I recently came across The Google Analytics Blog via the Twitter feed of my favorite Googler and found it pretty useful for understanding some of the program’s more ambiguous features. I’ve worked in analytics quite a bit in my days, and it can be a powerful tool if you know how to use it to it’s potential. Moreover, if you’re obsessed with online network dynamics like me, it’s a form of entertainment. I used to explore the ScienceBlogs analytics data for hours in the evenings after work marveling at quirky things, like how the traffic from one person’s personal blog that hasn’t been updated in months could contribute more incoming traffic to the site than a highly funded campaign. I do like me some irony.
Particularly interesting on the Analytics blog was this post titled 10 Myths About Google Analytics. While some of the “myths” are clearly an excuse to trumpet their selling points, there are some tidbits of useful advice in there.
One good and crucial thing about this blog is that it links to the Google Analytics support forum, in which reside employees who know the intricacies of that system and get paid to respond to your queries.
Regarding MYTH 2: Google Analytics is basic and doesn’t have any “advanced” features or metrics, if this is a real complaint, whoever said that clearly did not actually log in, let alone attempt to drilldown to specific areas of content and explore different metrics (Hey, did you know you can click things on the Internet??).
Another good thing I have to report about this post is the reminder via MYTH 8: Google Analytics does not support A/B or multivariate testing and isn’t well-integrated with other tools, that you can use Google Website Optimizer to test different features that you’re thinking about implementing on your site. A lot of web development decisions are made according to flimsy reasons, like that something “looks good” and are based on the personal preference of a few people. But I like to approach development like a science by starting out with a hypothesis (about a design aspect or wording on a heading) then running tests with both scenarios to let the numbers show which is better received by the masses. Of course, if they were MY personal preferences, they would almost certainly always agree with the science. But not everyone has the instincts of the Queen of the Internets when it comes to navigating sites, so quantifiable data is always nice.
Though top ten lists are all the rage, it sounds like most of the “myths” about Google Analytics come from people who are intimidated by the system and don’t do the proper research to find what they’re looking for before they call tech support to complain. But this isn’t a bad thing — it just means there’s more jobs for nerds like me!