Spam within Google analytics is a real pain – it does make your site look really well used but you’re kidding yourself. It’s the reason you’re getting so many visits from tiny towns in the middle of Russia and the Philippines.
I’ve just done some work for a client that involves getting them some closer to real numbers from Google Analytics and I found it really useful to refer to
Especially useful was they way they used Segments to test and view the percentage of traffic which is spam, before creating filters to remove it from this point on.
On this site only about 10% of all traffic is not some form of spam visits either trying to drum up business for their wonder SEO service or sell me something.
For intranet and digital analytics teams it is worth thinking about the same sort of issues with your analytics tools. Are the numbers you’re looking at real – how could you confirm if they are real people or some other artifact showing up in the data puke your analytics platform shows you ?
Have you tried
- Looking at traffic during a holiday period ? Is their still traffic being reported ? If so try to work out why ?
- Comparing what you know about your network environment to what you see with analytics ? In one case a client that couldn’t track username thought the IT team were huge users of the intranet, when they didn’t use it at all. The cause, IP addresses were used to identify which teams were using the intranet and anyone who logged in remotely got an IP address that had been labeled IT team.