He Built It
on His Lunch Breaks.
Players Spread It
Without Being Asked.

Every issue of The Real How follows the same structure. The Setup. The Constraint. The Gap. The Playbook. The Traction Framework. The Mistake. What You Can Apply. And your move for the week.

Most people see $2.5 billion and assume there was a grand strategy. A VC-backed roadmap. A marketing machine. A team of growth hackers running A/B tests.

There was none of that. There was a programmer building something on his lunch breaks because it interested him, posting it to a forum of 40 people who understood it before it had any polish, and then watching what communities do when you give them something genuinely worth caring about.

That is a story worth understanding in full. Because the mechanism that made Minecraft the best-selling video game in human history is available to you right now, in whatever category you are already building in. Let's get into it.

The Setup: Five Years Building Someone Else's Games

It is 2005. Markus Persson starts working at King in Stockholm. Flash games, one every couple of months. Builds it, hands it over, moves to the next brief. King will later become the studio behind Candy Crush and list on the New York Stock Exchange for billions. Right now it is just a small office and Persson is the guy who finishes on time and goes home.

He does not hate it. He is good at it. But here is what nobody tells you about people who learned to code at seven and built their first game at eight: the instinct does not disappear when you clock in for someone else. It just waits.

He starts feeding it quietly. Building small things on the side. Management notices. Tensions build. In 2009 he moves to jAlbum, a smaller photo sharing company in Stockholm. The work is lighter. Nobody there cares what he builds after hours.

He has not quit to start a company. He has moved sideways to buy himself the one thing that mattered more than anything at that stage: time, and the freedom to use it honestly. That move — not dramatic, not a leap of faith, just lateral — is available to most people reading this right now.

In 2026, the tools available to you compress the timeline even further. AI-assisted development can take you from idea to working prototype in days. What took Persson a weekend might now take you an evening. The time advantage has never been greater for a solo builder with the drive to use it.

The Constraint: No Budget. No Team. No Plan. Just What He Already Knew.

He was not building Minecraft with a team or a budget. He was building it between shifts. During lunch. In the evenings. On weekends. Alone. In Java, because that was the language he already knew — not because it was optimal, because it was what he had.

Look at what each constraint forced him to do. He could not spend six months polishing, so he shipped fast. He could not run a QA team, so he released early and let real players find the problems. He could not afford marketing, so he had to build something so genuinely interesting that people would tell each other about it without being asked.

The constraints did not limit the product. They shaped it into exactly what the market wanted.

Most builders wait until conditions are ideal before starting. Better tools. More time. A co-founder. A savings buffer. Persson built with the worst possible conditions — alone, between shifts, in a language that was not optimal — and those conditions became his advantages. He shipped faster than anyone polishing. He got real feedback faster than anyone testing in private. He built distribution into the product because he could not afford to run ads.

Your constraints right now are not what is stopping you. They are the curriculum. Sound familiar?

Finding the Gap: Two Half-Products and One Clear Question

Here is how the idea actually happened. This is the part that matters most for anyone trying to find something worth building, and it is the part most coverage skips entirely.

Persson had been experimenting with a personal project called RubyDung. A 3D building game. He tried a first-person view but thought the blocky graphics looked too rough and dropped it.

Then he found Infiniminer. An indie game by Zachary Barth. Block-based. First-person. You mine, you build, you place things. Clean chunky visual style. Anyone could sit down and understand it in sixty seconds. Persson was part of TIGSource, an independent developer forum where Infiniminer was being played and discussed. Then the source code leaked. Barth open-sourced everything and walked away.

Persson had also been deep in Dwarf Fortress. A simulation game of almost incomprehensible depth. Entire civilisations generate from scratch. Extraordinary. But the interface was pure ASCII text and getting started required documentation longer than most novels. Most people bounced off it within twenty minutes.

He could see the gap clearly.

He went back to RubyDung. Brought in the first-person view. Added the blocks. Added survival and exploration. Called it Cave Game. Built the first working version over a single weekend in May 2009.

That weekend is where the best-selling video game in human history started. Not with a pitch deck. Not with funding. With a question and two days.

The Playbook: Post Before It Is Ready. Charge Before It Is Finished. Leave Gaps on Purpose.

On May 17, 2009, Persson posted the first version of Cave Game on TIGSource. Not to the world. Not to journalists. To a forum of independent game developers who had the context to understand what they were looking at before it had any marketing or polish.

He did not announce a finished product. He shared a prototype and let people respond. Within hours the thread was alive. Screenshots of things people had built. Suggestions arriving in real time. Energy from real people who got it immediately.

