How to Lie with Statistics 2.0: Using Pageviews as a Metric

Earlier this week comscore came out with numbers that showed Fox/MySpace had surpassed Yahoo in pageviews for the very first time. There was a lot of buzz in the blogosphere about this and many people took these numbers to mean that MySpace was more popular than Yahoo.

This sort of reminds me of 1999 when people used to use “hits” as a metric to compare websites. Back then you could slice an image into 4 different files and get your “hits” to triple overnight.

Well today, the equivalent of that is splitting your content between many more pages, just like Netscape has done, just for the sake of inflating your pageviews (so that Comscore and others can use that to spout off meaningless statistics).

Yahoo has been using Ajax to create faster and friendlier user experiences across their properties. Using ajax (properly) results in better user experience, but it reduces the number of pages a viewer has to click through. Myspace has notoriously bad user experience and they haven’t bothered using ajax to better it. Using pageviews as a metric in the web 2.0 space is stupid.

More pageviews = crappier experience = better Comscore numbers

Some of the smarter marketers are aware of this, but we need new metrics so journalists and ordinary people stop drawing false conclusions.