Eight million copies sold. Nine Game Awards, including Game of the Year: a record for a single title in the ceremony’s history. A 93 on Metacritic. A soundtrack with 617 million streams. A live-action film already in development. Hideo Kojima calling the development team his ideal size. All of this from a studio that had never shipped a game before.
I put a few hours into Expedition 33 last month, squeezed in between revision sessions, which is probably the worst possible way to experience a game that rewards patience and attention. It pulled me in anyway. That is the thing about it: it does not need ideal conditions. It just works.
Clair Obscur: Expedition 33 launched on April 24, 2025. It hit one million sales in three days, 3.3 million in 33 days, and eight million within its first year. Notable figures for any game, but especially notable for an original IP in a genre that legacy franchises like Persona and Pokémon have historically dominated. It was the most downloaded new third-party title on Xbox Game Pass in 2025, and those Game Pass players did not count toward the sales figure. It was the highest-rated game of 2025 on every major review aggregator. It won the Golden Joystick Ultimate Game of the Year. It is on the BAFTA 2026 Game of the Year longlist.
None of this was supposed to happen. Not because the game was not good, it clearly was, but because the specific combination of factors that produced it broke almost every conventional assumption about what succeeds in this industry. And most of the coverage explaining its success got the explanation slightly wrong.
What Sandfall Interactive actually is
Sandfall Interactive is a French studio based in Montpellier. Expedition 33 was their first game. Their core team was approximately 30 people, many with little or no professional game development experience before joining. Creative director Guillaume Broche built the prototype alone before forming the studio around it.
There is a version of this story where the point is the underdog angle: scrappy first-timers outpunching a bloated industry. That version is not wrong. It is just not the useful version. The more useful version is about what the constraints produced rather than what they overcame.
A 30-person studio cannot make a game by committee. There are not enough people for the layers of approval, brand safety review, market testing, and executive sign-off that characterize large studio development. Every creative decision is visible to everyone involved. The vision either holds together or it does not, and there is nowhere to hide if it does not. What Expedition 33 has, a coherent world, a specific tone, a story that knows exactly what it wants to say, is a direct consequence of that constraint. It did not survive a committee process because it never went through one.
The things it should not have been able to do
Turn-based combat. In 2025. The dominant industry narrative for roughly a decade has been that turn-based RPGs are a dying format: that modern players lack the patience, that the action RPG had permanently replaced the traditional formula. Expedition 33 launched the same week as Oblivion Remastered and sold 3.3 million copies in 33 days. Oblivion Remastered is an action RPG. Expedition 33 is turn-based with real-time parry mechanics layered on top. Both sold exceptionally well. The turn-based-is-dead narrative did not survive contact with a game that executed the format brilliantly.
Non-English development. The gaming industry’s dominant creative centers are American, Japanese, and British. Expedition 33 comes from France. Its world is built from Belle Époque aesthetics: the painterly visual style, the mythology, the specific gravity of French romanticism. Sandfall did not file down its Frenchness to appeal to a global audience. It committed to it completely. The result felt unlike anything else available, precisely because it drew from a different cultural source than most of its competition.
A debut game at AAA production values. The voice cast includes Ben Starr, Jennifer English, Charlie Cox, and Andy Serkis. The Unreal Engine 5 visuals compete with studios ten times the size. The soundtrack by Lorien Testard topped the Billboard Classical Albums chart, sold out a live concert tour, and accumulated 617 million streams, rising to 645 million when including additional EPs. Sandfall extended their core team’s capability through smart outsourcing: animation studios, QA partners, localization teams, without surrendering creative control. The result looks, sounds, and feels like a game built by people who had been doing this for twenty years. It was built by people doing it for the first time.
What Kojima understood that most coverage missed
At Summer Game Fest 2025, Geoff Keighley described Expedition 33 as “a monumental achievement from a team of under 30 developers.” Hideo Kojima later said Sandfall’s size, “33 team members and a dog”, was his ideal. Both statements point at something real without quite naming it.
The point is not the headcount. It is where creative authority sits.
In a 30-person studio, the creative director can see every part of the game. Broche described the philosophy as wanting to make “stories like an art house, crazy projects because we want to make them.” That is not a mission statement on a wall. It is a description of the actual working condition. At 300 people, an art house philosophy becomes an aspiration in a slide deck. At 30, it is operational reality.
The reason Expedition 33 feels the way it does, specific, committed, uncompromised, is inseparable from the structure that made it. Most coverage treated the team size as a trivia point. It is closer to an explanation.
The industry context that makes it matter
2025 was not a kind year for large-studio game development. Layoffs continued across the industry. Several major AAA releases underperformed commercially despite enormous budgets. The structural economics of large-studio games have become genuinely precarious: the gap between what they cost to make and what they need to sell to break even keeps widening, which makes creative risk-taking harder to justify with every passing year.
Into that environment, Sandfall released a turn-based RPG set in a French Romantic fantasy world, priced it at $49.99, and sold eight million copies. The game cost a fraction of what a comparable large-studio project would require. Its break-even point was achievable. Its creative vision was intact because nobody with a quarterly revenue target had the authority to dilute it.
Sandfall’s response to their success is the most instructive part. When asked about expansion plans, lead programmer Tom Guillermin said he preferred staying small. Broche said the studio wants to “remain very small and agile with a strong creative vision.” They are not using Expedition 33’s success to become the kind of studio that Expedition 33 succeeded against. That is either principled stubbornness or genuine wisdom. The evidence so far points firmly toward the latter.
What it actually means
Expedition 33 is not proof that any small team can do this. The specific combination of talent, timing, vision, and execution that produced it is not a formula. It will not be cleanly replicated. Broche has said himself that the game is “the first in a new wave” of small teams punching above their weight, not a template for guaranteed success.
What it does prove is narrower and more useful: a game does not need a massive budget, a large team, or an established franchise to compete at the highest level of the industry. The creative risks that large studios avoid, unusual cultural settings, turn-based mechanics in a post-Soulsborne market, stories about grief and mortality without ironic distance, are available to smaller teams precisely because smaller teams are not institutionally pressured to avoid them.
The most important thing about Clair Obscur: Expedition 33 is not the nine awards or the eight million copies. It is that this game could not have been made by a large studio. The specific qualities that make it exceptional required a structure where compromise was difficult. Sandfall did not succeed despite being small. They succeeded in part because of it.
The industry will spend years arguing about what that means. Most of those arguments will focus on the wrong things: budget ceilings, team size targets, pipeline efficiency. The actual lesson is about creative authority, who holds it, how many people share it, and what happens to a game when the answer to both questions is as few as possible.
For more on why this pattern keeps repeating, the piece on why small teams keep making the best games covers the structural reasons, and the sustainability reality that success stories like this one tend to obscure. The question of how AI tools might change what small studios can build is directly relevant here too: the AI in game development piece covers both the opportunity and the risk. And notably, the accessibility choices Sandfall made with Expedition 33 are worth examining alongside the accessibility features piece, which covers what good looks like across the industry.
