I'm a solo UX designer from London, and I built Dopamine Detox to solve my own problem: losing 3+ hours a day to mindless scrolling.
What it does:
- Tracks time on TikTok, Instagram, YouTube, Twitter, Facebook in real-time
- One-click blocking (30/60/120 min sessions)
- AI predictions warn you BEFORE you fall into a binge (premium)
- Build focus streaks with gamified rewards
- 100% privacy-first – all data stored locally, never leaves your device
Technical stack:
- Chrome Extension Manifest V3
- Vanilla JS (no frameworks – keeping it lightweight)
- Local storage only (no backend, no servers, no data collection)
- Dynamic site tracking with content scripts
- Declarative net request API for blocking
Early results (2 weeks live):
- 56 installs from 85 page views (65.9% conversion rate)
- 12 weekly active users (89% retention)
- 58% of users are on ChromeOS (Chromebook students love it!)
- 0 uninstalls in last week
The most interesting technical challenge was building AI predictions without sending data to a server. I use on-device pattern recognition to predict when you're about to waste time based on your historical behavior.
Pricing: Free forever (5 sites), $19.99 one-time for premium (no subscription).
Would love feedback from the HN community – especially on:
1. Privacy approach (is local-only storage compelling enough?)
2. Pricing ($19.99 one-time vs subscription)
3. Features you'd want to see
4. Technical implementation suggestions
Happy to answer any questions about the build process, challenges, or anything else!
Ogbon•1h ago
I'm a solo UX designer from London, and I built Dopamine Detox to solve my own problem: losing 3+ hours a day to mindless scrolling.
What it does: - Tracks time on TikTok, Instagram, YouTube, Twitter, Facebook in real-time - One-click blocking (30/60/120 min sessions) - AI predictions warn you BEFORE you fall into a binge (premium) - Build focus streaks with gamified rewards - 100% privacy-first – all data stored locally, never leaves your device
Technical stack: - Chrome Extension Manifest V3 - Vanilla JS (no frameworks – keeping it lightweight) - Local storage only (no backend, no servers, no data collection) - Dynamic site tracking with content scripts - Declarative net request API for blocking
Early results (2 weeks live): - 56 installs from 85 page views (65.9% conversion rate) - 12 weekly active users (89% retention) - 58% of users are on ChromeOS (Chromebook students love it!) - 0 uninstalls in last week
The most interesting technical challenge was building AI predictions without sending data to a server. I use on-device pattern recognition to predict when you're about to waste time based on your historical behavior.
Pricing: Free forever (5 sites), $19.99 one-time for premium (no subscription).
Chrome Store: https://chromewebstore.google.com/detail/dopamine-detox/emoh...
Website: https://dopaminedetox.uk
Would love feedback from the HN community – especially on: 1. Privacy approach (is local-only storage compelling enough?) 2. Pricing ($19.99 one-time vs subscription) 3. Features you'd want to see 4. Technical implementation suggestions
Happy to answer any questions about the build process, challenges, or anything else!