Go to Blog

Was It Good for You? How to Use Analytics to Measure User Experience

Marianne Sweeny

I am tempted to think that the arrival of Ray Kurzweil and his singular focus on the Singularity is what brought out the “sensitive New Age geek” in Google. However, the touchy feely user experienced-focused Search Borg came on the scene long before Ray took the position of Director of Engineering at Google with a project to teach computers to write poetry. If nothing else, we will have something to look forward to at poetry slams.

It was that cuddly Panda update that selected the inferiority, complex-ridden user experience community as prom kings and queens of SEO, while the thriving community of Web professionals dedicated to reverse engineering Google’s engineering stood by in disbelief. And, it only got worse from there with future “enhancements” that focused on content freshness, aboutness and other “nesses” that brought content strategy to every interactive agency service offerings and a spotlight on beleaguered Web writers.

There are many reasons to mourn keyword optimization. Those phrases and numbers were the building blocks for what came to be highly accessible, yet wholly dysfunctional, SEO.

As our users got smarter, lazier and more distracted, it became clear that the match game of query to document was doomed to fail. Let's give ourselves credit that we did not give up this binky easily. No, it took three major algorithm changes to pry our fingers away from keyword ranking.

So, let’s start by agreeing that it might be a good thing that SEO has become something you need to know something about in order to succeed. And, figuring out why users do something is much more rewarding than link bombing blog comments. Optimizing the experience will be rewarding and can be fun. And, for you metrics mavens, here are a few data points to keep an eye on.


How many unique visitors are coming to the site and where are they coming from? If referrals and social are sending more traffic than search engines, you might want to revisit those Title tag and Meta Descriptions to entice more clicks.

 user experience visitors chart

The Visitor flow diagram found under Audience in Google Analytics reveals the dominant traffic lanes through your site. You can display as many interactions as you would like to display with the ability to track by location, event or medium. Important UX data to look for is: where do customers drop off (Weak call to action, anyone?) and the pages others move on to. This is valuable data in developing a page layout that encourages visitor interaction for a low bounce rate and directs traffic where you want it to go.

visitor flow diagram


Content seems like a no brainer. Pick the top pages and leverage the heck out of them. Deconstructing user experience means digging a little deeper. Top Landing Pages shows that #10 has an average time on page of 8+ minutes. A visit to Top Content under Behavior reveals a bounce rate of 84%. Looks like they are reading the page and that the call to action or engagement opportunity presented is not interesting. Further support for this theory is found in the 80% exit rate. This is a page that needs some layout or related linking help.

 UX landing pages


Event tracking is user experience crack cocaine. It enables us to see when visitors abandon forms. This tells us how many form fields are too many for that white paper download. Measuring other user interactions, e.g. scrolling, can be done through more sophisticated event tracking configurations.

These are just a start on how you can measure the quality of user experience on your sites using the same data that Google uses: click-through (Do they select your page from the results?), bounce rate (Do they do anything when they get there?) and conversion (Did the page solve their information need?). And yes, I am developing a UX dashboard for Google Analytics with these and other players yet to be named.

While metrics are much like runway lights in illuminating potential problems, they are data points, not information. They tell us what, not why. That is our role: to figure out the why.

Author bio:

Marianne Sweeny is a Search Information Architect at Portent Inc. She is passionate about optimizing the user search experience on the Web or inside the enterprise firewall. Marianne’s last article for SEMrush was "Enterprise Search: Where You Get to Be Larry Page...No, Really!"

Like this post? Follow us on RSS and read more interesting posts:

Marianne Sweeny is VP of Internet Marketing for Strategic Edge Partners, a medical marketing agency. She is passionate about optimizing the user search experience on the Web or inside the enterprise firewall. Marianne’s most recent article for SEMrush was entitled: "SEOs, Guns & Religion."
Share this post


2000 symbols remain