School Alumni Facebook Experiment Results

Halls Of Horror
Creative Commons License photo: DownTown Pictures

On Sunday I introduced my experiment with building a community through Facebook for a school alumni group. I’m pleased to be able to say that the Facebook application was complete yesterday and I can now discuss some of the results.

First of all, a little about the target audience. We are looking at sixth form college students in Britain (between 18-19 years old). In grand total, there are 200 students in total and that’s the total number of people we are aiming at. I would estimate that about 25% of the target audience doesn’t use the internet on a regular basis so the target figure of course is about 150.

We’ve had two attempts at creating a community around the school alumni group.

Attempt 1: Drupal

The first attempt involved using the open source Drupal CMS. The community featured forums, latest news and so on. The community was advertised primarily through real world posters and presentations. There was also a small campaign online to promote it via e-mail.

After one year, we managed about 50 registrations (33% of the target) of which about 5 are active. This means the community is actively used by about 4% of the target audience. In total, about 30 messages were received on the forums in the one year the community has been active.

I believe the main barrier to the success of this online community is the difficulty in creating an account. You need to set up yet another username, password and validate your e-mail address. That means people have to go to a lot of effort to join the community and the lack of active members in the community also limits the usefulness (more on this later).

Attempt 2: Facebook

This is the new alumni community which I launched yesterday lunchtime. The community has various features which are similar to the first attempt. The community again was promoted through real world posters and presentations, as well as online via email and messaging. This time, viral marketing also took place through various Facebook channels.

After just 24 hours, the community reached 75 users. That means we’re already half way towards the target of 150. Are people engaging with the application? We received 30 messages in just one day. According to Facebook statistics we received 1,100 page views in just the first 12 hours. For me, that’s astonishing growth – through a Facebook application we achieved the same results in under a day as it took a year to achieve on an external Drupal-based website.

From my perspective, the amount of work that it took to promote the Facebook application was much less than that of the external community.

Conclusions

Bright Atlanta
Creative Commons License photo: Nrbelex

A sceptic will point out that this is an unfair test because the two communities were obviously different in their designs and feature sets. I will say that when it came to launching the second community, I was much more experienced at creating and designing online communities so I was able to create a much more engaging community. But it is the fact that we opted for a Facebook application which allowed us to develop these engaging features. Many features have a critical mass of users below which it would not be useful.

The growth of community is proportional to its size. The constant of growth (k) depends on various factors:

  • The proportion of users who are active in the community
  • The proportion of users who engage in the community
  • The proportion of users who will let their friends know about the community
  • The conversion rate (invites to registrations)

I would say that Facebook applications have a much higher growth constant than other communities, for several reasons:

  • Easy to access; no additional URL, username, password to remember
  • There are additional communication channels through Facebook
  • It is much easier to build engaging community features through Facebook
  • Facebook has an invite system for users to share their favourite apps
  • Users already have Facebook accounts meaning the “barriers to registering” are almost zero: all they have to do is to tick a box granting an application access to their personal data.

Of course, it depends on the demographic of your userbase. This experiment was run with British teenagers, many of which have Facebook. Across other demographics, I believe implementing OpenSocial or OpenID could lead to similar benefits for online communities.

I’m very pleased with the results of this experiment and I would say it has confirmed some of the suspicions I had when I wrote my previous post. Of course, it’s something you can rip apart but I believe it’s well worth a think when you come to create your next online community!

Leave a Reply

Your email address will not be published. Required fields are marked *