Earlier today, Mashable posted a great article that casually predicted Adam Lambert as winner of tonight’s American Idol finale. The authors of the post didn’t use a crystal ball to come to that conclusion, rather the increasingly buzzy methodology behind search trend analyzation. This process uses things like geography and search frequency to predict social behaviors, and has shown an exciting amount of promise, particularly for people in areas like marketing or politics.

In this particular case, the authors tracked search trends by region to gauge interest in both Adam Lambert and Kris Allen, and their findings seemed to imply that because people were searching more for Lambert, his fans could likely depend on him stealing the golden microphone. However, fans this side of Universal Studios (read: OneRiot’s intensely masculine, pop rock-ridiculing, ‘Kill Your TV’ search team (cough)) begged to differ. Instead of looking solely at searches, our team was watching what people were actually saying about Idol’s two final contestants on the realtime web - and they seemed to be telling a different story.

Check out the above graph, which tracks social media mentions of both Lambert and Allen over the time period around their last performance. Instead of tracking the total number of mentions, this chart tracks the proportion of mentions in respect to each individual’s average fan traffic on the social web. While at times Lambert does appear to be inching ahead of Allen and vice versa, they are by most accounts neck-and-neck.
This is an interesting graph, but much like a general search trend analysis it is somewhat incomplete. When people search for or share something online, you can fairly assume that they are at least interested in the subject. What you can’t assume is why they are interested in it, because just knowing that someone searched for something won’t always tell you what they were hoping to find. Likewise, just because someone shares something online doesn’t always mean they care about it - in fact, it might just mean they’re talking crap.

So, what’s in a search? A lot of things, and getting to the bottom of them is tricky. Did people search more for Adam because they intend to vote for him, or were they simply hunting down more info on the more sensationalist of the two singers? Without getting inside the collective brains of America, It’s nearly impossible to say. Thankfully, “What’s in a social media mention?” is a question we can answer just by digging a little deeper.

In the graph above, you’ll notice that Kris Allen seems to be faring better than Adam Lambert (Allen fans, go get your high fives on). What you’re looking at is a filtered graph of positive social signals surrounding each candidate, meaning this accounts for not only for mentions of either candidate on places like Twitter and OneRiot, but also for the sentiments behind those mentions. Overall, while people seem to talk about both candidates equally as much (as indicated by the first graph), it looks like they’re much more likely to talk favorably about Kris Allan, using words like “love” and “must win.” Predictably, this indicates that they will really vote for him in the final battle.

In other words - remember all those searches for Adam Lambert? They might not mean that much, because the actual sentiment of the social web is predicting a different winner. Of course, we’ll have to wait a few more hours to find out, but until then - our money’s on Allen.
Buddy Up