The Salary Is
Never the Trap.
The Story You Tell Yourself
About It Is.

Every issue of The Real How follows the same structure. The Setup. The Direction. The First Bet That Failed. The Playbook. The Traction Framework. The Mistake. What You Can Apply. And your move for the week.

Most people think the trap is the job. It is not. The trap is the story you tell yourself about what you would lose by leaving it.

Daniel Vassallo had the $511,000 salary. The promotions. The respect. The clear path to principal engineer and everything the system promises if you wait long enough. He had everything.

He left anyway. Not for another job. Not for a funded startup. Not for anything with a guaranteed outcome. He left because he stopped believing the next year was going to feel different from the last one.

What happened after is one of the cleanest demonstrations I have found of what happens when someone with real expertise stops trading that expertise for a salary and starts building small things from it instead. A book made $140,000. A 100-minute course made $310,000. A community sold for $3.6 million. None of it was a startup. None of it had investors. Every single thing started as a small bet. And the bets kept compounding. Let's get into it.

The Setup: Eight Years Building Amazon's Infrastructure

Vassallo grew up in Malta. Small island in the Mediterranean. Population around 500,000. He got hooked on computers at five, started programming early, and dropped out of university partway through a physics and computer science degree to shift to a correspondence program through the University of London. He took whatever work he could find between studying. Waiting tables. A school lab. QA for a winery.

His first real programming job was part-time at a sports betting company in Malta around 2005. He walked in and found the problem immediately. Their developer had let them down and the team was manually inputting large amounts of data every day. He automated it. Saved them hours of work before anyone expected him to. That moment taught him something he did not forget: the people who make themselves genuinely useful do not need permission to matter.

In 2010 he joined Amazon. Entry level. Building tools for AWS. Within three and a half years he had been promoted twice to senior engineer. He stayed for eight years. His salary started at $75,000. By his final year it was $511,000 — most of that in Amazon RSUs that vested and landed in his account like cash.

He had saved approximately five and a half years of living expenses before he left. He is honest about that. The savings made the decision easier. But he has said plainly that he believes he would have found a way regardless. He left Amazon in February 2019. He walked away from approximately one million dollars in unvested stock options. The salary was never the trap. The story was.

The Direction: Not a Plan. A Set of Constraints.

Vassallo did not leave Amazon with a plan. He left with a direction. And there is a real difference between those two things.

The direction was this: stay self-employed. Build things from what he already knew. Do not bet everything on one outcome. Keep the downside small enough that a failure does not end the experiment.

That is a completely different optimisation target than what most builder content tells you to aim for. When your goal is staying in the game rather than reaching a specific number, every decision changes. You do not double down on a failing bet because doubling down might end the experiment. You build in small units so no single failure can eliminate you. You choose structures that generate income without requiring your constant presence.

The motivation problem is not something a promotion fixes. It is something a promotion reveals. Every new level brings more reward and the same emptiness. Which means the problem is not the level. The problem is the system. He was honest enough to admit that before the next promotion arrived. Most people are not.

The First Bet That Failed: What $10,000 a Year Taught Him That $511,000 Could Not

His first attempt after leaving was Userbase. A SaaS product for developers. End-to-end encrypted user authentication, making it simple to build secure web apps without managing the complexity. Technically interesting. Real problem. He got 1,000 people on a waitlist. Hit number one on Product Hunt on launch day. Made the front page of Hacker News simultaneously.

And then almost nothing. Userbase made around $10,000 a year. Not per month. Per year.

A different kind of person doubles down here. Raises money. Hires a team. Runs the startup playbook. Pushes through because the alternative is admitting the first bet failed. Vassallo did something different. He accepted the evidence, kept Userbase running as a small side thing, and did not let it consume him.

While building Userbase he had been sharing everything on X. The process. The decisions. The numbers. The problems he was hitting. Not as a marketing strategy. Because transparency was the natural way he processed what he was building. His following grew. Not fans — people who recognised themselves in what he was sharing. Employed engineers. Developers with expertise and a job they were not sure about anymore. That audience was not a business yet. But it was going to be.

The Playbook: Package What You Know. Ship Before You Overthink It. Let the Market Decide.

By mid-2019 Vassallo had eight years of AWS experience. He knew which services actually mattered and which were noise. He knew things that took a decade inside Amazon to learn and that most developers outside it were guessing about. He and his former Amazon colleague Josh Pschorr turned that knowledge into a book.

