In early 1992, the project stopped. Game Freak had been working on a concept called Capsule Monsters for Nintendo since 1990, but the money ran out. The team made Yoshi — a block-matching puzzle game — in six months to keep the company alive. The larger project went on hold. No one knew when it would start again.
Game Freak had not begun as a game company. In March 1983, a sixteen-year-old named Satoshi Tajiri published the first issue of a handwritten fanzine about arcade games. He called it Game Freak, after the name he used on high-score boards. The magazine covered titles that mainstream journalism ignored. An illustrator named Ken Sugimori found the fanzine and offered to contribute art from the ninth issue onward. They sold about a thousand copies per issue through game shops and by hand.
On April 26, 1989, Tajiri and Sugimori incorporated Game Freak Inc. in Tokyo. Junichi Masuda, recruited by Tajiri the same year as a composer, became the third founding member. The studio's first work was Quinty — a puzzle game published by Namco in 1989, known in North America as Mendel Palace. It took Tajiri two years to learn programming and another year to build the game. It was competent work. It proved the studio could deliver. It did not point toward anything larger.
The next projects followed the same pattern. Yoshi in 1991, Mario & Wario in 1993, both for Nintendo. Small puzzle and action titles that paid the bills. The company rarely exceeded a dozen people. The financial runway was short. There was no room for a long bet.
In 1990, Tajiri pitched a different kind of game to Shigeru Miyamoto and Nintendo's planning department. He called it Capsule Monsters. The core idea was simple: the Game Boy's link cable was not just for multiplayer — it was a tunnel through which creatures could travel between two worlds. You caught creatures in the wild, trained them, and traded them with a friend through the cable. The social structure of childhood insect-collecting, turned into a portable game. Nintendo agreed to fund development.
What followed was six years that came close, on multiple occasions, to ending the project entirely. Staff turnover was significant. The core remained Tajiri, Sugimori, and Masuda, with a rotating cast of contributors around them. The budget was covered mainly by payments from Nintendo, but what remained for salaries was minimal. There were months when Tajiri went without pay. The mechanics were reconceived multiple times as the Game Boy's technical limits were tested. By the time a working version existed, the original Game Boy was nearing the end of its commercial life, and Nintendo's marketing department was not enthusiastic about the odds.
Sugimori designed the original 151 creatures. He worked almost entirely alone, producing several designs per week across multiple years. Each design had to function as both a believable animal and a character a child would want. The Game Boy's tiny monochrome screen forced a clarity of silhouette that would define the series for decades. There was no second draft. There was no budget for one.
On February 27, 1996, Pokémon Red Version and Green Version were released in Japan. The initial print run was modest. The game found its audience slowly — through playgrounds, through the link cable trades the design had been built to activate, through word of mouth. By the end of 1996, the two versions had sold millions of copies in Japan. An anime series launched in April 1997. The international release came in September 1998 under the names Red and Blue, selling approximately ten million copies in North America alone by year's end.
By measured revenue, the Pokémon franchise became the highest-grossing media franchise in recorded history. Combined merchandise, game, film, and licensing revenue has been estimated at over 150 billion USD as of 2022. Game Freak, the studio that built it, has employed between 100 and 200 people for most of the franchise's life — roughly the headcount of a mid-sized accounting firm. For six years, it was three people and a rotating door.
The chapter between 1990 and 1996 is not the one that gets told. The story people know begins with the release, with the phenomenon, with the numbers that kept growing. But before any of that, there were six years when the project nearly died multiple times, when there was no salary, when the only certainty was that the work was not yet done. The question the story asks is not whether you can build something worth the wait. The question is: how long can you wait when you do not know if it will be worth it?