Picking the Right Web Analytics Package: Getting Started

You have a website, but what’s really going on with it? The only way to really tell is through web analytics. Web analytics packages are the tools that report on how a site is used, when, and by whom. This data is vital to tailoring a site to the desired audience. A solid web analytics package is vital to getting the best results from a website.

There are two basic kinds of analytics packages: ones that operate by analyzing the log files produced by the web server, and external ones that operate via a script or image embedded in the page, a practice known as page tagging.

An example of the first type is AWStats, an open source log analysis package available at SourceForge. Like all log analysis packages, AWStats is limited in that it cannot capture the user’s activity on the page itself. Only actions that cause a file to be loaded from the server are available to log analysis packages. However, log analysis packages work regardless of the end user’s operating environment. The information sent by the users to the server is always available to a log parser.

An example of the second type is Google Analytics. Google Analytics operates via a piece of JavaScript code embedded into all pages of the site. This JavaScript allows the Google Analytics to track not only activities that result in a page load, but also the user’s activities within a page. The downside of this is that the script is more vulnerable than a log parser. If a user disables JavaScript or blocks the inclusion of external content, the analytics code can’t run, and won’t record anything about the user or their activities.

Both types of analytics packages have their strengths and their weaknesses. It is common to use both types of analytics packages on the same site. Since log parsers run on the server and JavaScript based analytics packages run on the client, they will not interfere with each other.

For more information on AWStats, visit the AWStats project at SourceForge.

For more information on Google Analytics, visit the Google Analytics homepage.