American Idol’s Top 11: Explaining the Unexpected

This post was written by Phil Kam, Cision Analysis’ director of new services who is integrating social media analysis into general PR measurement protocol.

In earlier weeks of American Idol, social media volume offered accurate predictions of nationwide voting, resting on the premise that those who care enough to mention contestants on Twitter or Facebook are also likely to text “vote” in their support. However, as viewers developed stronger connections with contestants, their opinions and actions become more complex, and the correlation weakened. This is an example of the unfortunate truth that not all buzz translates to sales.

Judging by what happened during the Top 11 elimination, it seems a number of observations are worth sharing. In case you missed it, Casey Abrams received the lowest votes despite his high popularity compared to the other male contestants. He was duly eliminated, but saved at the last moment by the three Idol judges.

For five days before his performance, Casey ranked an enviable second place collectively in social media buzz. He dipped to fourth place (behind Scotty McCreery, James Durbin and Jacob Lusk) going into the day of the performance, but he would have still been a sure bet to avoid elimination. So why did he receive the lowest share of the 30 million votes cast that week?

Social media is organic and fast-paced, so data that maps changes and fluctuations is invaluable when it comes to spotting early warnings. Looking at trending data among male contestants during the week, you will see Casey’s popularity fluctuated most among male contestants and received the least “refreshed” buzz after he performed. It might be an indication that he is least successful in acquiring new fans or he is losing his fans in higher rates. In sum and/or in isolation, Casey’s buzz alone neglected to tell the entire story.

  • Casey’s buzz volume experienced the most significant “dip” in share from March 18th to March 22nd. While his volume dropped nearly 85% in 4 days, one of his closest competitors, James Durbin, fluctuated less than 10%.
  • Similarly, Jacob Lusk’s volume dropped in excess of 80%, but was boosted to more than 10 times after his performance. Casey, in comparison, received less than six times the increase in post-performance buzz.

 

Avoiding a dramatic elimination and with a save from the judges, Casey’s popularity was renewed to a new level.  Studying the residual effect of the buzz in coming weeks will be necessary to see if America will continue to vote to support him.

  • During the period between the March 30th and 31st shows, Casey received nearly 35% more buzz than the closest male contestant, Scotty McCreery.  In fact, after avoiding near elimination, Casey dethroned Scotty as the most mentioned male contestants for the first time since March 1st during days when the shows are aired.
  • Votes cast the week after the elimination nearly doubled to 55 million.
  • As of April 1st, Casey’s official American Idol twitter profile is followed by 54,000 people – 2nd only to Scotty’s 55,250 followers.  According to data collected from Twittercounter.com, the elimination buzz only marginally boosted Casey’s followers “growth rate”.

 

As said, a comprehensive PR measurement program that is actionable should consider various data input and factors.  Using this incident as an example, the following data and info could be considered:

  1. Knowing which competing contestants’ buzz correlate best.  The audience will continue to support by voting—knowing how and to whom fan loyalty shifts can predict gains as more popular contestants are eliminated week after week.  In PR, this is priceless—knowing how to capture the void left by other brands if consumers were to switch.
  2. New vs. old – who is gaining the most new audience attention?  And who is capturing the steadiest loyalty from fans?  Knowing the statistical correlation between which fans will vote and which will not is similar to identifying influencers or opinion leader in PR.


  • Doug Smith

    Vote for James !