Yesterday I discussed the advantages of measuring positive actions which visitors to your website might take, by setting up “conversions” (or “goals”) in your visitor analytics software. As the second of the three aspects of website traffic analysis which I think you should look at, I’d like to suggest a related area which is possibly even more fundamental, and that’s only looking at the relevant visitors to your website in the first place.
What do I mean by this? Well, many of you will be responsible for marketing in, say, just the UK and Ireland. So you’re simply not interested in any visitors to your website from anywhere else in the world. And if you look at the location of your visitors (something you can do right now), you’ll see that quite a proportion are from overseas. They’re simply not relevant.
Or you might say: “I only want to look at visitors to my site who have never been here before”, something which is particularly interesting when you’re comparing the quality of the traffic from advertising. And you’ll almost certainly want to focus on visits which seemed to be fairly engaged with the site (i.e ignoring anyone who took one glance and left).
All this is done by “segmenting” the website traffic in your visitor analytics software. It’s pretty straightforward to do, and you can do it yourself without having to touch the website or add code. Here’s how “advanced segments” work in Google Analytics.
I get the privilege of working with the Google Analytics for most of the companies who use us to manage their Google AdWords campaigns, and for most of them, I set up a segment which looks something like this:
Visitors who had not been to the site before, AND who were from Europe, AND who spent at least 20 seconds on the site OR who looked at more than one page.
We call these the “quality visitors”. From this, when analysing the source of visitors, we might find that a link on one website sent 23 people, of which 8 were “quality”, whereas Facebook sent as many visitors but only 3 were of any use. Or whatever. This is the real insight which you need into your data – and it’s something you can set up today.