Sunday, February 15, 2009

Experience Project Class Notes - February 13th

We heard from Julio Vasconcellos from the Experience Project on Friday. Here are a few of the discussion topics that I found most interesting:

- As someone who doesn’t actually use social technology much beyond e-mail and Google, I was interested in hearing Julio’s short list of websites that can help create buzz. A key to creating tipping points here is the idea of “all or nothing” – either you get the votes and get to the top story on a site (via others voting on it), or you don’t . . . and if you don’t, your article really doesn’t take off. There is also a difference between being popular and becoming viral.
  • Viral Equation: (% people who pass on the story) x (% invitations accepted) x (average number of people invited)
  • If > 1, then goes viral
- Some of the sites Julio mentioned:
  • Digg
  • Reddit
  • Stumbleupon (this one seems to be a service that finds a relays sites that you might like)
  • Yahoo!Buzz
  • Twitter
  • Kirtsy (targeted toward women)
- As in traditional media, it is important to think about the match between your message/issue and the audience of the social technology you are targeting. You can look up demographic information and topic use (e.g. categories like news, animals, etc.) for most of the sites. Quantcast (free) and Comscore (paid service) are two places that do monitoring. You can also get to know audiences by tracking different bloggers, spending time on these different sites, etc. Bottom line: know your audience and target your message.

- There also some interesting ways to message-test. Julio used Facebook ads, which are relatively cheap, to test combinations of messages/headlines and pictures. You can set this up to monitor impressions and click through rates. There are some specific issues here with regards to tracking whether someone just saw an ad vs. acted upon that ad . . . CPMs vs CPCs (if you are interested in exploring this further, I would talk to Shal). Jennifer also sent an email with information on message testing techniques, and the Google analytics information is really helpful too. As someone who has worked on non-profit communication campaigns for many years, the idea of having so many ways to do cheap message testing (as opposed to more traditional focus group work) is very exciting.

1 comment:

  1. In the context of us thinking about what actions to request people to take on the behalf of our cause, Julio suggested another interesting ask... "please Digg this." This is very relevant to the discussion that we had at the end of the class, especially with the college football reform team asking for input on what the action they request people to take should be.

    The other insight I thought was useful to consider is that tipping points should be thought of in series, using the momentum of one to set off another instead of thinking of them as discreet events.

    The final additional point that resonated with me was offering different levels of engagement - this sparked some discussion at the end of class around the trade-off of offering options that people feel comfortable with vs. keeping things simple. Kaan made the point about if he was given options for action like total fasting vs. not eating dinner vs. juice fast, etc, it would just be overwhelming and he would do nothing. I agree that simplicity should be the first priority and then offering options for engagement should be the second order goal after the simplicity is achieved.