A year ago, I needed a word counter. Not a complicated one — just something that counted words and characters as I typed. I searched, found pages covered in ads with blinking popups, and closed the tab. So I built one myself in an afternoon.
That first tool was ugly. No dark mode, no mobile layout, just a textarea and a few stats. But it worked. I shared it with a few friends, and someone said: "Can you add a sentence counter?" So I did. Then someone else asked for reading time. Then JSON formatting. Then a QR generator.
I kept building because each request felt like a real problem. Someone needed to check their essay word count at 2 AM. Someone wanted to format JSON without pasting it into a random site. Someone needed a quick mortgage estimate without creating an account.
Ninety tools later, I've learned things that surprised me.
I spent weeks on a complex PDF editor with merge, split, and OCR features. It's technically impressive. It gets a few hundred visits a week.
The word counter I built in one afternoon? Tens of thousands of visits. Same with the password generator and the QR code maker.
The pattern is clear: people don't want feature-rich tools. They want the one thing they need right now, with no loading, no sign-up, and no confusion. A tool that does one thing well beats a suite that does ten things okay.
"Every tool you add is a decision you force on the user. The best tools have zero decisions — just open and use."
When I first added "runs in your browser" to the homepage, I expected nobody to care. I was wrong. Users email me specifically because tools don't send data to a server. People paste sensitive text, financial numbers, and private documents into these tools.
Building everything as client-side JavaScript was more work — especially for complex tools like image compression and PDF processing. But it's the single most common reason people tell me they use Knexio over alternatives.
After building 60+ browser games, I have enormous respect for game developers. A tool either works or it doesn't. A game needs to feel right — the timing, the physics, the controls, the feedback loop. A game that works correctly can still be unfun.
The biggest surprise was which games took off. I expected Chess and Sudoku to be popular. I did not expect a simple reaction speed test to be one of the most-used pages. People love quick, repeatable challenges they can improve at.
Early on, I read every email and tried to build whatever was requested. That led to scope creep and unfinished tools. Now I treat feedback differently:
The best features I've built came from noticing the same request three times from unconnected people. That's a signal worth following.
Building the tools was the fun part. Writing blog posts and guides to help people find them was a completely different skill. I learned that:
If I could go back, I'd start writing useful guides earlier and worry less about tool quantity.
Building a tool takes a day. Keeping it working for a year takes ongoing attention. Browsers update. APIs change. New devices ship with different screen sizes. A tool that worked perfectly on Chrome last month might break on the latest Safari update.
I now spend about as much time maintaining existing tools as building new ones. Automated testing helps, but there's no substitute for manually checking tools across browsers every few weeks.
Keeping 90+ tools free requires some form of funding. Advertising is the most practical option for a site like this, but it comes with trade-offs:
I've chosen to keep ads minimal and accept that the site will grow slower as a result. If you're reading this and wondering how to fund a free tool site: ad revenue works, but it takes volume to be sustainable. A single popular tool won't pay the bills — you need a library that brings people back.
I'm focusing on three areas for the coming months:
I also want to write more posts like this one — less structured, more personal. Not every piece of content needs to be a ranked list or a step-by-step tutorial. Sometimes the most useful thing you can share is what you actually learned along the way.
If you've used any of the tools on this site, thank you. If you found a bug and reported it, double thank you. And if you're thinking about building your own set of free tools — do it. Start with one thing you need, make it work well, and see where it goes.
— Kai Zhao, June 2026