Filters, when used properly, can unlock a myriad of useful data to help you better understand your website visitors. Profile filters can help you organize your data and drill down to specific metrics or traffic segments you care about.
However, there is one caveat to be aware of. Adding unnecessary, incorrect, or too many filters can actually compromise your data. Why? Google Analytics filters actually work by permanently removing any data from your profile that doesn’t match the specific conditions of your applied filters, with no way of getting that information back.
You have to be careful implementing filters because too many filters or inaccurate filters can actually lead to:
- Skewed or corrupted data
- Complete loss of accurate data
- Strange historical comparison data
In order to avoid these issues, there are a couple things to keep in mind when adding filters to your Google Analytics profile in order to maintain the integrity of your data.
Always create and maintain a completely unfiltered profile view. This unfiltered view will act as the “control”, and will contain all your data just in case you run into issues (like the ones listed above). It’s also a good idea to test filters out before applying them to the main view to help troubleshoot problems. Many companies create a test profile view that is only used for testing new filters and settings, not for reporting.
When possible, use advanced segments, custom reports, or secondary dimensions instead of filters. Since adding too many filters can cause issues, it’s a good idea to use other ways to find the data you are looking for when you can. For example, a common filter that many companies utilize is to show the entire domain name in their reports. So, instead of just “/about/” it will list “www.exampledomain.com/about/”. The domain can be useful to see in reporting if a company is tracking several domains and sub-domains. However, you can actually find this data very easily without using any filters. By simply clicking on the “Secondary dimension” button, choosing the “Behavior” options, and then clicking on “Hostname”, a new column will be added to the report to show the full domain name associated with each page. Work-arounds like this will allow you to easily segment and analyze your data without using filters.
Which filters does your B2B use to help you analyze your data? Have you had any issues after using too many filters or implementing incorrect filters? Let us know!