Political Web Sites Post-Election Not Doing As Bad As You'd Think

|

ObamaMcCain_AP_10_07_08.jpgNow that the presidential election is over and Sarah Palin has skulked back to Alaska, we had wondered how politics sites could survive a massive drop-off in post-election traffic.

But according to number-crunching whiz kid Nate Silver of fivethirtyeight.com (itself a fledgling politics site), the predicted drop-off in traffic hasn't happened to everyone. A few politics-heavy Web sites (The Atlantic, Drudge, HuffPo) are even doing better.

Here's Nate's analysis of how well sites have done retaining their pre-election audiences (see his site for the full methodology). But big disclaimer up front: traffic numbers are based on Alexa, at best an imperfect tool.

Doing well: (Percentage of pre-election traffic retained post-election.)

  • TheAtlantic.com: 125%
  • DrudgeReport.com: 124%
  • HuffingtonPost.com: 109%
  • NYTimes.com: 97%

Of course, not everyone is surviving. Ouch:

  • Electoral-Vote.com: 30%
  • RealClearPolitics.com: 25%
  • RasmussenReports.com: 15%
  • Gallup.com: 12%
  • JohnMcCain.com: 9%

And fivethirtyeight.com itself? Retaining 59% of its traffic. But Nate, you have access to better analytics for your own site than Alexa. Why don't you give publish the real numbers so we know how accurate your analysis is?

See Also:
How Politics Sites Plan To Survive Post-Election
Three Reasons Political Sites Are Glad The Election Is Over



< Prev. Story
Next Story >

1 Comment

Garbanzo said:
Alexa? Puh-lease! Those numbers bear little resemblance to reality, particularly for small and midsized sites (below 2 million monthly uniques). Quancast and Compete are better, but unless it's a directly tagged Quantcast site, those numbers are wacky (so are Hitwise, for the record). And even for the big sites, it can be off by a factor of two or three. You'd think that these third-party services would at least reflect directionally what's going on, but upon looking at a number of major events and comparing these sites to page-based stats (e.g., Omniture, Google Analytics, etc.), they only occasionally hit the mark.

There is a vacuum of good numbers in the competitive space, but Nate of all people should know that you don't fill a vacuum with crappy numbers.

Join the discussion