My latest app has two users. One of them is me.
To be honest, this isn’t new for me. I’ve made many apps that never took off, and given a limited pool of energy, how does it make sense to maintain an app that will never be adopted by anyone else? The only difference now is that my pool of energy has deepened. AI agents are in their element when they are stitching together pre-built UI components and building dashboards, and frankly most of what we build is not new. One of my long-time collaborators said to me 10 years ago, “all we do is ifs and loops”.
The upside of building for yourself is:
- Uncompromising feature set
- No subscription, no extra password to manage
- Local-first, accessible only over WireGuard
- Complete data ownership
I was maintaining a primitive household management system through a matrix of spreadsheets, which is very difficult to hand over to anyone else. Most people know their way around a dashboard app, so my fiancée will not feel like she’s been thrown in the deep end. So I built Tuis (Afrikaans for “at home”), an app tuned to our requirements, encompassing the disparate parts of life.
A userbase of two used to mean “don’t bother”, but now it means “no compromises”.
Some things obviously still make sense as SaaS. I’m never going to build or self-host my own Bitwarden because the stakes are too high, and I’m not going to build my own peer-to-peer chat app, because there’s no way I’ll convince people to install yet another chat app.
But the agentic coding era means tuning software to your own requirements is cheap enough to throw away. Give it a try. Just don't leak your Stripe keys in your network requests.
Have you built anything like this?
Cross-posted from LinkedIn.