Search Marketing Insomniac…The Blog of Adam Henige

Of all the additions to Google Analytics within the past 12 months, none have made waves with analytics professionals more than the segmentation tool.  Obviously, this provides a million different ways for users to slice up data, but for beginners (and I run into this regularly) all of the possibilities can be confusing.  So now that you can create all of these segments, what are some practical ways to use segments?

One of the first things I point out to people is the ability to understand user intent.  What do people want when they come to your site, and how well are you connecting with them?  This is good information to know, and segments can give you some great information in this regard.  I will use the Netvantage site as an example.

We offer three primary services, obviously - SEO, paid search management and web analytics.  We target a variety of terms in each category from an SEO standpoint, and with that comes an assortment of long tail terms as well.  Our paid search management page will bring in traffic for terms like “paid search management“, “ppc michigan”, etc.  All of these terms imply a basic need for paid search management and it’s valuable for us to know if our landing page is engagin our visitors.  So here’s what we do in such a situation.

As you can see above, I’ve added three variations of terms that land on our paid search page.  This segment will cover all users who have landed on our site from a search containing the terms “paid search”, “ppc” and “adwords”.

Applying this segment will allow me to look at these users from an aggregate as well as a refined level with ease.  At the highest level I can look at the segment’s overall bounce rate and begin identifying if there is room for improvement.  If so, I can look at the keyword report and start breaking it down further to individual keywords and see if there are issues between sub segments - which may require me creating new segments.  If at first glance of the keyword report I notice that Adwords related keywords are underperforming, I may break that segment out individually and compare it to the other subsegments.

In the case that we have poorly performing visits for Adwords related terms, then it would make sense to start experimenting with the page to speak more directly to users who may not be familiar with terminology like “PPC” and “paid search” (a distinct possiblity).  Again, these people have the same intent (a need for our services!) but we may not be addressing them correctly.  Using Analytics, however, can help us uncover these missed opportunities and push more users through the conversion funnel.