Custom Segments in Google Analytics
Google Analytics has a very useful feature called Custom Segments which allows users to create and filter custom datasets for all your reporting with a few simple steps. What makes Custom Segments different from the stock search and advanced filtering dimensions and metrics are twofold:
- Users are able to select and SAVE the segment, and not have to re-apply manually with each report viewed.
- Users are capable of creating reports that the stock reporting system is not set up to do. An example of this could be as simple as UNIQUE sessions from a unique referral source.
Example Case: Custom Segments to Exclude Ghost and Spam Bot Referrals
This example of using Custom Segments is based on cleanup work we did for our clients recently to exclude Ghost and Spam Bot Referral (source medium) from their Google Analytics data.
*Note that this method outlined below will NOT fix the problem. It is a starting point that you can use to modify the rules based on your specific spam referral traffic sources, and then view historic data that has already been negatively effected in order to see actual user data. This segment will eventually become outdated as new spam sources arise.
Step 1: Use Existing Segment URL
We’ve already set up this custom segment that you can use to test with you analytics views.
Visit this URL to start the process with your analytic account login: Custom Segments to Exclude Ghost and Spam Bot Referrals (last updated June 26th 2015).
*Note that these custom segments can not be automatically applied to your global website account. It MUST be applied by the EACH user for EACH profile view you wish to use the segment with.
Step 2: Add Custom Segment to Your Website'(s) Analytics ‘View’(s)
Step 3: View Reporting to Apply Custom Segment
Open your reporting and click ‘+ Add Segment’ at the top when you want to use this filter to review historic data without the spam (fake) referral data.
Step 4: Select the Custom Segment and “Apply”
Step 5: View Segmented Results
Now view your reports to see what historic traffic is actually real referral traffic and what is/was spam.