Adaptive Path Blog

PM: How did you go from making a game to making Flickr?

CF: We were building Game NeverEnding, which was browser-based. Not swords-n-sorcery. Trying to create a situation in which the game was a context for socializing. People need a reason to play games and hang out. If you are playing a game, you have an excuse to hang out. A very social game–wandered around environments and found different objects. Unfortunately, in 2002, it was a very bad time to do this. People didn’t understand this. We had just enough money left to do one last thing. One last shot was this photosharing thing. Emerged out of the game. We hadn’t done our research; if we had, we would have known it was a loss-leader for photo printing. We inadvertently created this massive multiplayer photosharing game. As a result, created this new thing. Pieces certainly existed and it made sense to people. Cameraphones helped–in 2003-4, these began to take off at the same time.

PM: None of the other sites were taking advantage of camera phones. Created an environment where cameraphones photos could be used almost like printed photos.

CF: In a new technology, you inherit all the metaphors from the old technology. We didn’t have that. It wasn’t an album, it was a photo stream or photo set. All “digital native.” In 2004, people had gotten past the need for an album. People wanted public photos.

PM: What were the internal changes you had to make to make this transition happen?

CF: It was very hard–we all loved the game, but it simply was not succeeding. We we’re going to make it. We had 30,000 passionate, active users. We had to make a big decision: for a while, we were building both products in parallel–with 6 people. It came down to a vote and it was evenly split. Stewart and I convinced Eric to change his vote by guilt. Then we had to go through the process of shutting down the game. Users were crushed by the closing of the game. It was a very difficult thing. From the ashes came this new site.

PM: How in the early days did design happen? How did things make it onto the screen and how has it evolved?

CF: You used to drag photos into a conversation. It was a neat idea, but not a great product. No easy to bootstrap a company out of this. In order to share photos, you had to be logged in with people you knew to share photos. It didn’t take off initially. We kept on iterating and building. In June 2004, you could finally put a photo on a webpage, and then it finally took off. We added tagging, the open APIs, badges, blog this photo. All that stuff.

PM: How are your design decisions made? Do you do a lot of sketching? Right to code? All of the above?

CF: We were way too design heavy. We had too much design overhead. Started out fairly clear–straightforward design. Great deal of functionality. Biggest design challenge was letting the photograph be the thing that was shown, that stood out the most. But then again, hugely functional. The trick is how do you expose the functionality but let it be subtle enough to not detract from the photo. We did very little sketching. Photoshop right into code. Doing this so fast. Design the icon, put it up there, right the code. Doing 10 pushes a day–incredibly fast. Did the first version of Flickr in 8 weeks. All of the product was designed in the next four months. Done very much in collaboration with the users. Posting an average of 50 posts a day in the forums in that first year between the six of us. We’d respond immediately. Moved navigation from the right side to the left, launched it, within a minute forums lit up “this is a disaster.” We moved it back. Design lasted only 2-3 hours.

PM: The About Page on Flickr. The importance of expressing a product’s vision from the user’s point of view. How did this statement emerge?

CF: I think we sat down and did some soul-searching and wrote down all the reasons and broke them down into clusters and this was what emerged. We needed a way to explain it to people–site introduced whole new behaviors. It helped focus people. This is why we have what we have. To direct back at the users.

PM: How useful was it internally?

CF: It keeps you on the path. It’s indispensable. These are the things that will put you on the true path.

Liz Dunn: How did you not push users down the “print tunnel”? How were you going to make money?

CF: We knew we were going to have two accounts: free and premium. Knew this from the beginning. When you start something as a business (instead of as a hobby), you need a business strategy. If you add it later, it’s difficult. We were also trying to raise money, so we needed the plan. The real challenge is making it happen at a big company like Yahoo, now THAT is hard. Six people in a garage is easy.

PM: You are now at Yahoo and your role is to bring innovation to Yahoo. How do you get a large organization to adopt more innovative techniques?

CF: I have never worked at a big company. I brought with me all these practices and prejudices from working at start-ups–I didn’t know how to work any other way. Documentation? Meetings? With six people you don’t need meetings. So trying to take the best of those practices and see what was in common with the people at Yahoo. Hack Yahoo Group was formed. Buy a lot of Red Bull and a lot of pizza and you get a lot of motivated, creative engineers around and you say build stuff. Build something in 24 hours. Demo at the end of 24 hours. Simple thing, but it’s like an art. You need just enough control, but not too much. Balance it so you keep the Powerpoint Parade at bay. Happens once every quarter.

Jonathan Rasmusson: How do you handle users’ feedback? How do you know when your customers are wrong?
CF: If you spend too much time listening to users, you can destroy your product. The needs can’t be too specific to individuals. Also, people don’t always know what they want. You have to use the suggestion box, but you have your set of features you know you are going to add. You can throw those out and say, which is more important to you? Do you prefer this feature or that feature? Immediate feedback is also good. Usually a very vocal, passionate group of people will be very attached to certain things. We listen, but we try to listen with reason.

Sam Felder: When Flickr rolls something out, it’s just a blog post and new button. Is this gentleness planned for and the sense of stumbling onto something?

CF: We don’t want to push things on people if we think they are going to be specialty items. The slivered almonds of Flickr. Gives people a sense of discovery and ownership. You don’t want to put big banners: Go Try These Features. Similarly, when we launched Pipes, it was a wild success. We underestimated that. The low-key release has been a strategy, but we’re very much a early-adopter type product and we’ve retained that feeling.

This entry was posted on Monday, February 12th, 2007 at 3:27 pm and is filed under Adaptive Path, Conference Commentary, Experience Strategy. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.

This post is licensed under a Attribution-Noncommercial-No Derivative Works 3.0 Unported License.