Google analytics is free, it gives us a great deal of information that we, as webmasters, can use to assess site performance, keywords and on-site metrics that indicate user engagement.
However, any data is only truly useful if you understand exactly where it’s come from, how it’s been compiled and what elements of it could be inaccurate. Here are some important things to know when use GA to measure your site performance:-
- Javascript Reliant – Analytics uses “page tagging”. Webmasters have to add javascript snippets to every page they want tracked. Javascript code is executed on the visitor’s machine and can easily be blocked. Around 5% of internet browsers blocked javascript in 2008 – these visitors go unaccounted for with GA.
On top of this an unknown percentage of users have extensions such as “Adblock” that also blocks the GA tracking code. Furthermore only a small percentage of mobile users have phones that can execute javascript properly.
- Cookie Reliant – Analytics usually drops 4 cookies (cookies are simply small files of information about the visitor’s activities) to help determine unique visitors, referral paths and other useful information.
Cookies are dropped and stored on the client machine until the cookie either expires or is removed. I haven’t been able to find any reliable information about the %age of users that block cookies, but I’d say around 10% do. If cookies aren’t accepted, or are refreshed, GA will either not report the visitor or provide false information (ie. count the visitor as a unique when in actual fact they are a return visitor).
- No Spider Information – It is really important for a webmaster to understand how search engines are crawling their site. Information about the regularity and depth of site crawls can help understand effectiveness of site structure and what content is deemed “important” by the search engines. Unfortunately Google doesn’t provide any of this information or allow webmasters to upload log files containing the info.
- Analytics Under Reports – Webmasters often report that Google Analytics under reports traffic statistics. The client side javascript/cookie based model is inherently less accurate than a server side tracking option such as using a log file analyser.
- Reliance on Google – One of the biggest issues that companies face with SaaS (Software as a Service) based tools is that of reliance and data security. By using Google’s tracking software your data is stored on somebody else’s servers and can be removed/deleted at any point in time. It would be a real kick in the nadgers to lose years and years worth of web analytics data – and this has to be considered a risk.
- Stats aren’t compiled at Real Time – Analytics used to update once a day…nowadays it usually updates multiple times a day – but the data isn’t real time. If you operate a site with lots of traffic I’d recommend using a real time tracking solution to understand exactly what’s happening on your site at any given point in time.
Don’t Get Me Wrong – Analytics is Great…
It’s very powerful, free, has incredible extended functionality, conversion tracking, integrates with Adwords and Adsense, allows report creation, segments, filtering etc etc. It’s just important that we, as webmasters, understand just how accurate the data in Google Analytics is and what may be missed out!