browser games aren’t necessarily competing with app games. They’re another surface through which the same mobile audience encounters games. This fits remarkably well with YouTube Playables, instant-play experiments and social platforms adding playable content. For casual publishers, we think no-install distribution deserves much more serious attention again.
What to learn: web-game SEO may be considerably more important than conventional portal loyalty. The product unit increasingly isn’t “the portal”; it’s the individual game discoverable from Google, YouTube, social feeds or embedded contexts. For casual publishers, we think no-install distribution deserves much more serious attention again.
dig deeper
Design watch: Control Resonant deliberately rejects yellow paint and replaces it with systemic orientation
Control Resonant director Mikael Kasurinen has explained one of the game’s more interesting level-design decisions: Remedy deliberately avoided the ubiquitous coloured ledges and markers used to tell players where traversal is possible.
That’s particularly risky because Resonant’s Manhattan allows players to run, fly and manipulate gravity through spaces where conventional ideas of “up” and “down” don’t necessarily apply.
Remedy instead uses environmental systems—including consistent gravity cues affecting rain and wind—to help players orient themselves. Kasurinen’s argument is that overly explicit signposting converts exploration from “Where should I go?” into “The designer told me where to go.”
Why it matters: the interesting lesson isn’t “yellow paint bad.” It’s that affordance readability should follow the intended player fantasy. If the fantasy is discovery, removing uncertainty can destroy part of the game.
What we’d study: how often players actually become lost versus how often the ambiguity creates genuine exploration. Resonant is a useful contemporary case study in environmental readability without explicit UI-like signposting.
dig deeper
AI-production watch: The9 says it built a commercially launched mobile game in eight months instead of 2–3 years
The9 launched The Greed of Man on September 24. The company says the title reached No. 1 in Hong Kong’s simulation-game category on both iOS and Google Play. More interestingly, CEO George Lai says development took roughly eight months, versus an estimated two to three years under its traditional production model.
The game was built using The9’s proprietary the9bit system, which orchestrates several AI models across content generation, debugging, asset creation and integration, while allowing creators to alter gameplay and visuals through natural-language instructions. The chart result is company-reported and category-specific, so it shouldn’t be mistaken for evidence of major commercial success yet. But the production claim is much more interesting.
Why it matters: this is closer to an AI-native production pipeline than “our artists use Midjourney” or “programmers use Copilot.”
Industry implication: the metric worth tracking isn’t how many assets AI generates. It’s time from prototype to market validation. If teams can test commercially credible games in eight months instead of two years, portfolio economics change dramatically even if AI never replaces entire disciplines.
dig deeper
Casual design watch: hybrid-casual puzzle design is converging around “dismantling” rather than matching
There is a useful little design signal buried in this week’s mobile data. ABI Games’ Colony Flow, released in June, has already generated an estimated $11 million IAP from 7.6 million installs. Its core mechanic borrows from Pixel Flow: players gradually dismantle an image, but Colony Flow replaces conveyor belts with ants. It recently ranked fourth among top-grossing hybrid-casual puzzle titles. At the same time, Circle Games, whose Sort Express has generated an estimated $5.8 million, just raised $25 million led by Tencent.
There’s a pattern here beyond another sorting-game cycle. Traditional Match-3 creates satisfaction through:
identify combination → trigger removal → cascade.
A growing group of hybrid-casual puzzles instead uses:
inspect object → identify obstruction → progressively dismantle complex visual object.
That produces an immediately readable before/after state, which works exceptionally well in playable/video advertising.
What we’d watch: how developers layer long-term progression onto these extremely tactile cores without destroying their simplicity. That’s currently one of the most interesting problems in casual design.
dig deeper
AppLovin vs. Unity has turned mobile ad-tech competition into a fight over who owns auction data
AppLovin has sued Unity, alleging that Unity’s free Ad Quality SDK collects information from auctions mediated by AppLovin MAX—including clearing prices, impression IDs and waterfall data—and uses that information competitively. Unity rejects the allegation and argues that customers voluntarily share the data to help its system identify harmful ads.
