Version One in a Day
Buchheit: It was 2001. Google did a big reorg, gave engineers projects. They knew I had an interest in email — just go build it. So having learned from prior experience, and knowing I have a short attention span, the very first version of Gmail I wrote in a day.
It wasn't very good. I took code I'd been working on for Google Groups, shoved all my email into that indexing engine, and I kind of had email search. Back then, you couldn't really search email. You could find an email and click on it, but searching was a thing.
The first version of Gmail — I wrote it in a day. It wasn't very good. But it existed.
Feature Request: Search Someone Else's Email
Buchheit: I emailed it to the engineering team — hey, I built this email search, let me know what you think. People replied: yeah, it's okay, but it would be better with my email instead of yours.
So version zero only searched my email. The next version I made it search other people's email. Then people said — all right, now I want to reply to one of the messages. The whole thing was just iterating, step by step, trying to make something that would make people happy.
Version zero only searched my email. People said — it'd be better with mine. Feature request.
Are You Happy? Yes or No.
Buchheit: We had as our launch gate this idea that we needed 100 happy users inside Google. I embedded in the interface this thing that would pop up: 'Are you happy with Gmail? Yes or No.'
I'd get a list of who said yes and who said no. Then I'd go to all the nos and ask — what's it going to take to make you a happy user? Some people basically wanted a clone of Outlook. I'm like — you're never going to be a happy user. But other people, it would just be a minor thing — add one feature, fix a bug, and they're happy. We slogged through one user at a time until we got to 100.
Are you happy? Yes or No. I'd go to every No and ask: what's it going to take?
Deep and Narrow, Not Broad and Meh
Buchheit: When you're entering an established category — email was 30 years old — it's pretty much impossible to make something that appeals to everyone. If you try, you end up with a mediocre product nobody really loves.
Figure out a thing that will have really deep appeal, even if it's to a tiny fraction of people. If you can make that small fraction obsessively love what you're doing, it's easier to grow that group. There's always people at the margin — make something slightly better and they join. It's easier to start with deep but narrow appeal and broaden over time, than to start with broad meh and try to convert people from meh to loving your thing.
Start with deep narrow appeal. Broad meh is impossible to convert to love.