Few weeks ago, one of my clients started seeing spike in referral traffic in Google Analytics. Of course he got excited, thinking that his website is getting popular, people are linking and visitors are coming. Unfortunately that wasn`t true. You don`t have to be a Google Analytics expert to see that all the traffic my client was so excited about was, in fact referral spam.
Referral spam is used as a lead generation technique or as a way to drive traffic to shady websites and gain affiliate commission. In this article you`ll learn how exactly referral spam works and how to recognize it in your Google Analytics account.
Types of Referral Spam
Referral spam is coming to Google Analytics in two different forms:
- as traffic from spammy web crawlers and
- as ghost referral traffic.
There are many spammy websites that uses bots (web crawlers) to crawl the web and index webpage data.
Most respectable web crawlers identify themselves as such to web servers and therefore they are excluded from the Google Analytics reports.
The spammy web crawlers, however, don’t identify themselves as robots. In that way, the websites that sent the spammy crawler end up in your Google Analytics referrals report and trick you into visiting them (or visiting other spammy websites to which they redirect).
If you open one of these fake referral URLs (not recommended), you will probably be redirected to an online store, online marketing scam or a porn site. The owner of the fake referral website receives a commission if you purchase something from the website he`s redirecting to.
Below is an infographic that explains how referral spam is used in affiliate marketing:
The other type of referral spam in GA is called ghost referral traffic .
Ghost referral spam doesn`t even visit your website. It is generated by simple programs that sends fake HTTP requests to multiple Google Analytics properties. These fake HTTP requests are translated as fake hits and sessions in your GA reports. The automated script that sends the fake HTTP requests can set the referrer to be any URL.
So basically, once the spammer has your Google Analytics property ID, he can promote any website by generating page views in Google Analytics without sending requests to the actual web site.
Below you`ll find a list of domains that are known to be produced using ghost referrals:
How to recognize referral spam in Google Analytics
The fake referral traffic is not difficult to recognize in GA reports. It typically has several of the following characteristics:
- Bounce Rate of 0% or 100%
- Avg. session duration of 0s
- % New Sessions of 0%
- Landing Page is /
But typically doesn`t mean always. For all four criteria I mentioned, you could also see counter-examples, such as the following:
Almost all of the referral sessions of this website are coming from fake referrals. We see that most of the visits have a 100% bounce rate and 0s session duration. The first website, however (floating-share-buttons.com) is also a notorious referral spam but it has a bounce rate 87.29% and 57s session duration.
That`s why why bounce rate, average session time and % new sessions are not always the most reliable signals to recognize referral spam.
Another way to recognize referral spam in Google Analytics is by checking if :
- the fake referrals are reported with fake or (not set) host names or
- the fake referrals have fake page titles
If we take the example of above and we add “Host name” as secondary dimension, we can clearly see that 5 of the referral websites have real host names:
Normally, as secondary dimension you should see the host names under which your Web site is reachable. But in the screenshot above, most of the host names are either (not set) or fake.
Most probably the type of referral spam we`re dealing with here is ghost referral spam. But unfortunately, it`s difficult to know for sure.
And a third way to recognize fake referral traffic is by simply visiting the referrer site.
In most of the cases you will land on a referring page that doesn`t have your backlink or you`ll be redirected to another domain.
Impact of fake referrals to your Google Analytics account
The big problem with fake referral traffic is that it`s skewing your website statistics.
It`s especially annoying on new websites, which don`t yet receive a lot of referral traffic. Let`s see an example of a new website affected by referral spam:
This website has received 421 visits (sessions) from referrals during the selected date range. Unfortunately all of these sessions are sent from fake referrals. As a result all Avg. behavior metrics for referral traffic are not accurate.
Moreover, in the case of a new and unpopular website, the fake referral traffic could quickly become the main acquisition channel:
In the eyes of an inexperienced website owner, this report says that the most effective acquisition channel is referral (which is not true).
The website avg. session duration, avg. pages / session per view and bounce rate are also inaccurate because they are heavily impacted by the metrics from the fake referral sessions.
And there’s more: spam traffic does not convert on any of your goals, therefore the conversion rate will be also impacted.
As a result, referral spam traffic not only inflates your session count, it also dilutes some of your key performance metrics, such as engagement and your conversion rate.
In summary, if your website is heavily impacted by spam referral traffic, your Google Analytics data is not trustworthy.
And, as we know, many important marketing decisions are taken based on web analytics data, inaccurate date could lead to bad business decisions and create an untrue idea about your website performance.
Now, after we’ve leaned what are the biggest problems that spam referrals can bring to your website, it’s time to see how to deal with this problem.
In my next article, I’ll give 5 solutions to remove referral spam from Google Analytics that will hopefully help you to deal with referral spam once and forever!
Meanwhile, feel free to share your experience with ghost referrals and spammy crawlers in the comments below!