Paul Buchheit at Startup School 08
Key ideas
Buchheit argues that listening means decoding what users, advisers and your own reactions actually signal, rather than acting on the surface of what is said.
From a machine transcript of the audio. Quotes link to the moment they are said.
- 1
Hear the problem under the request
Buchheit argues users tend to be right about the problem and not about the solution they propose, using a faster-horse quote he says he could not confirm came from Henry Ford. His Gmail example: a run of odd requests for replying from the inbox turned out to mean the product was too slow, and the requests stopped once performance was fixed.
- 2
Understand advice instead of obeying it
Buchheit says he does not mean advice should be ignored; he asks listeners to work out why someone believes it and how their situation differs. He retells a Paul Graham essay in which an adviser, successful with boxed software, told the founders of a website to sell it in cardboard boxes. He reads 'that's impossible' as a statement of someone's limited experience, and warns that his own lessons come from consumer web software and may be terrible advice for enterprise.
- 3
Find one person who cares
Buchheit says most launches meet silence, and the job then is to find even one enthusiastic user, follow them and look for more like them. He sees this as a reason to launch quickly, and says that if nobody can be found it may be the wrong product. At Gmail the target was 100 happy users inside Google, gained one person at a time, which he found harder than he expected.
- 4
Treat your own irritation and energy as data
Buchheit advises noticing each wait, confusion or small annoyance in a product, and says most of what went into Gmail began with something that annoyed him. He applies the same test to jobs: he describes feeling drained at Intel, excited at early Google, and quitting Google years later when the drained feeling arrived there too.
- 5
Weigh features against the other 99%
Asked how to prioritize with limited resources, Buchheit says he uses a mix of what is easy and what he wants himself. He warns that adding a feature for 1% of users is probably usually a mistake, because a more complicated interface and harder code hurt everyone else, unless that 1% is really important. He favors trying things, saying probably most of what he put into Gmail did not survive and that some apparently bad ideas turned out well.
In Paul’s words
“So the problem part of what they're telling you is correct. But they probably gave it like three seconds of thought”
Paul Buchheit·Paul Buchheit at Startup School 08·Watch at 2:23(opens the original in a new tab)“You don't need a million happy users, you just need one person who's really enthusiastic and keeps sending you email”
Paul Buchheit·Paul Buchheit at Startup School 08·Watch at 9:43(opens the original in a new tab)“And so most of the things we put into Gmail were just like I was just annoyed with something and we'd try to think of a solution for it.”
Paul Buchheit·Paul Buchheit at Startup School 08·Watch at 12:38(opens the original in a new tab)About this entry
- Published
- Jan 30, 2009
- Channel
- startupschool
- Type
- Video
- Length
- 28:13
- Language
- English
- Added
- Nov 2, 2026
- Source last checked
- Nov 2, 2026
What’s available here
Original source
Linked. Opens on YouTube.
Player on this page
Available. YouTube’s own player, loaded only when you press play.
Full text or transcript
Not reviewed. No license to show a text version here.
Our summary
Draft. Written by us with AI from a machine transcript of the video’s audio (Whisper) and checked against it. Machine transcripts can mishear; each quote links to the moment so you can hear it. Not yet rights-reviewed.
Ask questions with AI
Not reviewed. Off until there is licensed text to work from.