The Six-Year Development of Pokémon Red & Blue

In 1990, an obscure Japanese magazine writer named Satoshi Tajiri pitched Nintendo on a game where children would catch insects, trade them between handhelds, and battle them. He had no programming team capable of shipping it. He had no money. He had a dying console (the Game Boy was being declared obsolete by every analyst that year) as his target platform. Five of his six employees would quit during development, and one of them attempted suicide because of the project's stress. The game shipped in 1996 and sold 31.05 million copies. The franchise it started has, as of 2026, generated over $90 billion in revenue — making it the highest-grossing entertainment franchise in human history. This is how Pokémon Red & Blue actually got made.

Satoshi Tajiri's bug-collecting childhood

Tajiri was born in Machida, Tokyo in 1965. The neighborhood, when he was a child, was semi-rural — rice paddies, ponds, undeveloped lots. Tajiri's hobby, like many Japanese boys of the era, was insect collecting. He spent so much of his childhood catching beetles, dragonflies and tadpoles that his neighbors gave him the nickname "Dr. Bug."

By the late 1970s, Machida was being paved over for housing developments. The ponds were filled in. The empty lots became parking lots. By the time Tajiri was a teenager, there were no insects left to catch. This loss — the disappearance of his entire childhood hobby — is the foundational emotional fact of the entire Pokémon franchise. Multiple interviews across decades have him saying the same thing: I wanted to give kids something to collect, when their world doesn't have insects anymore.

Game Freak: a magazine, not a studio

In 1981, Tajiri started a fanzine called Game Freak covering arcade game strategy. The zine was run out of his bedroom and consisted of him and his friends writing tip sheets for arcade games like Xevious. One of the people who replied to a fan letter was Ken Sugimori, a freelance illustrator who would later become the lead designer of every Pokémon you've ever seen.

By 1989, Tajiri's writing convinced him that there was a gap in the market that could only be filled by people who actually played games. He convinced Sugimori, programmer Junichi Masuda, and a handful of others to convert Game Freak the magazine into Game Freak the development studio. They had no industry experience. They had no money. They got a contract to develop a Famicom puzzle game called Mendel Palace (1989). It made just enough money to fund their next project.

The link cable epiphany (1990)

The Game Boy launched in April 1989. By early 1990, Tajiri was carrying a Game Boy everywhere. He noticed the link cable port — a feature Nintendo included for two-player games like the bundled Tetris. Tajiri's specific mental image, repeated in every interview, was watching the link cable physically connect two Game Boys and imagining "an insect crawling along the cable from one device to the other."

"I imagined the bug actually moving through the cable. I knew immediately we had to make a game about it."

The original concept was called Capsule Monsters (Kapuseru Monsutaa) — pocket-sized creatures stored in capsules, traded over link cable, fought in arena battles. The pitch document Tajiri brought to Nintendo in early 1991 was a single page. He had no script for how to build it. The first version of the design even had every Pokémon being trapped not in Poké Balls but in tiny glass capsules — the "ball" came later when Nintendo's legal team flagged trademark issues with another company's "capsule" product.

Shigeru Miyamoto says yes

Most accounts of Pokémon's history emphasize Miyamoto's mentorship. The reality is more complicated. Miyamoto saw the original pitch and was skeptical — he later said in interviews that he didn't fully understand it but liked Tajiri's energy. The actual decision to greenlight Capsule Monsters was made by then-Nintendo president Hiroshi Yamauchi, who reportedly told Miyamoto: "Tajiri reminds me of you. Help him." Miyamoto became the project's official supervisor and his core contribution was practical: he forced Game Freak to ship two versions of the same game.

Miyamoto's reasoning, recorded in Iwata Asks interviews decades later, was simple. Trading was the game's central mechanic. If only one version existed, kids could play the game with strangers but not with friends — strangers would have nothing different. Two versions, with version-exclusive monsters, would force friends to talk to each other. Trade among siblings, trade at school, trade at the arcade. The two-version design was Miyamoto's single most important contribution to Pokémon and it shaped every Pokémon main game since.

