Facebook Platform Starts Living Up To Its Potential

This guest post was written by Joe Green, the Founder and President of Causes. Causes is the largest online platform for activism and one of the original Facebook Applications. Green’s approach to building online tools is rooted in his offline experience as a grassroots political organizer for federal, state, and local campaigns. He recently wrote a guest post on Facebook Smart Lists, and, as a launch parter at today’s f8 conference, has significant insight into how Facebook’s announcements will impact its developer platform. Disclosure: Green was an early advisor to Facebook and has a small amount of stock.


I was one of the first people to join Facebook in February of 2004, and launched one of the inaugural applications on the platform in May 2007. The new Facebook profile and Open Graph announced today, along with the launch of smart friend lists last week, is going to usher in a new era of the Facebook platform. And I believe entire industries will potentially be revolutionized by social, from travel to reviews to health to e-commerce, and of course charity.

Wait. Wasn’t this supposed to happen 4 years ago, when Facebook launched platform at the first F8? Yes it was, but it didn’t. Today Facebook Platform is known primarily for gaming. This is not to say that Facebook Platform has not accomplished a lot so far; we, for one, owe the existence of Causes to it. But it is still far from achieving its potential.

The original dream of Facebook Platform was to enable developers to build experiences that were social at their core, like Facebook Photos, without having to build their own standalone social network. Remember all of the “me to” social networks built just to have a social feature Facebook and Myspace didn’t have? I built one for political discussion called Essembly. It enabled unique and potentially transformative social interactions, but only 20,000 people ever used it.

It was in this context that Facebook launched platform, with the noble goal of making core parts of your life that are social offline equally social online, as Facebook had already done when it upended the already-mature world of photos. This aspiration was evident in many of the Platform launch apps: iLike in music, Flixster in movies, Where I’ve Been in travel, Causes in charity. (In fact, Where I’ve Been seems to be a new FB in-house app, maps). We launched Causes on Platform precisely because we believe that altruism is fundamentally social and should be core to online identity, as it is for most people offline.

When it first launched Platform, Facebook opened up every integration point they could think of, including friends lists, profile, news feed, email, and they even created new channels like requests and notifications. The main value Facebook intended to offer was providing everything a developer would need to make his or her product social. And to sweeten the deal (and I think as something of an afterthought), Facebook offered these developers some distribution. This last point, as it turned out, was key.

There were two main reasons that the first 4 years of platform played out as they did: 1) Communication Channels were overwhelmed by the distribution Facebook offered developers and 2) Facebook Profile did not offer an enduring presence for applications.

Communication Channels:

When Platform launched early in the morning on May 25th, 2007, from a ballroom at the W hotel with engineers sitting on the floor, apps grew faster than anything else in the history of the Internet. Causes was at a 1 million installs inside of a week. And the race was on. This massive growth was enabled by the opening of so many friend-to-friend promotional channels, but fueled by apps aggressively pushing users to invite their friends any way they could.

Apps grew very fast, but mostly because they would create lightweight but broad engagement, rather than deep engagement with friends.  Facebook spent the next couple years focusing on clamping down on spam on Platform, and slowly eliminated or restricted communication channels they had previously opened to developers . This was unfortunately necessary, but it also stifled the development of revolutionary applications in core utility areas because the same channels that are most ripe for abuse are also the ones providing the deepest connection between users.

Facebook’s attempts to deal with these issues also often ended up increasing spam, because as conversion rates of communications went down, developers increased volume. If you want 1 click and the conversion rate is 10% you need to send 10 messages, if you have a 1% conversion rate you need to send 100 to get that same 1 click. What succeeded on platform were, for the most part, applications that could get users to frequently and somewhat indiscriminately communicate with lots of friends.

The changes made over the past few weeks, along with those announced today, will fundamentally alter the paradigm for Facebook communication channels. Specifically, there are two main types of communication on Facebook: direct friend-to-friend, and feed posts. For the direct friend-to-friend communications, friend lists allow for appropriate targeting, therefore making more high bandwidth channels acceptable. Instead of having to send requests to donate to my birthday wish to 50 indiscriminately chosen friends, I can send an FB message or email to 8 of my close friends.

For feed posts, the new Open Graph with specific edge types allows Facebook to know a lot more about data coming into the feed and to filter accordingly. I also can do more to filter my feed manually, and I have been seeing much more interesting content in my feed as the people in my close friends list get more prominence. It turns out, perhaps against Facebook’s better judgment, that people are willing to do a bit of work to improve their experience.

Finally, it is worth mentioning that for the first time since live stream came and went, Facebook has increased the total amount of feed stories consumed with the rollout of a mixed real-time and top stories stream, which is a win for everyone publishing into the stream.

Profile

From the day Facebook launched in 2004 the profile was the most critical, and really, the only page on the site.  Only with the addition of News Feed in 2006 did attention shift away from profile.  People used to navigate the site by surfing friends’ profiles, and used their own profile as their navigational starting point.  The profile used to consist mostly of a list of interests, actively curated by the profile owner.  With the addition of news feed the entirely of your Facebook experience has moved toward representing primarily your recent and frequent activities, which is great for capturing someone in a given moment, but not the whole of the person.

For example, your favorite movie is likely not the one you most recently watched, and there is no way to represent that on the current profile except buried on the info tab. Likewise, you might only donate once a year to your favorite charity, but that one act may mean more to you than 50 moves you’ve made in a game. In other words, the current profile benefits applications that have frequent usage, such as games, but offers no way to show how deeply someone cares about something.

This was not always the case. Of all the integration points opened at the original F8, profile boxes were the ones both developers and users were most excited about. There was an entire category of apps that users did not interact with frequently but loved to have on their profile. One great application, called “Where I’ve Been”, let you put a map of the world on your profile and highlight every country you had visited. But like the communication channels, profile boxes quickly overwhelmed the profile and Facebook justifiably pared them back and then eliminated them altogether. The day the last profile boxes disappeared we received the most user complaints in our history, making it very clear to us that people want their giving to be represented as a core part of their identity.

This new version of profile, if it works correctly, should achieve a balance between displaying things you have done recently and representing a comprehensive view of who you are. Key to this will be understanding which actions you really care about, even if they happen infrequently, and which frequent actions you care less about. Facebook seems to be intending to do this algorithmically, but I bet they will end up incorporating a big dose of direct user curation, as they are now doing with friend lists. Being able to ‘star’ a story is a good start, but it will need to be able to happen more in the flow of posting.

Whatever you think of Facebook, you cannot fault it for lack of ambition. In the course of a month they have fundamentally altered pretty much every major part of their site—profile, feed, friends—sometimes more than once. User reaction will likely follow the usual pattern of initial outrage followed by gradual acceptance and embrace. The more open question is how the rest of the Internet ecosystem will react. Facebook is making its usual appeal to developers: “give us your data, and we will give you distribution and engagement.” This tradeoff has not always worked out as intended, but I think it will this time.

I am confident we will see major sectors, from music to reviews to commerce, revolutionized by authentic friend-to-friend interactions. We are fortunate at Causes to have a big start in one of the largest markets around, the $300 billion giving market. It is anyone’s guess if the other major categories will go social with their current leading companies, or if entirely new ones will emerge, like Zynga in gaming. Either way, it will be a fun ride.