How Flickr Made It To The Next Level

The second in our series, The Pivot, about the crucial changes in direction that saved some of the most successful startups in recent history. In this installment: Flickr.

How Flickr Made It To The Next Level

As with many web innovations, Flickr, the site that became synonymous with online photo sharing, started out as one idea only to pivot to something completely different.


Ten years ago, newlyweds Caterina Fake and Stewart Butterfield were developing a massively multiplayer online game called Game Neverending. Both were early bloggers with a creative streak and a flair for literature, and were intrigued by the concept of a virtual world ruled by social interactions, a precursor to World of Warcraft and Second Life. The way Fake (her real name, which has caused countless hassles at airports) once described it to me, Game Neverending sounds pretty far out. Part whimsy (a player would find a wombat whistle, blow it and baby wombats would cuddle up against her) part meta-reality: Users could work alone or together to create a business or product, raise venture capital, hype it, and everyone would grow rich. Players could cheer up one another by sprinkling magic sparkle powder in a crowded room. You can see screencaps of the game here.

Born on the web, it wasn’t the kind of game they could shrink-wrap and sell at Walmart, though, and they had trouble attracting venture money. The young couple mortgaged their home to partially self-fund the project, which they added to friends’ and family money they raised, but they didn’t have enough to complete development. So they had to scale back any grandiose ambitions. Because an important aspect of the game hinged on instant messaging, Butterfield suggested they develop it with the capability of dropping photos into conversations, which they agreed would be “very cool.” In early 2004 they built a prototype and stuck it up on a webpage.

It went nowhere.

Image: ReadySetRocket

They confronted the same chicken-egg problem any inherently social business faces. For a user to benefit, others had to be on the same instant messaging client. But how do you get that initial stream of users? The answer was to create a primitive social network with photo-share capability and incorporate tagging to organize photographs. This took advantage of two powerful trends that were converging. First, digital cameras were getting cheaper and more powerful by the month, and naturally people wanted to display them online. It meant that photography had become, as Fake put it, “delaminated from its delivery mechanisms.” Second, now there existed a convenient way to organize thousands of photographs online, which had grown out of a social bookmark site called Delicious, the first to deploy a non-hierarchical classification system in which users could label bookmarks to better organize them.

Fake, an amateur photographer and professional web designer, knew firsthand the nightmare of organizing thousands of photos on a hard drive. Before tagging she was faced with a seemingly intractable organizational mess–endless lists of JPEG files with no effective way to place them into context. She could, of course, organize them into folders, but that was merely a binary solution to an infinitely more comple organizational challenge. And since anyone could add a tag and comment, it became a democratic mode of organization that gibed with Flickr’s mission, which was to be public in the way a blog was.

The earliest members came from the Game Neverending community, and Fake made it a point to greet everyone who entered the site to introduce him or her to someone else. “I see you are into yoga–here’s someone else who’s dabbled in Bikram,” Fake might say. She viewed online social interaction as if it was a cocktail party and she was the hostess. If you come to a party and nobody offers you a drink, she reasoned, you’d leave to a happening party somewhere else. Her job was to ensure you stuck around.


While Flickr was free, it wouldn’t always remain that way since serving photos online suck up expensive bandwidth. Why not use that to help seed the site? Fake thought. She instituted a viral policy: Invite five friends to join and get three months free. The implicit carrot-stick worked better than she could have imagined.

For the first few months signups increased 75- to 100% month over month, then spiked as bloggers became hip to the site. As the blogosphere exploded, so did Flickr. Every blogger who used it acted as an advertisement, since to view pictures readers had to click to Flickr, as illustrated in this famous graphic called the FlickrVerse, which shows how Flickr spread. Its growth accelerated to 30- to 50% month over month, a product of the social and conversational-fueled viral loop that was created. The more blogs that displayed photos that were stored and shared on Flickr, the more people were exposed to Flickr. This led to more blogs and more blog readers adopting the service.

Soon the site that Butterfield and Fake had started as an afterthought to their multiplayer game became the 5th most popular site on the Internet, generating revenue by charging a fee to heavy users wishing to store large numbers of photos. A year after launching, the formerly debt-riddled couple sold Flickr to Yahoo. Six months after that the site hosted its one-millionth photo. Five years later it was approaching 3 billion images.

Although Flickr might be a billion-dolla business today, Fake says she has no regrets selling out for a reported $40 million: “If you throw a rock in Silicon Valley you can find somebody that could have been a contender,” she says. “I think it’s a waste of time to sit around and think of what could have been.”

Adam L. Penenberg is a journalism professor at NYU and a contributing writer to Fast Company. Follow him on Twitter: @penenberg.

[Image: Sanzhar Murzin via ShutterStock]

About the author

Adam L. Penenberg is a journalism professor at New York University and author of several books.