The financial crisis (1991-1995)

Game Freak signed the Pokémon contract in 1991. The original delivery date was 1993. They missed it by three years. Multiple things went wrong:

First, the team was too small. At peak, the entire Pokémon Red & Blue development team was eight people — including Tajiri (director), Sugimori (lead artist), Masuda (programmer/composer), and one of the most underappreciated figures in game history, programmer Sōta Fukushi, who wrote most of the actual battle engine. By comparison, Super Mario World on the SNES had a team of about 16 at peak.

Second, Game Freak had no money. They were funded by Nintendo's milestone payments, which arrived only when deliverables shipped. When deliverables slipped (and they always slipped), Game Freak's bank account ran out. Tajiri later said in Time magazine that he sometimes ate one meal a day during the worst of the development.

Five of Game Freak's eight original employees quit between 1992 and 1995, citing burnout. Tajiri replaced them slowly. In 1995, with the game still unfinished, Game Freak was three months from bankruptcy. Nintendo would not give them an advance against unshipped milestones. The studio was saved only when Creatures Inc., a company founded specifically for the Pokémon project, agreed to share investment in exchange for shared IP rights — which is why every Pokémon game still has a "© Game Freak / Creatures / Nintendo" copyright triplet on the title screen.

The Game Boy was supposedly dead

The single most-cited business decision in Pokémon's history is that Game Freak shipped Pokémon on a "dead" platform. By 1995, the Game Boy had been on the market for six years. The Game Gear and Atari Lynx had failed to dethrone it. The handheld market was, to outside analysts, fully saturated. Nintendo was internally debating whether to even produce more Game Boy units.

Tajiri, who had pitched the project on Game Boy in 1991 and never wavered, refused to migrate to color. The reasoning he later gave was specifically about the link cable: he wanted as many kids as possible to have the device, and the Game Boy had a Japanese installed base of about 35 million units in 1995. Migrating to a newer platform would have meant fewer trades possible. The "dead" platform was, in fact, the most-used handheld in human history at the moment Tajiri shipped his game.

February 27, 1996: Pokémon Red & Green

The original Japanese release was Pokémon Akai (Red) and Pokémon Midori (Green). The blue version came later — it was a re-release in October 1996 with improved sprites, used as the basis for the international "Pokémon Blue" most Western players know.

Initial Japanese sales were soft. The first month moved about 200,000 copies — respectable for a Game Boy game in 1996, but not a hit. What changed everything was an obscure feature called Mew.

Mew was the 151st Pokémon in the game. It wasn't catchable through normal play. Programmer Shigeki Morimoto had inserted it during the final debugging pass, in code space that became free when other features were cut. He told no one outside the dev team. After release, it was discovered by glitch hunters and word spread through Japanese playgrounds. Then, in April 1996, Nintendo Japan's monthly magazine CoroCoro Comic announced a contest: 20 readers would receive Mew distributed via mail-in event cartridges. They got 78,000 entries.

The 78,000 entries kicked off the cultural snowball. Sales went from 200,000 in February to 1.5 million by Christmas. By end of 1997, the Japanese versions had sold 6 million copies. By end of 1998 (when the international versions launched), 10+ million.

MissingNo. and the glitches that became legend

The development crunch left bugs in the shipped game. Some of the most famous glitches in video game history are direct artifacts of Game Freak's tiny budget:

MissingNo. ("Missing Number") was the result of the game generating Pokémon based on memory location. When the game's encounter table read invalid data, it pulled from RAM and assembled a creature out of garbage. MissingNo. is technically every Pokémon that doesn't fit into the legitimate 151. Players discovered that catching MissingNo. duplicated whatever item was in the 6th slot of your bag — turning Rare Candies and Master Balls into infinite stacks. Game Freak never patched it because the Game Boy had no patching mechanism. It became a cultural fixture.