They called it The Good Parts of AWS. 173 pages. A deliberately opinionated take on which AWS services you would be foolish not to use and which you could safely ignore. Not a comprehensive reference. A specific, experienced perspective — the exact thing no documentation anywhere was giving developers. They launched it on Gumroad on December 25, priced at $65.

In the first 14 days it made over $45,000. Total lifetime revenue crossed $140,000. Over 13,000 copies sold. For roughly 160 hours of work between two people.

Then came the Twitter course. In April 2020, Vassallo released Everyone Can Build a Twitter Audience on Gumroad. Priced at $25. He put it together in 16 hours. One take. No editing. A screen recording of him talking through exactly what he had done to build his audience from zero to 24,000 followers.

It made $100,000 in the first two months. Total revenue crossed $310,000. Over 13,000 sales.

Sit with the arithmetic. $310,000 from 16 hours of work. That is more money per hour than almost any knowledge worker on earth earns from a salary — including the $511,000 Amazon salary he left behind. But the 16 hours is not the real number. The real number is the two years of transparent building on X before he pressed record. The product succeeded because the person selling it had built the exact thing they were teaching, in public, in front of the people buying it. The credibility was already there. The product was just the packaging.

Then the community arrived, by accident. In November 2021 he ran a cohort course — A Portfolio of Small Bets. Six live Zoom sessions over two weeks. Around 30 people per cohort. He ran 29 cohorts. As demand kept coming he realised the value was not in the sessions. It was in what happened between them. The Discord. The people who stayed and kept building together. He repackaged it: join the community, the course is included, lifetime access, one-time fee.

Between November 2021 and October 2023 Small Bets generated $824,000 in revenue and $600,000 in profit, growing entirely by word of mouth from members who had built real things inside it. In April 2025 he sold it to Gumroad for $3.6 million. The Amazon job paid $511,000 a year and required eight years to reach. Small Bets made $3.6 million and required four years to build from a cohort course he started as one more small bet.

The Traction Framework: How Vassallo Found His Channel and How You Find Yours

Weinberg and Mares, in Traction, argue that most businesses fail not because of bad products but because of bad distribution. The book identifies 19 channels. The Bullseye framework tells you how to find the one that works for your specific product and audience. Here is what Vassallo ran against that framework — entirely by instinct.

Weinberg's rule — most businesses get zero distribution channels to work, and if you can get even one to work you have a real business — applies perfectly here. Vassallo's channel was X. Not for volume or follower count. For the specific, irreplaceable quality of witnessed experience. Nobody else in the world could have launched that Twitter course and had it perform the way it did. Because nobody else had been transparently building that specific audience for two years in front of the exact people who were about to buy it.

In 2026, the same mechanism is available to you in whatever domain you already work in. The tools to build a transparent, specific, daily presence in your field have never been cheaper or more powerful. AI can help you structure your thinking before you post. The platforms are there. The audiences are looking. The only thing missing is the decision to start before you have a product to sell.

The Mistake: Aim at a Life. The Number Is What Happens on the Way.

The $3.6 million exit was not the goal. It was the output of hundreds of small decisions made while optimising for a completely different thing. Vassallo has said clearly that he did not leave Amazon to build the next billion-dollar company. He left to remain self-employed.

That reframe changes every decision. A $140,000 book is a win. A $310,000 course is a win. A SaaS making $10,000 a year is a data point, not a catastrophe. The metric is staying in the game with your autonomy intact. When your target is self-employment rather than scale, you stop asking whether each thing is big enough and start asking whether it keeps the experiment running.

The second mistake he identifies is waiting too long to believe the numbers. When the AWS book made $45,000 in 14 days, most people would have called it luck and moved on. He watched it carefully, understood what it meant, and built the Twitter course using the exact same logic. Package specific expertise. Sell it to the audience that has been watching you build. Keep the production simple enough that the downside is measured in hours. Let the market tell you whether it was worth doing.

What You Can Apply: Four Moves, In Order

Your Move This Week

Vassallo did not leave Amazon because he was brave. He left because he was honest enough to admit that the next promotion was not going to fix the thing the last one did not fix. The salary is never the trap. The story you tell yourself about what you would lose by leaving is the trap.

He stopped telling that story in February 2019. What came after is the record.

In 2026, with AI compressing the distance from expert knowledge to packaged product to paying customer, the only variable in that record is when you decide to start.

Next issue: Justin Kan — how strapping a camera to his head and broadcasting his entire life accidentally created the infrastructure for Twitch, and what that teaches builders about committing fully to the strange idea before the market understands it.