He Built 12 Startups in 12 Months and Failed Most of Them. The One That Survived Changed How the World Works Remotely.
By Creatives Takeover · June 8, 2026
Building fast, failing often, succeeding once.
In 2014, a 28-year-old Dutch programmer named Pieter Levels was running out of money.
He had sold everything he owned, packed what was left into a single backpack, and started traveling the world. He had tried making music. He had tried freelancing. He had a growing sense that the internet could give him the freedom he was looking for, but nothing had clicked yet. He was burning through savings with no clear plan, working from hostels and coffee shops across Asia, and the clock was ticking.
Most people in that situation would have gone home. Pieter Levels made a public promise instead.
He announced he would build 12 startups in 12 months. One new product, every single month, shipped and live for anyone to use. No safety net. No team. No investor. Just him, a laptop, and a deadline he had told the entire internet about.
The Challenge Nobody Thought Would Work
The goal was not to build 12 successful companies. Pieter was honest about that from the beginning. The goal was to force himself to finish things, to break the habit of endlessly tweaking ideas in private and never actually shipping them.
He had noticed something about himself that a lot of builders recognise. The hardest part was not the building. It was the launching. So he created a system that made launching non-negotiable.
Most of the 12 products went nowhere. Some found a small audience and quietly faded. A few never even made it to the end of their month. But the challenge did something more important than produce hits. It trained Pieter to move fast, to kill his attachment to any single idea, and to treat every launch as a data point rather than a verdict.
By the time he reached the seventh startup on his list, that training had made him sharp enough to see something most people were missing entirely.
A Spreadsheet That Became a Movement
Nomad List started as a Google spreadsheet.
Pieter had been living as a digital nomad for months, and he kept running into the same problem. He never knew where to go next. How much does it cost to live in Chiang Mai? Is the internet in Lisbon actually reliable? Is Buenos Aires safe for solo travelers working late nights? He was piecing together answers from forums, Reddit threads, and guesswork.
So he built a crowdsourced database. Cities ranked by cost of living, internet speed, weather, safety, and quality of life. He threw it online in a weekend. It was rough. It was unpolished. It worked.
Within days, thousands of remote workers had discovered it and started contributing their own data. Within weeks, it had a life of its own. Nomad List had found the thing every founder spends years looking for: a real problem, felt deeply by a specific group of people, with no good solution in existence.
It took four years to reach one million dollars in annual revenue. That might sound slow until you remember that Pieter built it entirely alone, never raised a single dollar of outside funding, and kept his costs close to zero the entire time.
Remote OK and the Job Board Nobody Knew They Needed
The year after Nomad List, in 2015, Pieter launched Remote OK.
The premise was simple. If tens of thousands of people wanted to work remotely, someone needed to build the place where companies could actually hire them. He launched it quickly, kept it lean, and watched it climb to number one on Product Hunt on launch day. It became one of the highest-voted products the platform had ever seen.
Remote OK became the largest remote job board in the world. At the time, remote work was still considered a perk, a nice-to-have for companies trying to attract engineers who did not want to relocate to San Francisco. Five years later, a global pandemic would force every company on earth to confront the reality that Pieter had already built infrastructure for.
He did not predict COVID. He just understood the direction things were heading before most people were willing to look.
Then AI Changed Everything Again
By 2022, Pieter had been building and shipping for nearly a decade. He had launched over 40 projects. Most had failed. A handful had made him financially independent. He had lived in more than 40 countries and 150 cities, running his entire operation from his laptop.
Then he watched the AI wave arrive and did what he always did. He shipped something.
In early 2023 he launched Photo AI, a platform that lets people generate professional-quality photos of themselves using artificial intelligence. No photographer. No studio. No expensive shoot. Just upload a few images, train a model on your face, and receive hundreds of polished portraits.
Within 60 days it was generating $40,000 a month. Within 18 months it had crossed $100,000 a month. By late 2025 it was bringing in $138,000 every single month, representing roughly 70 percent of his total income, which by then had grown to over $3 million a year. His net worth is estimated to be between $40 and $60 million as of 2026.
He built Photo AI in raw PHP, jQuery, and SQLite. No React. No microservices. No engineering team. Just one person who understood the problem and moved before anyone else.
The Number That Explains Everything
On the Lex Fridman podcast, Pieter said something that stopped a lot of people in their tracks.
Out of more than 70 projects he had ever built, only four made real money and grew. His success rate, by his own calculation, was roughly 5 percent.
That number sounds brutal. But it is actually the most important part of his story. Because Pieter did not succeed despite his failure rate. He succeeded because of it. The failures were not detours from the path. They were the path. Every product that went nowhere taught him something. Every launch that flopped sharpened his instincts. Every month spent shipping something the world did not want made him faster and more precise the next time.
Most people build one thing, wait to see if it works, and stop if it does not. Pieter built 70 things and kept going until four of them changed his life.
What He Never Did
Pieter Levels never raised venture capital. He never hired a team. He never rented an office. He never waited for permission or validation from anyone in a position of institutional power.
He built in public, sharing his revenue numbers openly on X, his failures as loudly as his wins, his code when people asked. That transparency was not just a personality choice. It was a strategy. The trust it built became distribution. The audience that followed his journey became his first customers every time he launched something new.
He also never let his tech stack become an identity. While the startup world argued about frameworks and architecture, Pieter shipped working products in languages that other developers quietly looked down on. The products made money. The debates did not.
Five Things Pieter Levels Teaches Without Trying To
Volume beats perfection. A 5 percent hit rate across 70 projects built a $3 million per year business. You cannot find your winner without launching your losers.
Shipping is the skill. The challenge was never really about 12 startups. It was about breaking the habit of not finishing. That habit, once broken, compounds.
Simplicity is an advantage. Vanilla PHP and SQLite built a $138,000 per month product. The best tool is the one that gets you to launch.
Build in public, build trust. Transparent revenue numbers and open failures built an audience that became distribution. The trust came before the product.
Your audience is your moat. Photo AI reached $40,000 per month in 60 days partly because Pieter had 700,000 followers who had watched him build honestly for a decade. You cannot copy that overnight, but you can start building it today.
The world did not need Pieter Levels to raise money, hire a team, or follow the rules. It needed him to keep shipping until something stuck.
His journey is the biggest lesson on how extreme self-ownership can empower any entrepreneur to achieve their milestones without external support.