Facebook Co-Founder Andrew McCollum speaks in Oxford

November 7, 2007

 

Andrew McCollum

Andrew McCollum

At 6.15pm on Monday this week, Andrew McCollum, Co-Founder of Facebook spoke at the Said Business School in Oxford. Afterwards, I had the honour of meeting him along with Andy Young and some other Oxford students.

Andrew first introduced the genesis of Facebook referring to 3 founding members: Mark Zuckerberg, Dustin Moskovitz and Andrew McCollum. In Andrew’s words:

“Mark was the high-level thinker, Dustin did operations – stuff like looking after the servers and communicating with the users, while I did all the graphics for the first version of the site, lots of backend development and also built up a reputation as ‘the solver of hard problems’.”

The majority of Andrew’s talk focused on product strategy. With Friendster and plenty of other social networks already up and running when thefacebook.com was launched on 4th February 2004, many people have wondered what led to Facebook’s ridiculous penetration of well over 50 million people’s lives. Andrew explained that it came down to what is called the “social graph” – a notional graph of everyone in the world where each node is a person and each edge is a connection between two people. By trying to simulate pre-existing relationships and interactions in the offline world, they were making it easier for users to carry out tasks they wanted to do anyway – except it was more efficient and easier on Facebook.

Another key to Facebook’s success in the early days was that they only implemented what Andrew described as “everybody features”. Things like profiles, private messaging, the wall, photos, groups and events. Despite receiving many requests from people to implement them earlier, features like videos and notes (a.k.a. blogs) were deferred because they wouldn’t have been used by a particularly high proportion of users. They were trying to keep the user interface as clean and intuitive as possible for everyone.

However, Facebook don’t always mollycoddle users. Their implementation of the News Feed on 5th September 2006 is a good example of this. Product managed by Ruchi Sanghvi, release of this feature totally disrupted the user experience for everyone on Facebook and within two days the backlash from users led to over 1 million people joining groups in rebellion. This even included one entitled “Ruchi is the devil” – an issue Mark clarified in his first apologetic blog post in response to users. But Facebook didn’t remove the feature, they improved the privacy settings, stood their ground, and today most users probably regard the News Feed as one of the most useful features on the site. I certainly do.

The point Andrew illustrated when he talked about this was that you need to push users in the direction you know they want to go in – even if they don’t yet know it themselves. Why? Because it will ultimately lead to a better product.

The final part of Andrew’s talk I’ll mention is about the key form of virality – word of mouth. He said the correlation between:

  • success of a web application, and
  • probability of a user telling a potential user about the application

is close to 1. And what is it that will cause someone to tell someone else about a website? A compelling experience. Whether this is using a feature that makes your life so much easier, or even being bitten by the zombie reincarnation of some guy you went to primary school with, working out how to provide compelling, emotive experiences is the key challenge for anyone looking to build a viral Web application. It’s something that Andy and I have recently been focusing on with GroupSpaces, and we’re just beginning to see some results from our e-mail newsletter feature.