Skip to main content

Solve big enough problems

Dropbox logo

Dropbox

Show how the product would work, then build

Dropbox was born on a bus trip. It's an idea for a remote storage system, available from any place through the Internet. Drew Houston often forgot his USB disks, so needed some workaround.

He thought that he could solve his own problem by himself. He wrote the code, but before he launched Dropbox, he made a short video. It was a presentation of Dropbox's capabilities, with Drew's voice commenting on what he was doing on-screen. He uploaded it along with a waitlist to Hacker News and Digg (Reddit's predecessor).

The video got 70k first signups the waitlist next day, mainly thanks to the inside jokes of the Digg's community that he included

What the video didn't solve was capacity. The waitlist held about 5,000 names before it went up, and the team expected it might reach 15,000 after. It jumped to around 75,000 overnight instead, and nobody knew how much load the servers could handle, so they kept the product invite only rather than opening it to everyone at once. The scarcity helped: a waiting list made getting into Dropbox feel exclusive instead of making the delay feel like a problem.

Dropbox's early adopters were almost all Silicon Valley regulars, and growth flattened once that crowd was tapped out. The team brought in Sean Ellis, the marketer credited with coining the term growth hacking, to push past it. In 2009 they tested a two-sided referral program: an extra 250MB of storage for the person who referred a friend, and 250MB for the friend who signed up. Houston has said he modeled it on PayPal's old referral bonus, a design he later admitted he had likely misremembered.

  • Signups nearly doubled in the first 10 days after the referral program launched, from about 100,000 to 200,000 users.
  • By 2010, Dropbox users were sending 2.8 million referral invites a month.
  • The referral program lifted signups by 60%, and the company grew from 100,000 to 4 million users in 15 months, with no paid advertising.

Want to build growth systems like this?

The playbooks and AI skills behind how I run these plays.