Comments on: Website Optimization and Data Analytics http://dsanalytics.com/dsblog/website-optimization-and-data-analytics_90 Data Analytics- the art and science of analyzing data Sun, 06 Jul 2008 04:19:29 +0000 http://wordpress.org/?v=2.0.1 by: » getting a handle on Web Analytics [ Data Sciences Analytics ] http://dsanalytics.com/dsblog/website-optimization-and-data-analytics_90#comment-25 Mon, 09 Apr 2007 02:09:05 +0000 http://dsanalytics.com/dsblog/website-optimization-and-data-analytics_90#comment-25 [...] It is not so clear what you can do with the data other than count it, and I have expressed some skepticism in my post “Website Optimization and Data Analytics”, but if you are interested the Immeria site gives an insight into what the commercial software Omniture does in this post Instances vs. Visits in Omniture [...] […] It is not so clear what you can do with the data other than count it, and I have expressed some skepticism in my post “Website Optimization and Data Analytics”, but if you are interested the Immeria site gives an insight into what the commercial software Omniture does in this post Instances vs. Visits in Omniture […]

]]> by: John Aitchison http://dsanalytics.com/dsblog/website-optimization-and-data-analytics_90#comment-22 Thu, 22 Mar 2007 01:33:56 +0000 http://dsanalytics.com/dsblog/website-optimization-and-data-analytics_90#comment-22 I just came across Peter Norvig's post " the Alexa Toolbar and the Problem of Experiment Design" at http://norvig.com/logs-alexa.html His point is bias. The extent to which (Alexa) data truly represents a random sample of internet users .. "in fact it only represents those who have installed the Alexa toolbar, and that sample is not random. The samplees must be sophisticated enough to know how to install the toolbar, and they must have some reason to want it. It turns out that the toolbar tells you things about web sites, so it is useful to people in the SEO (Search Engine Optimization) industry, so it overrepresents those people." He goes on to talk about the size of the bias, and it appears that it can be huge. So before we start build web analytics edifices, it might be a good idea to have yet another look at the quality of the data on which our analyses rest. I just came across Peter Norvig’s post ” the Alexa Toolbar and the Problem of Experiment Design” at http://norvig.com/logs-alexa.html

His point is bias. The extent to which (Alexa) data truly represents a random sample of internet users .. “in fact it only represents those who have installed the Alexa toolbar, and that sample is not random. The samplees must be sophisticated enough to know how to install the toolbar, and they must have some reason to want it. It turns out that the toolbar tells you things about web sites, so it is useful to people in the SEO (Search Engine Optimization) industry, so it overrepresents those people.”

He goes on to talk about the size of the bias, and it appears that it can be huge.

So before we start build web analytics edifices, it might be a good idea to have yet another look at the quality of the data on which our analyses rest.

]]>