[mhofstadt]

I'd have to disagree a bit with the basic assertion of this article. While focusing on higher quality output of content is clearly useful, it can't be said that it is necessarily more useful than analytics. First, how do you quantify the quality of content? Is it by personal preference or by the number of site users engaging with that content? I'd probably use engagement (the stuff that leads to revenue) as the metric of choice for ascertaining the qualitative value of content, and , of course, that would be most accurately measured by the site's analytic data. Analytics help objectify what content people like and removes the aspect of personal preference.

Additionally, we need to understand why our traffic does certain things and, more importantly, why they do not do certain things. Analytics can help identify deficiencies in our site's user interface that cause friction for users to do the things we need. Without the proper data to show how users are interacting on a site, UI designers may be shooting in the dark if the goals are conversion rates and not just slick designs.