Gmail Version Zero Searched the Wrong Email
Paul Buchheit wrote the first version of Gmail in a single day. It was not good.
He sent it to the engineering team. Their first feedback: 'It would be better with my email instead of yours.' Version zero only searched Paul's inbox.
Gmail v0 only searched Paul's email. Feature request number one: search mine instead.
One Feature Request at a Time
Version two searched your email. Then people wanted to reply to messages. Then they wanted more.
The entire product was built through iteration. Step by step. Feature request by feature request. No grand master plan.
No master plan. Just one feature request at a time.
The Happy User Gate
Google set a launch requirement: 100 happy users inside the company. Not downloads. Not signups. Happy users.
Buchheit embedded a popup: 'Are you happy? Yes or No.' He tracked who said no. Then he walked to their desk.
Some wanted a clone of Outlook. Lost causes. Others needed one bug fix. He converted them one at a time.
A popup asked: Are you happy? Yes or no. Then Paul walked to the Nos.
Deep Love Beats Broad Meh
Entering email was insane. The category was 30 years old. Everyone had opinions. You could not please them all.
Buchheit's philosophy: find deep appeal with a tiny fraction. Make them obsessed. Then expand from there.
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.
Deep narrow appeal first. Broad meh is impossible to convert to love.