Lecture 16 - How to Run a User Interview (Emmett Shear)
Emmett Shear explains how conversations with broadcasters helped Twitch move beyond assumptions made at Justin.tv. In this historical Stanford lecture, he demonstrates a user interview, separates current behavior from feature requests and compares existing users, competitors' users and people who have not adopted any service. Notes cover selected public auto-caption passages for the 46-minute lecture, with speaker identity checked against the official Stanford CS183B archive. The date is the November 13, 2014 video upload. Twitch examples, platform comparisons and language-market observations reflect 2014. The captions include audience questions; interview-method claims are Shear's guidance rather than universal research rules.
Executive Summary (TL;DR)
Emmett Shear explains how conversations with broadcasters helped Twitch move beyond assumptions made at Justin.tv. In this historical Stanford lecture, he demonstrates a user interview, separates current behavior from feature requests and compares existing users, competitors' users and people who have not adopted any service. Notes cover selected public auto-caption passages for the 46-minute lecture, with speaker identity checked against the official Stanford CS183B archive. The date is the November 13, 2014 video upload. Twitch examples, platform comparisons and language-market observations reflect 2014. The captions include audience questions; interview-method claims are Shear's guidance rather than universal research rules.
Key Takeaways
Founders avoid talking to users
His first startup, Kiko Calendar, was built by two non-calendar users who never talked to any, and it was sold on eBay. His second, Justin.tv, used the trick of being its own consumer.