Citapp

I want to share what I've been building, partly to put it out there and partly because I could use this community's brain on the next part.

A while back I was writing a biology essay. Somewhere around the hundredth citation I realized I was spending more time on references than on the actual essay. Two things in particular drove me up the wall.

First, my sources were never in one place. I'd start on my desktop, switch to my school laptop, and half my links and notes just weren't there, so I'd waste twenty minutes re-finding stuff I'd already found.

Second, and this was the worse one, I'd read the perfect quote, keep writing, and then completely forget where it came from. I cannot tell you how many nights I spent at 1am with thirty tabs open trying to track down one sentence.

At some point I stopped complaining about it and started building. That became CitApp.

The idea is simple. You highlight any text on a page, save it in one click, and it grabs the source, files it into the right project, and turns it into a correct citation in whatever style you need. Everything syncs across your devices, so the thing I kept losing just stops getting lost.

I built it solo, mostly with AI doing the heavy lifting on the code. The stack is Next.js, Supabase, and Gemini for the parts where a page won't hand over clean metadata. I'm not an engineer by trade and I went back and forth for a while on whether that mattered. I decided it doesn't. Nobody using a citation tool cares how it was built. They care that the citations are right and that it's free.

The hard parts have been less about features and more about the messy real world. A lot of sites return an empty shell to a server or sit behind a bot wall, so pulling an author and a date reliably is genuinely hard. Dates are the worst. Wikipedia gives you a creation year, some sites give you a modified date, and you have to make peace with getting it wrong sometimes and letting the user fix it fast.

Where I'm at now: it's live, it's free for the core stuff (five sources a week, all the citation styles), and there's a Pro tier for unlimited use. I wired up analytics this week and started watching real people use it, which is equal parts exciting and terrifying.

Here's where I'd love your take. My product is for students, and students are a weird audience to reach. They don't read SaaS newsletters or hang out on Product Hunt. I've started answering questions where they actually are, like Reddit, study forums, and TikTok, being upfront that I built the thing and trying to be useful first instead of dropping links. It's slow going.

So my question: if you've built something for students, or any audience that doesn't live in the usual indie hacker channels, how did you find your first real users? What actually moved the needle, and what was a waste of time?

Happy to share more about the build if anyone's curious. Thanks for reading.