CEM Analytics: it’s not all about the UI



May 24, 2011 | 3 min read

May marketing madness usability week, post #24

By Gary Angel

Co-founder, President & CTO, Semphonic


CEM Analytics tools provide a field-day for classic Web and UI analysts. To an analyst, being able to see what happened INSIDE a page, how users scrolled, where they looked, how long they spent, feels like being kid on Christmas morning with all those un-opened presents under the tree.

It’s great, but there’s another side to CEM Analytics that I think often gets ignored and should get more attention both from the tools and the practitioners.

Two-tiered Segmentation

In Semphonic’s practice, we focus heavily on segmentation. We believe that every Web analytics metric and KPI should be placed in the context of two simple questions. If you tell me that traffic on the web site was down, my first question to you is, With whom? If you answer Customers, I’m going to ask what type of Customer visit  is it Customer Support (that might be good) or re-Purchase (that would probably be bad) or something else entirely. Until I know the Who and the What, a metric simply doesn’t have any meaning. We call this a Two-tiered Segmentation and think it’s ideal for Digital Analytics.

Detailed CEM Data

What’s this got to do with CEM Analytics? Tools like Contentsquare increasingly let analysts see CEM data by segment. That’s great, but there’s another aspect to CEM data that’s almost completely unexplored. You see, the hardest aspect of Digital Segmentation isn’t dreaming up segments, it’s figuring out how they can be inferred from Web behavior. The more behavior you have, the easier it is to infer the segment. Detailed CEM data gives you more behavior to choose from.

Visitor Cues and Patterns of Behavior

In traditional Web Analytics, we know whether someone viewed a page. In some cases, we know how long they spent there. That’s about it. With CEM, we can potentially know quite a bit more.

We call behaviors that tell us something interesting about a visitor and the patterns of behavior that let us categorize them into segments we call signatures.

CEM data provides a multitude of potentially interesting cues.

Naturally, content interest based on hover time is one of the most consistently important cues. Did a visitor take in a promotion or is that promo box just wasted real estate? Categorizing your visitors by their promotion receptivity is a fascinating segmentation opportunity. There are also strong targeting cues in hover times. Knowing visitors who at least considered an offer/product is great for potential re-targeting and personalization.

Target and Personalize

When building traditional visitor segmentations, we often create a set of style variables based on the visitor’s navigational preferences. Is the visitor a user of internal search or top navigation? Does a user take advantage of video or just text? Where does the visitor like to start his/her session? These styles help us figure out potential targeting and personalization options.

CEM data extends the possibilities for style segmentation. One of the purposes of style segmentation is to help us identify a visitor’s skill level, particularly on operational and transactional Web sites. CEM data provides a wealth of cues based on how quickly a visitor navigates, the methods they choose, and the focus they display per page. Understanding all of these opens up new vistas and opportunities into identifying your visitors and appropriately personalizing their experience.

New Horizon For CEM Analysis

Of course, CEM solutions haven’t been generous about opening this data up; even less generous than their miserly cousins, Web analytics tools. That’s changing slowing in the Web analytics world, and it’s a whole new horizon for CEM analysis.

It’s a whole new horizon for CEM analysis.

Every advance in data collection opens up new opportunities to enrich our understanding not just of our sites, but of our customers. That, after all, is what Customer Experience Management is all about. Segmentation, not just of traditional data to see UI patterns, but of UI patterns to identify important types of visitors, is a whole new vista for Customer Experience Management.


About the Author

Bringing over twenty years of experience in decision support, CRM, and software development, Gary co-founded Semphonic and is president and chief technology officer. He’s responsible for leading Semphonic’s development of Web analytics and SEM decision making tools for web marketing professionals. In addition, he helps companies like WebMD, Intuit, American Express and Charles Schwab maximize their web channel marketing through intelligent use of Enterprise Web Analytics.

Gary has published articles on Web and SEM Analytics in DM News, American Demographics, CRM Guru, CRM Buyer, IMediaConnection, Business Geographics and Business Insurance. He graduated, with honors, from Duke University and lives in San Francisco with his wife and two young girls.


Clicktale was acquired by Contentsquare in 2019. Since then, tools and features mentioned in this blog may have evolved. Learn more about our Digital Experience Analytics Platform.