Analytics Tips: Using Filters in Analytics Reporting

Ever strain water from a boiling pot of spaghetti? Well, you just have the basics for filters in web analytics.

Filters are a screening featuring within web analytics tools. They are meant screen certain aspect of the traffic that arrives to your site. Imagine filters as a strainer for draining water away from cooked pasta, and you have a good idea of how a filter should work.

In a website environment, filters include or exclude specific text information such as specific subdomains or directory in a URL, as well as a range of IP addresses.  Filters can also be programmed to rename URLs to make them more easily recognizable to the analyst.

To set them within Google Analytics, go to the admin page, then select the profile and view to which the filter will be applied.

To make sure the filter is recording the right data correctly, it is always a great idea to keep an unfiltered version of the data - data can not be recalculated once the filter is applied. Plus maintaining an unaltered version can be a diagnostic tool. Baseline trends from a profile without filters can be compared to reports with a filter to ensue the filter works as expected. This is especially important when Javascript expressions are applied as a filtering mechanism. Javascript expressions can become a bit complex depending on the amount of data being filtered.

For filters with regular expressions, text characters are used, such as  a slash / or brackets [ ]. The characters are designed to tell the analytics code what to included and exclude from the data.   For more details on regular expressions, view this Zimana blog post on what regular expressions are available.

Note that filters differ from the automatic segments available in analytic solutions, such as referral traffic or new vs returning visitors.  The idea behind a filter is to view a traffic segment based on technical aspects of the site.

Google Analytics Filter Selector