"I Built 30 Projects. They All Failed Except One"

Thomas Sonis launched more than 30 products over several years before one of them—a launch platform called Unid—finally clicked and began generating $8,000 to $10,000 per month. His GitHub account is littered with abandoned repositories: a marketplace connecting Gumroad sellers with affiliates that made about $500 total before he gave up after two years, a feedback exchange platform called Frisbee that never even launched, a simple website builder named Gridley, a bookmark manager, and a web app to manage plants that only he ever used. The pattern is familiar to many indie hackers, but Thomas's willingness to articulate exactly why each one failed—and how he finally broke the cycle—is what makes his story useful.

The Five Reasons Projects Fail

Thomas distilled his failures into five recurring causes. The first is giving up too early. He sees people launch a product and abandon it after a week or two because they haven't gotten registrations or sales, but he insists no product works overnight. You need to verify the market exists and then iterate repeatedly, learning from mistakes.