Monday, October 22, 2007
Traffic Jammed by Survey DataThe New York Times outlines the fight between web publishers and ratings services Nielsen/NetRatings and comScore over the "true" amount of traffic to websites.
Advertisers are caught between data from ad and web servers that give a more accurate representation of eyeballs versus the ratings services that approximate the traffic. This is an irrational argument to me, and not just because I'm on the media side.
Why by gosh would you ever believe a "survey panel" of a small percentage of users over audited data that tracks the pieces of content delivered to a computer screen? This is like having a phone survey asking how many people attended a concert instead of looking at ticket sales. While ticket sales maybe be slightly off when compared to filled chairs, it's much more reflective than a survey.
Assuming that robots are filtered out, ad servers and web servers are adroit at counting the who and how many. The advantage of a survey is it can eliminate the small percentage of people who visit a website both at home and work. Useful, yes, but not nearly as important as impressions and unique visitors as tracked by web servers.
Advertisers should put their faith in always-improving web server data and listen to surveys as supporting material to better understand demographics and repeat consumers.
By John Gartner at 09:05 AM | Comments (0)