User experience is everything
summary
A browser WebRTC product called appear.in launches after a six‑week intern demo and grows via Hacker News and Reddit because it is no login, no install and sharing a room link just works. Retention falls so the team targets teens with stickers and chat, which increases viral growth. A notification feature scales badly — one room causes 2,000 database reads per message — so queries are rewritten and performance recovers; simple UX changes (visible sign‑up) eliminate repeated support cases.
AI-written summary · about
chapters (21)
- 0:00Intro
- 1:17What is appear.in?
- 2:29It all started with the interns
- 3:223 students, 6 weeks
- 7:17The world's biggest WebRTC-based video-conferencing service
- 7:46The average user couldn't care less
- 8:19What could we build that was better?
- 8:54Just share the link
- 9:35User experience made us better, but technology was the enabler
- 9:45The user doesn't care about technology
- 11:57Talk to your users!
- 13:03Quantitative vs. Qualitative
- 15:48People have different expectations
- 16:58Get a target market
- 17:50Choosing a target does not mean your product will only fit those people
- 19:16"Access profile" is the new login
- 20:41Keep innovations to a minimum
- 20:56Users need to feel familiar
- 22:06Listen for feedback
- 24:35Performance was horrible, simple requests taking 5 seconds
- 28:19Summary
related talks
31mHow to think `User first`
41mHacking Culture with Javascript
21mShipping Faster: Behind the Scenes of How We Built
06mApplication Chats
21mOptimizing the customer experience
26mThe Key Ingredient Retaining Users in Your App and Reducing Churn
17mBuild for ALL of Your Users
06mIf You Build It They Might Not Come