Measure Users, not Sessions – Powerful Segmentations in Google Analytics

Show all Close
November 25, 2013

As many analytics experts point out – segmentation is key to making sense of your data and coming up with actionable insights. Hence, Advanced Segmentation, which lets you slice and dice your data on-the-fly, is one of the most useful features in Google Analytics.



Until recently, Advanced Segments in GA were based on visits only. This means that we would tie conversions with behaviors that occurred in the same visit. However, we know that people’s relationships with our website are (hopefully) not just one-offs. We also know that user decisions which drive conversions may be made on the basis of several interactions. Therefore, in order to really understand user behavior, we should focus our analysis on users, across visits. This is a paradigm shift, and it underlies a whole set of new capabilities that will help you draw powerful insights to drive smarter decisions.

There’s an extensive help section explaining how to get started with the new Advanced Segments, so we’ll skip that and focus on some of the strongest use cases that we found, with instructions on how to create these segments.

MVP – the Most Valuable People on our website

Rather than just looking at discrete transactions, we might like to look into the group of users who buy (or convert otherwise) a lot over time. Depending on what defines online user value for your specific business, you can define your MVP segment as users who spent more than X amount of dollars over time (Ecommerce filter by user), contribute content frequently, or completed a specific goal (Advanced conditionuser filter).

How much of the business are they driving (in comparison to single-purchase users)? What was their first purchase, which turned them into a long-term customer?

Frequent visitors

Another possibility is to filter for people who visit your site frequently. These are people that know your brand well and probably trust it. Pay special attention to the frequent visitors who don’t convert: use a Behavior filter with Visits > x condition, and an Advanced Condition with 0 goal completions for the user.

What brings them to your site over and over? Are they searching for specific brand models and not finding them? Do they look at specific products, and perhaps buy them elsewhere for a lower price?

Cohort Analysis

In a previous post, we described how to use cohort analysis for measuring user retention / churn. This involved keeping a custom variable with the date of first visit for each user. Now this comes out-of-the-box, so no special coding is required – simply use the Date of First Visit filter type.

However, do read our previous post if you’ like to learn how to turn cohort data to actionable retention metrics.

Optimizing first impression

“You never get a second chance to make a first impression”. Use user-based segments to see the effect of landing pages over time, i.e. how the first page that a visitor ever sees on your website is likely to drive conversions, even if they occur in subsequent visits.

For this, create a segment using an Advanced Condition with two filters: include users who completed a specific goal, and visits where Visitor Type = “New Visitor”. Compare this to a similar segment, but this time excluding users who completed the goal. The Landing Pages report will show you the “first impression” pages that lead to conversions, vs. the ones that don’t.

You can use the Google Analytics Solutions Gallery to find more ideas for Advanced Segments, Custom Reports and Dashboards that provide valuable information.