I shipped Nala 18 months ago. It is a meditation, sleep, and hypnosis app on Android, fully bilingual (English and French), with 13 specialist narrators, more than 300 guided audio sessions, 12 structured programs, 15 free SOS sessions, and a 7 day free trial on the paid tier. I built it solo. No team. No outside funding. No ads ever, in the app or in the marketing.

This is the post I wish I had read before starting. Not a victory lap. A list of things I got wrong, what I would do differently, and the technical decisions that turned out to matter way more than I expected.

Decision 1: 13 human narrators, not synthetic voices

The cheapest path in 2024 (when I started) was to use AI generated voices. Eleven dollars a month, near unlimited audio output, decent quality. I ran the experiment for three weeks. The result was an app that technically worked and emotionally did not. Listeners can tell. Not on a single sentence, but on a 12 minute meditation.…