When Building Is No Longer the Hard Part
Notes from Gamescom, Gamesforum London and Antler’s European Founders Conference
I spent the last few weeks at Gamescom in Cologne, Gamesforum in London and Antler’s European Founders Conference. Across games and startups, the pattern was remarkably similar: smaller teams are building more, testing more and getting to market with less capital.
Most of the conversation around AI focuses on that productivity gain. But when the economics of building change, everything around them starts to move too. The bar for what counts as traction rises. Ideas that were previously too expensive to pursue become viable. Old advantages, including where things are built, begin to shift. And some of the signals investors and publishers have relied on start to mean something different.
More founders, fewer checks
Antler put some striking numbers behind this. Since 2021, the number of new founders in Europe has increased 54%, while pre-seed deals have fallen 38%, Seed deals 41% and Series A deals 45%. The number of institutional investors actively deploying early-stage capital has fallen 42% since 2022.
AI hasn’t created that funding contraction, but it is contributing to the other side of the equation. The tools for starting companies are becoming dramatically more accessible, allowing more founders to get further with smaller teams and less money. At the same time, competition for the capital to scale those companies is getting tougher.
That changes expectations too. One founder at Antler described how a few years ago, 3–4x growth could make a company stand out, while today that can increasingly feel like the starting point. Antler’s data shows the same pressure further down the funnel. Historically, around 23% of European startups that raised a Seed round went on to raise a Series A. For the 2023 cohort, that figure had fallen to 9%.
The U.S. isn’t following exactly the same trajectory, and it would be wrong to extrapolate Europe’s funding contraction globally. But the broader pressure looks familiar. Carta’s data shows that the path from Seed to Series A has become substantially harder than it was for earlier cohorts, while PitchBook and NVCA data shows capital increasingly concentrating into a small number of enormous rounds, particularly in AI. In 2025, half of all U.S. venture deal value went into just 0.05% of completed deals, while AI accounted for nearly two-thirds of total deal value.
Different markets, different dynamics, but something similar is happening underneath. Technology is making it easier to start and build at the same time the market is raising its expectations for what deserves to scale.
What this looks like in games
At Gamescom and Gamesforum, one thing was clear: building games is getting cheaper, testing ideas is getting faster, and reaching players is becoming easier. Small teams can build more with fewer people, test far more ideas before committing significant resources, and reach players across more platforms. AI is accelerating development and content production, while also making it faster and cheaper to create and test ads, produce playable ads, optimize UA and run creator campaigns.
None of that changes the basic playbook. Mobile studios have been testing concepts, creatives and audiences before committing significant development resources for years. What’s changing is the speed, cost and volume at which they can do it. A team that could previously afford to test a handful of ideas can increasingly test many more, learn faster and kill weak ideas before spending heavily on them.
Distribution is opening up too. HTML5 and web games are seeing renewed interest, driven partly by lower costs and the ability to build once and deploy across multiple platforms. Poki, CrazyGames and Playgama came up repeatedly, with particularly positive feedback from developers I spoke to who had launched on CrazyGames and Playgama. TikTok Mini Games and YouTube Playables had real momentum, with several developers particularly bullish on YouTube.
Chinese studios are especially interesting here. Many are bringing mature WeChat and Douyin mini-game playbooks West, with small teams testing lots of concepts, learning quickly and putting serious resources behind the ones that work. Again, the playbook itself isn’t new. What’s changing is how accessible it’s becoming and how many channels teams can now test against.
This is what makes the current moment in games fascinating. A small team can build more games, test more ideas and reach more players without a corresponding increase in people or capital. That’s a huge advantage, but it also means there are going to be a lot more things competing for the same players.
The rules are changing
When technology makes something dramatically cheaper, it doesn’t necessarily make the market easier. It changes where the constraint sits. Ten startups suddenly able to accomplish what previously required much larger teams doesn’t create more venture capital. Game developers able to test five times as many concepts don’t create five times as much player attention.
This is where I think conversations about AI that stop at productivity miss something important. A developer coding faster or a founder building a company with fewer people can create an enormous advantage. But competitive advantages are relative. As those capabilities spread, the market adjusts.
The productivity gain hasn’t disappeared. It’s being absorbed into higher expectations.
That means some of the benchmarks we’ve used to judge progress are going to have to change. For founders, the bar for traction is rising as smaller teams are expected to get further, faster and with less capital. For game developers, the shift creates opportunity as much as pressure. As games become cheaper to build, ideas faster to test and new distribution channels open up, the economics of what can work begin to change. Concepts, audiences and platforms that might not have justified the investment before can suddenly become viable. And as those economics continue to shift, there will be new waves of opportunity around things that simply weren’t practical to build before.
The same is true on the other side of the table. Investors and publishers will have to recalibrate what good looks like. Team size, development speed and even the quality of an early product tell you something different when the economics of producing that product have changed. The more capability technology gives us, the more important it becomes to judge results relative to what is now possible.
Even geographic arbitrage is shifting. Lower-cost development markets have long offered a significant economic advantage, but AI is changing the amount of labor required to build and test games in the first place. At the same time, teams in those markets have access to the same leverage. Geography will still matter, but where the advantage lies may look very different in a few years.
AI doesn’t just change what teams can accomplish. It changes the economics around them. The benchmarks move, old advantages shift and things that didn’t make sense before become viable. The interesting part of the next few years won’t just be watching teams do more with less. It will be seeing what becomes possible because they can.
Written by Rudy Koch, co-founder & CEO of JOA and co-founder of Mythical Games. More: Homepage · Writing · The GreenPark Series · Media.