Michael Lynch left a developer job at Google in 2018 to build his own bootstrapped software businesses. The first stretch went badly. For about two years he shipped products that barely found paying customers and lost money, and he openly questioned whether quitting a comfortable job had been a mistake. He kept blogging through all of it, publishing honest retrospectives that slowly built an audience of other founders following along.

Halfway through his third year, he created TinyPilot, a small device that plugs into a computer and lets you control it remotely without installing anything. The software was open source and ran on a Raspberry Pi, but the demand was for a ready-made unit people could just buy. Lynch started selling assembled kits, and after a launch that spread through Hacker News and tech YouTube reviewers, orders came in faster than he expected. The product earned about $54k in its partial first year.

From there the numbers climbed steadily: $460k in 2021, $807k in 2022, and $993k in 2023. The catch was that for the first few years the business barely cleared a profit. Raw materials, payroll for a growing team, and repeated hardware redesigns ate almost everything. A switch to metal cases in 2023 changed that, letting him charge more and produce more, and net profit jumped to $236k that year on a team of seven.

Running a seven-person hardware company started to feel like too much. Lynch and his wife wanted to start a family, and he did not think he could be the sole operator and a present new father at the same time. He listed TinyPilot through the Quiet Light brokerage, presenting it mostly as an ecommerce business at roughly 3x trailing earnings. Two serious buyers competed, and in April 2024 he sold for $598,000, an all-cash deal with no earnout. After broker commission and legal fees his take was about $491k, bringing his lifetime profit from the business to around $920k over four years.

What stands out is how much of the outcome came from the documentation habit Lynch built early. Following advice from the book Built to Sell, he wrote playbooks in Notion for every recurring task so the company could run without him. When the sale closed, the buyer needed only about 25 hours of his consulting time because the team already knew how to run everything. The same transparency that drew his audience also drew a buyer who had read his public retrospectives and trusted what he was selling.