The Mew Glitch let players catch Mew without the official event by manipulating a stat-modifying battle in the Pokémon Tower. This worked because of how the game stored battle pointers in memory — a side effect of crunching the entire battle engine into 8 KB of RAM.

The Truck on Cinnabar Coast: a single inaccessible truck sprite in a corner of the map that triggered an entire generation of "the truck has Mew under it" conspiracy theories. The truck was placeholder art that Game Freak forgot to remove before ship. There is nothing under it. It was just a missed cleanup commit on a project that ran out of QA time.

The international release and the anime

Nintendo of America was, internally, opposed to Pokémon. The marketing team's read was that American kids would not be interested in collecting Japanese cartoon animals. The president of Nintendo of America at the time, Howard Lincoln, reportedly made the localization team focus-test the game with American children for six months. The kids loved it. Nintendo greenlit the localization but allocated almost no marketing budget.

Two things saved Pokémon's American launch. First, the Pokémon anime had been a hit on Japanese TV starting April 1997 — the same anime episode that famously caused 685 children to have epileptic seizures from a strobing scene in December 1997. Warner Bros. licensed the dub for syndication starting September 1998. Second, the trading card game (developed by Wizards of the Coast in the US) launched in January 1999 and became the must-have toy of that Christmas.

The Pokémon Red & Blue Game Boy carts shipped in North America on September 28, 1998. By Christmas, they were the top-selling games in the US. By spring 1999, parents were lining up around blocks for trading cards. By 2000, Pokémon was a $5 billion-a-year franchise. Game Freak's eight-person team in a Tokyo apartment had built the largest entertainment property in history.

Why the original game still works

Played in 2026, on a Game Boy emulator, Pokémon Red & Blue is rough. The sprites are ugly. The animations don't exist. The battle music repeats. Most importantly, the type chart is broken — Psychic-type Pokémon are absurdly overpowered because the game only checks for "is the move super-effective" without applying the Psychic vs. Bug weakness correctly (a bug fixed in Gold/Silver). Speedrunners can finish the entire game in under two hours.

And yet. The core loop — explore, encounter, capture, evolve, trade — works. The 151 creature designs, almost all sketched by Sugimori in 1992-1995, hold up. Charizard, Blastoise, Venusaur, Pikachu, Eevee, Dragonite, Mewtwo, Mew. None of them have been visually replaced in 30 years. Nintendo can't replace them because every kid who ever owned the original Red or Blue cart has them imprinted as the foundational images of "monster." That permanence is, in a sense, what's kept the franchise alive: every generation since has been a remix of the iconography Game Freak's eight-person team locked in during 1992-1995.

The legacy

Tajiri stepped back from active directing duties after Pokémon Diamond & Pearl (2006). He's still listed as Executive Director on every mainline Pokémon game and remains Game Freak's president. Sugimori still leads art direction. Masuda directed the games until Sword & Shield (2019) and now serves as Creative Fellow.

The eight people who shipped Pokémon Red & Blue in 1996 are now individually wealthy beyond what most game developers ever achieve. The studio that almost went bankrupt is one of the most valuable independent game studios in the world. The "dead" platform Tajiri shipped on became, partly because of Pokémon, the best-selling handheld console family in history.

The single most-quoted line from Tajiri, in every interview about how Pokémon happened: "If a kid in 2026 catches a Pokémon and feels what I felt catching beetles in 1975, then everything was worth it." The kids do. Tens of millions of them, every day, on phones their parents carry to work. The bug Tajiri saw crawling along the link cable is still moving.

Play the original games right now in your browser, no downloads needed: Pokémon Red, Pokémon Blue, Pokémon Yellow, or jump to the Game Boy generation that followed: Pokémon Gold, Pokémon Silver, Pokémon Crystal.