Everyone

Reduce Referral Spam in Google Analytics Without Using Filters

Have you ever looked at your Google Analytics and found that most of the traffic was referral spam? Did you even know that website spam existed? Referral spam can destroy the accuracy of the data you’re receiving such as visitors, bounce rate, and page duration.

This is data that must be accurate to help you make data-driven improvements to your website and minimize the cost or even the need of a redesign.

What is Referral Spam?

Referral spam targets search engines. The technique involves making repeated website requests using a fake referrer URL to the site the spammer wishes to target.

Simply put, other websites are making requests to your website in hopes of improving their own search rankings. Except for polluting your statistics, this technique will not harm your website in any way. You can read a more thorough definition here.

Psst... if you sign up for my newsletter, I'll send my best advice and most popular articles for freelancers and designers right to your inbox.

I recently looked at the analytics on a client website and noticed that most of their traffic from the previous 12 months was spam.

60% of the traffic came from referrals. 90% of those visitors were on the website for less than 10 seconds. The bounce rate was over 80%. This was all indicative of spam. The 3rd most popular user language being set as “Vote for Trump!” was also a dead giveaway.

I once had this same problem on my own website and it ruined the accuracy of the data I was receiving. I saw a sudden spike in traffic and thought I had finally conquered SEO. Upon further investigation, it was just a bunch of ugly spam.

Identifying Referral Spam

The best way to identify referral spam is to sort your referrals by bounce rate in descending order. Alternatively, you can sort the referrals by page duration in descending order. Since these bots are usually automated, they only ping your site for a split-second. That makes the page duration extremely short and the bounce rate very high. Keep an eye out for something that looks like this:

ga-spam-languages

What Can You Do About It?

Most of the articles I’ve read require you to create complex filters for each referrer or referral spam URL. But it’s a tedious and time-consuming task to maintain filters for each new spam bot that attacks your website. You don’t have time for that. There must be a better way.

spam-exclusion-filter

What I found after researching this problem extensively is that with little exception, basic spam bots look for a Google Analytics tracking code that ends in “-1” like this:

  • ga(‘create’, ‘UA-11891287-1‘, ‘auto’);

To eliminate most of the spam you’re receiving, simply create a new property in Google Analytics, which will increase that number to 2 or more like this:

  • ga(‘create’, ‘UA-11891287-2‘, ‘auto’);
  • ga(‘create’, ‘UA-11891287-3‘, ‘auto’);

Once I implemented this new tracking code on my own website, my analytics data became much more accurate and I noticed most of the referral spam visits disappeared. It’s not a perfect solution, but it will be a big long-term improvement.

When Should You Make This Change?

This new tracking code will cause your analytics tracking to start at zero. Your data will still exist in the old property, but new data will be recorded on the timeline of the new property.

My advice? If your account is plagued with spam anyway, there’s no harm in deleting inaccurate data. You can export the last 12 months of data to be safe.

The best time to do this is at the beginning of a new year or when you release a new version of your website. Since the data is being recorded over time, it’s best to associate the change with a significant calendar event.

If you’re looking for even more details, check out The Definitive Guide to Removing All Google Analytics Spam

No thoughts yet. Share yours!

"Reduce Referral Spam in Google Analytics Without Using Filters"

Join the Discussion:

Comments post instantly. Your email will not be published. All comments are subject to moderation.