AppLovin is seeking restrictions on Unity collecting or using data tied to auctions AppLovin wins. Unity had reportedly offered to stop certain future collection from MAX-mediated auctions, but AppLovin rejected that because it did not address previously collected data or models potentially trained on it
Why it matters: mobile advertising is increasingly an optimization problem solved by machine-learning systems trained on enormous datasets. Consequently, auction data itself becomes strategic IP.
This isn’t merely two ad-tech companies fighting. The outcome could influence what information mediation platforms, SDKs and ad-quality tools are allowed to observe about one another.
Industry implication: publishers should pay much more attention to the data flows inside their monetisation stack. “Free SDK” increasingly means exchanging something other than money.
dig deeper
Mario Kart Tour’s shutdown is a useful post-mortem on what enormous reach is actually worth
Mario Kart Tour shut down on September 30 after seven years. AppMagic estimates put it at 276+ million lifetime downloads and $268.5 million in IAP revenue. The shape of that revenue curve is more interesting: IAP peaked at almost $22 million in October 2019, fell to around $5-6 million monthly during much of 2020–21, then continued declining. By the time Nintendo announced the closure, the game was generating around $650,000 per month. Nintendo removed its gacha mechanics in late 2022, after which monetisation declined further.
Why it matters: 276 million installs sounds extraordinary. But that works out to less than a dollar of estimated lifetime IAP revenue per install. Compare that with modern mobile successes built around sophisticated retention, segmentation and live-ops systems, and the difference between audience scale and audience value becomes obvious.
What to learn: famous IP can buy astonishing acquisition efficiency. It doesn’t automatically solve long-term monetisation design.
dig deeper
Hidden-gem/design watch: Tokyo Game Show’s strangest games point toward mundane-task games becoming a real design space
One of the more revealing indie signals coming out of TGS wasn’t another roguelite. Former PlayStation indie chief Shuhei Yoshida highlighted games including Finding Polka, Haunted Streamer and particularly Hair Removal Salon Simulator, a game about removing customers’ body hair using increasingly absurd tools. Yoshida explicitly compared its satisfaction loop to PowerWash Simulator.
We’re seeing an increasingly established design family around:
recognisable mundane task → highly legible progress → tactile interaction → visual cleanup/transformation → low failure pressure.
PowerWash Simulator, unpacking games, cleaning games, repair games and now even ASMR/unboxing mobile titles all exploit versions of this structure. The important ingredient isn’t “simulation.” It’s visible state change. That’s extremely powerful both psychologically and for streaming/social-media discovery.
Players can immediately see:
messy → clean
full → empty
broken → repaired
covered → uncovered.
Design implication: We’d expect considerably more experimentation around everyday physical processes that games historically considered too mundane to model.
dig deeper
Nerial’s closure is a reminder that critical distinction isn’t enough to sustain a modern indie studio
Nerial, the studio behind Reigns, Card Shark and The Crush House, announced on October 2 that it is entering “hibernation mode,” ending development and laying off its remaining staff after 13 years and 11 published games. Parent company Devolver Digital acquired the studio in 2021; Nerial had already undergone substantial layoffs in 2024.
This one matters precisely because Nerial wasn’t producing anonymous games. Reigns created an instantly recognisable interaction model. Tinder-like left/right decisions applied to narrative systems and Card Shark and The Crush House were similarly unusual.
Why it matters: one of the industry’s uncomfortable current realities is that creative differentiation and critical recognition aren’t necessarily enough to support a permanent studio organisation. A successful indie IP can finance projects, but maintaining salaries and infrastructure continuously between unpredictable releases remains difficult.
Industry implication: expect more acclaimed indie developers to behave less like permanent mini-publishers and more like project-based creative collectives. Scaling teams up around funded productions and shrinking between them. For an industry currently arguing that the answer to AAA inflation is “just make smaller, more distinctive games,” Nerial is an important reminder that smaller doesn’t automatically mean economically safe.
dig deeper
