Oscars Winners, How did Social Media Do?

Yesterday, I put up a quick post about WebTrends’ predictions about the Oscars based on social media mentions, in which I wrote that I thought that they would prove to be inaccurate.  The predictions that WebTrends made were as follows:

  • Best Movie: Inception
  • Best Actor: James Franco
  • Best Actress: Natalie Portman
  • Supporting Actor: Christian Bale
  • Supporting Actress: Helena Bonham Carter
  • Best Director: David Fincher

The actual winners in these categories were:

  • Best Movie: The King’s Speech
  • Best Actor: Colin Firth
  • Best Actress: Natalie Portman
  • Supporting Actor: Christian Bale
  • Supporting Actress: Melissa Leo
  • Best Director: Tom Hooper

In fairness to the forecasting work that WebTrends carried out, they were more correct that it looks.  In three of the four categories where they were wrong, it was the actor or director they had predicted in second place that won.  With the best film, they were “wrongest”, forecasting “The King’s Speech” as the 7th most likely to win.

As I wrote yesterday, I felt that the biggest challenge that is present when using social media data to predict what is happening in the wider world is that the population online is not representative of the wider population of the world.

Membership of the Academy is publicly known, and members are listed here amongst other places.  The fact is, that while Twitter has a fairly diverse population of users in terms of ages,  the academy does not, which means that their voting patterns and tastes will be less diverse.

What is interesting to consider, is: if WebTrends had been able to pin down a specific Age group within their study  would their results have been even closer?

When using Social Media data as part of your online marketing strategy, it is essential that you don’t just look at the whole picture, but look instead at the people who are your actual customers to see how they behave online, and what their opinions are.  They may differ from the herd, but not stand out from it unless you allow them to.