I use expense splitting apps constantly. Apartment bills with my partner. Trips with friends. Dinners, nights out, the kind of shared spending that's just part of life.
For a long time, Splitwise was the default. Everyone had it, it worked well enough, and I didn't think much about it.
Then it got worse.
The frustration built slowly
It wasn't one moment. It was a slow accumulation of friction that eventually made me dread opening the app.
The daily expense cap. The wait timer. The upgrade prompts that appeared exactly when you were trying to do something simple. The UI that felt increasingly designed to remind you that you were on the free tier.
At some point I realized I wasn't logging expenses anymore. Not because I forgot, because I didn't want to deal with it. And when you stop logging, you lose track. You don't know what you owe. You don't know what you're owed. The whole point of the app falls apart.
That's the real cost of bad UX. Not the subscription fee, it's the mental overhead of not knowing where you stand with people you care about.
One day I sat down and thought: this problem is not hard. Why does the best solution feel this bad to use?
So I built Tab
I started in the first week of March 2026. As of writing this, I've put about 25–30 hours into it.
The goal was simple: build the thing I actually want to use. No artificial limits. No dark patterns. No moment where the app makes you feel like you're doing something wrong for wanting to split a fourth expense.
Tab does what expense splitting apps should do: tabs, friends, expenses, settle up. It works on your phone, your laptop, your tablet. You can add expenses by voice if you don't want to tap through forms. You get notified when someone adds something or settles up. It handles multiple currencies for trips. The basics, done well, with nothing in the way.
Why it's free
Tab costs me almost nothing to run. The app is on Cloudflare Pages, the website on Vercel, both essentially free at this scale. The backend, database, and Redis run on a $10/month VPS. Backups go to Cloudflare R2, which is also basically free right now.
I'm also lucky that I don't need Tab to make money. I have other projects: Conncord and SnitchFeed, that cover my costs. Tab is something I build because I use it every day and I want it to exist. That's enough.
The AI voice expense feature is free for all users right now. I'm absorbing those costs. If AI usage grows significantly, I might introduce BYOK (bring your own key) as an option. But that would be an option, not a gate. No paywalls.
TLDR: the moment an expense app costs you something to log an expense, you stop logging. That defeats the entire purpose. Tab is free because it has to be.
Will it stay free?
Yes, for as long as I can make it work.
If infra costs ever get out of hand (think $100+/month), I'll ask the community to help cover them through optional donations. No paywalls, no tiers, no "premium" features held hostage. If you find Tab useful and want to chip in, great. If not, the app works exactly the same.
I'm not building this to exit. I'm building it because I use it.
Open source, self-hostable
Tab is fully open source. The code is on GitHub.
If you want to run it yourself, there's a Docker Compose setup that gets the full stack running: backend, frontend, database, Redis, notifications, wherever you want. Your data stays yours. If I ever shut Tab down or take it in a direction you don't like, you can fork it, host it, and keep going.
That matters to me. I've always wanted to contribute to open source by solving a real problem, not just submitting a PR to a project I don't use. This problem spoke to me. I'm pretty sure I'm not the only one who's felt this way about Splitwise.
If that's you, welcome. Tab is for you.