He had genuine feedback from real humans the same day he posted. Most builders sit on things for months trying to make them perfect before showing anyone. By then they have made hundreds of decisions alone. Persson made almost none alone. From day one he was building in direct response to real people in real time.

Then he did something that still makes people uncomfortable: he charged before the product was finished.

On June 13, 2009, Minecraft accepted first pre-orders at approximately €9.95. For an explicitly unfinished game with no tutorial, no proper menu, and a developer most people had never heard of. Within the first month he had over 1,000 paying customers and more than 20,000 registered players.

Then he did something nobody talks about as a distribution strategy, because it sounds like a product decision. He released no instructions. On purpose.

No tutorial. No guide. No onboarding flow. No explanation of how anything worked. Without instructions, players had to figure it out together. They went to forums. They made YouTube videos. They wrote wikis. They formed communities around the shared experience of discovering what the game could do. And Persson fed that loop constantly — things players asked for on Tuesday appearing in the game by Friday. That speed built something no marketing budget can manufacture. Players felt ownership. They were not just users. They were participants in something being built live.

He stayed employed until leaving became the obvious conclusion. He moved from full-time to part-time at jAlbum as sales increased. He did not quit on excitement. He watched the numbers and waited for the mathematics to become undeniable. In June 2010 the Alpha released and PayPal froze his account because the transaction volume flagged as fraud. A part-time programmer at a Stockholm photo sharing company was not supposed to have that kind of money moving through a personal account. He left jAlbum shortly after. Not as a leap of faith. As the obvious conclusion to a simple arithmetic problem.

Then he turned down Valve. In September 2010 he visited Valve Corporation, met Gabe Newell, participated in their technical exercises, and was offered a job. He said no. He also turned down venture capital. Mojang was built on revenue from the beginning. No investors in the room. No board telling Minecraft what it needed to become. Four years later Microsoft paid $2.5 billion for what he refused to give away.

The Traction Framework: How Minecraft Spread Without a Marketing Budget

Gabriel Weinberg and Justin Mares, in Traction, identify 19 channels through which any business can get customers. Their central argument: most businesses fail not because of bad products but because of bad distribution. Peter Thiel put it directly — poor distribution is the number one cause of startup failure.

Persson ran one of the most effective traction strategies in the history of software, entirely by instinct. Here is what he did mapped against the Bullseye framework.

Here is the distribution chain that made $2.5 billion possible — and that you can replicate in any category with the right product decisions made early.

The best distribution strategy is not a channel. It is a product that people cannot help talking about because using it makes them want to show other people what they made inside it. Persson did not design this. He built the world he personally wanted to explore, and the absence of instructions was not a marketing decision. It was just how the game felt most honest. The distribution was a byproduct of authentic product decisions.

In 2026, the equivalent of this mechanism looks different. A creative tool where the output belongs to the user, not the platform. An AI product where the creative act is yours, not the system's. A community where what you build inside it is shareable and identifiable as yours. The blank canvas principle — give people a world they can change and they will tell everyone about what they made in it — applies to every category. Not just games.

The Mistake: Know What You Are Optimising For Before You Start

Microsoft paid $2.5 billion. He bought a house in Beverly Hills for $70 million. He had more money than he could spend in ten lifetimes. Within a year he was posting publicly about depression and isolation. He wrote it plainly:

"I love games and I love to program, but I don't make games with the intention of them becoming huge hits. I've become a symbol. I don't want to be a symbol, responsible for something huge that I don't understand, that I don't want to work on."

He had spent his life building things. That was the substance of his days. When the company sold, the building stopped. What was left was the noise. The number did not replace the work.

The lesson is not that he was wrong to sell. He was burning out. The scale was real. The exhaustion was real. The lesson is that he did not know what he was optimising for before he started building, which meant he had no framework for deciding what to do when the outcome arrived.

What You Can Apply: Five Principles That Work in Any Market

Your Move This Week

Persson did not win because he was the most talented developer in Stockholm. He won because he saw what two incomplete things were each missing, built the version that had both, shipped it the same week he finished it, and trusted the community that showed up to do what communities always do when something is genuinely worth caring about.

The gap was always there. He just looked.

And in 2026, with AI compressing the distance between idea and prototype to days, the question of whether you can build it is no longer the hard one. The question is whether you will look for the gap in the first place.

Next issue: Daniel Vassallo — how an Amazon engineer earning $511,000 a year walked away, built a course in 16 hours, and made $310,000 from it.