Originally Posted October 8th, 2008, on MediaPost’s Red White & Blog
As a digital marketing company, it shouldn’t be surprising that iCrossing has done its fair share of Election ’08 analysis. As the manager of our advanced analytics team, I am always working to determine ways we can use our proprietary analytics techniques to analyze current events. I thought I would take a minute to recap some of the more interesting findings from our analysis, and discuss some of the current themes and my own take on the media’s influence in the current election.
In July 2007, we started looking at 41 candidates and 126 issues-related keywords we felt would influence the election. Since that time, we are now down to four names of candidates and their definitive running mates, and have increased the issues-related keyword list to 228 keywords. Beyond the changes in the competitive landscape, there have been some interesting changes in the demographics of those searching online, specifically that women increased their likelihood to use the internet to learn about candidates and the issues. Research shows that 56% of men and 54% of women in April reported using the internet to learn about a candidate’s status on the issues, up from the 47% of men and 38% of women during our original How America Searches: Election ’08 report in July 2007. You can find all of our findings online at iCrossing.com/Research.
Of course we run into the question of how important the internet is as a source of information to voters. We know that it is important, not many people will argue with that. Exactly how important the internet is to the election is an entirely different question. Our research shows that the internet is still second to TV, but as mentioned in the recent “McCain Winning Search-Ad Race” article in AdAge, 87% of internet users search for information about election issues, and 83 % of the issues-related keywords we researched had no visibility from the candidates. Not to mention that 85% conduct candidate research online to determine where they stand on the issues. The candidates are certainly playing the political media game online and paying attention to their demographics and media coverage based on some of the trends we are seeing in their strategies.
From our research at iCrossing, as well as my own consumption of the internet, it appears as though the parties are taking two very different approaches to online media and natural search optimization. Analysis has shown that McCain’s campaign is consistently spending more on paid search online while Obama seems to be focusing more on utilizing social media and natural search, using paid search less often. Obama is earning visibility, McCain is buying it. Without being able to measure the actual effectiveness of their paid search campaigns, Klaassen points out in the same AdAge article mentioned above that McCain is getting more paid traffic than Obama, 22% and 14% of web traffic respectively. McCain’s campaign is present in the social media space, but not as popular. Using Facebook as an example, some quick analysis this morning shows that Obama has nearly four times as many Facebook supporters as John McCain. Obama has more than 2 million Facebook supporters, an indication that he is playing into his online demographic, as is McCain by not focusing as heavily on social media and search. Democrats are more likely than Republicans to use both social media sites (46% versus 32%) as well as candidate websites (38% versus 24%) to find information about the issues while Republicans are more likely to find information online via news sites (81% versus 90%). Seems like the campaign strategists are paying attention to the numbers—maybe the strategists read Competing on Analytics: The New Science of Winning because this sounds like the makings of yet another anecdotal marketing story.
This also brings up the biggest challenge I see for Obama in this election when it comes to understanding your demographics. How exactly is he planning on getting the young voters out from behind their computers and into the polling booths on Election Day? I would be highly surprised by his campaign if he didn’t have a plan in the works. (If I had to guess, the plan involves a lot of text messages, e-mails, and Facebook updates.)
Due to these differing demographics, it also shouldn’t be surprising that the search volumes for Democrats have been significantly higher. Until Sarah Palin arrived on the campaign scene in early September, no candidate besides Hillary Clinton had ever had a higher search volume than Barack Obama, and even she only held that title for one month. Now, Palin’s numbers remain steep, check out Google Trends for all the details. Also interesting are the search volumes of the candidates’ wives — mirroring the conventions, Michelle Obama and Cindy McCain saw their highest numbers during their respective party conventions. Again, not surprising considering the spotlight, but neither of them stood a chance of remaining in the spotlight once Sarah Palin showed up.
How do I explain this increased search volume? Sarah Palin is running into the problem of establishing herself in the media and online right now. This is a problem Barack Obama ran into during the early phases of the election and likely lead to his high search volumes, but Palin is playing catch-up. Of the 41 candidates in iCrossing’s original How America Searches: Election ’08 piece, Sarah Palin was not one of them. I hesitate to put too much emphasis on search volumes because, while the analysis is quick and easy to produce, search volumes on candidate names do not show intent. It is likely that at this point in the campaign just as many people are looking for material to use against a candidate; more so than those looking for material to support one, although I have yet to see any research supporting this idea.
By far the biggest take away I see from this election when it comes to media is the fact that candidates are paying attention. They are looking at the numbers, bringing together online strategists and pairing them with their offline media strategists, and they are not risking their fate by ignoring emerging media. It will take many elections before we fully understand the impact of online media specifically, but MediaPost has done us all a service by providing an online platform for those of us analyzing the data to bring it together, document the response, and ultimately figure out what it all means when the results are known on Election Day. Issues change, candidates change, new media emerges, and just as the candidates are challenged on how to respond in the space, we are each challenged by how to measure their response.
