Although the field of web analytics is still in its infancy as it continues to change and adapt to new and emerging technologies it has made tremendous strides in the past fifteen years. Gone are the days where a simple counter at the bottom of almost every Geocities website was enough to get an indication of visitor interaction. This was a very primitive method of checking hits, and early counters simply allowed visitors to hit refresh to see the counter increase. Later counters fixed this problem and added information such as “X amount of hits since Y date” to give additional insight as to how long this data has been collecting.

In the mid 1990s some smart fellow looked at gleaming information from web server logs and log file analysis was started. Web server log files allowed for the introduction of metrics beyond hits, and included items such as page views and sessions. While not terribly important it allowed for greater insight as to visitor behavior than just hits which counters were collecting. In the late 1990s search engine spiders and other robots added to the complexity of tracking visitor behavior, and it made it hard to discern the difference between human and non human visitors. One such application that utilizes web server logs is Urchin 5 which is pictured below. Still heavily used throughout many companies it has a new version Urchin 6 in beta, and is one of the few surviving applications that still rely upon web server logs.

To compensate for the limitations of log file analysis companies developed the method of tagging every web page with a piece of java script to track visitor behavior. Every time that a web page is requested, including refreshes the java script code executes and sends that information back to the application vendor. The vendor also handles the sending of cookies to users to track metrics such as a returning visitors and unique visitors. Applications like Google Analytics pictured below offer a wide variety of insights that until recently were only available with the purchase of more enterprise level web analytics applications through java script implementation.

Today website owners have a multitude of choices when it comes to discerning visitor behavior information from their website. Choices include log file analysis through Clicktracks, Webalizer, Sitemeter, AWStats, and Analog. Java script solutions choices include Google Analytics, Microsoft Gatineau, Omniture Site Catalyst, Webtrends, and Coremetrics among others. For those of you who have blogs applications that are useful include MeasureMap, Mint, and Feedburner. My recommendation is to evaluate what information you require and then make adjustments to your software choice as necessary. If you’ve never used a web analytics application before then start off with a free one from Google or Microsoft to get an initial sense of what is happening on your website. There is no need to jump into a more complicated application that will sit collecting data without any insights being gathered simply from information overload. In later blog posts we’ll get into how to choose a web analytics applications and how to determine if a metric is worth monitoring.