Dealing with Ghost/Spam Traffic in Google Analytics
Over the past few months, a previously minor issue with Google Analytics has exploded into a huge problem. This GA referral spam, or ghost traffic, takes the form of fake traffic to your website. The apparent objective of these spammers is to get people to visit their websites, based on the fact that they show up as a referring website in GA results. The worst of them manage to show up in reports for almost every website that uses Google Analytics.
This isn’t even real traffic that hits your servers. Instead, it appears in your reports by tricking Google. We have seen this traffic for a few years, but until this past spring it rarely made up more than a small percentage of total traffic. At past levels, it was a very minor annoyance, and easy enough to ignore. But now, the problem has escalated to the point where over 50% of traffic to some sites is fake. And this gets in the way of using your data to make intelligent business decisions.
In general, Google has been fairly quiet on this issue, although they did recently add in a checkbox that allows you to filter out some of the junk. Unfortunately, they have not done a great job of keeping up with the spammers, although in recent weeks there are signs that they are getting better.
In June, a new spammer joined the fray, “trafficmonetize.org”. For a few days in mid-June, traffic from this site exploded like nothing I’ve ever seen, for most of the sites that we monitor. But then, on June 19th, it appears that Google cut them off. In addition, it seems that other referral spam slowed down at that point (although it is still too early to say for sure). So for now, the worst of this problem may be behind us.
Luckily, we are not completely at the mercy of Google on this issue. In addition to using the checkbox mentioned above, there are also 2 more technical options available to reduce, and nearly eliminate, this traffic. One method involves creating filters, which will eliminate all future junk traffic from hitting your reports. The other involves creating segments, which allows you to view a clean version of your data, while segmenting out the junk.
I won’t go into all the technical details here, as others have done a very thorough job already. Ohow.com has published this article on filters and this one on segments. And for a slightly different perspective, check out the analyticsedge.com filters and segments articles.
Not sure which you need? Ideally, you should set up both. Segments will allow you to get clean reports from past data, but take a little extra time whenever you want to run reports, among other limitations. Filters can take a little longer to set up, and will not clean up your historical data, but they are more efficient going forward.