Google Analytics Segmentation

Google Analytics Segmentation

Published by

October 29, 2017

What is segmentation? Let’s start there, though if you need to leap back to Google Analytics for E-Commerce (our first post in this series), I’ll wait. Also just a note Google Analytics Segmentation is a key feature of Google Analytics but is of course available in other web analytics tools as well.

FREE Download: Save yourself from the missed steps and trial + errors and get the checklist we follow and keep in mind when creating custom segments in Google Analytics.

Segmentation means isolating a part of a sample of some data and observing it in relation to a whole or other segment. For example, you could take a set of people visiting a web site and then select only the segment coming from a certain geographic area, or of certain age group. This segment’s behavior may or may not be similar to the whole set.

Segmentation allows us to understand the different types of people who come to an e-commerce site by key characteristics, like their location, age, and even interests. Those insights let you deliver targeted messaging that speaks specifically to the needs and wants of each segment, create personalized user experiences, design marketing and ads to be more effective and efficient, and so much more.

The object of the game is to isolate a segment of your visitors and see how their behavior differs (and by how much) from the entire visitor population, or how their behavior compares against a chosen metric.

For example, if we conduct customer research and find that one group, despite the best efforts of the marketing department, shows little interest in buying the product (while everyone else loves the product and can’t get enough), we can segment the non-buyers out and take a closer look at what is driving their decision.

Or, we can simply stop wasting our precious marketing budget trying to reach a segment that won’t budge – because segmenting revealed a core product-market misfit – and devote more resources to expanding the segment that already loves the product.

Online marketing allows for quick, virtually error free and comprehensive data gathering for segmentation purposes. The segmentation tool in Google Analytics is rightly considered one of its best features.

How Google Analytics Segmentation Works

Google Analytics makes segments very easy to add on the Reporting tab of your Google Analytics account. Once you add a segment, existing data in Google Analytics will automatically be applied. Your segmented report will be instantly visible and, by default, compare the segment to All Users.

Let’s see how it operates.

Google Analytics Add Segment
Audience Overview screen with ‘Add segment’ indicated

Clicking on Add Segment opens a window where you can select distinct criteria, or groups of visitors, that the segment will include. The more criteria you add, the narrower the segment becomes.

Google Analytics Segment Definition Screen
Google Analytics Segmentation Definition Screen

In this example, you can see that this segment has been set up to show all female visitors aged between 18 and 34 coming from United States.

But we can narrow this down even further using the self-explanatory options, and you have the choice to create custom segments or use predefined ones.

Of course, there will always be visitors who have accessed the site anonymously, using incognito mode on their browsers, so some of their data will be unavailable.

Other ways to segment include looking just at returning visitors, or new visitors, or visitors who have already made a purchase. You can also filter out visitors coming from organic search, or direct visitors or any combination thereof.

The data is all there, in real time, and yours to explore as you please. Google Analytics applies the segmentation to existing data, retroactively, unlike the filters that you apply at account, property or view level. For more on Filters, read our in-depth article.

Where does all of this information on your visitors come from? Google. Google uses third party cookies to track users and derive their interests, ages, genders and the other ‘Big Brother’ stuff you find in Google analytics. This data is used by Google Analytics to give some information on the users and is shared anonymously. No individual identification or personal data can be derived or accessed.

How to Use Segmentation for Conversion Optimization

How can you put all of this data to use?

You can use it as a diagnostic tool to see if some segments are failing to convert, or even to reach your website, at a higher rate. You might uncover a problem you can fix to win those customers back.

For CRO, we compare segments to each other, or to the average visitor, and check for inconsistencies or differences in behavior – then we look for a logical explanation for them. If we can find a good explanation, or infer one from the behavior, we can devise a remedy to drive the customers to convert at your average rate.

For example, if you find out that the teenage girls segment has a lower conversion rate than teenage boys, then you might change value proposition or call to action to appeal more to the lower performing group – or, a more advanced approach would be to create variants of your content that would appear only to their appropriate segments.

Another common example is if you realize, through using Segments in Google Analytics, that visitors speaking Spanish make up a significant portion of your audience, but have a high bounce rate. Then you might consider creating a version of your site in Spanish to better enable them to purchase.

Explore one segment at a time to identify potential problems and opportunities. Then use those insights to come up with remedies.

Segmentation and Marketing

Of course, most of us are somewhat familiar with segmentation from its uses in marketing.

Using segments, we can analyze the behavior of returning customers relative to new visitors, or relative to each other, segmenting by devices, interests, geographical or demographic groups. Targeting returning visitors with relevant and engaged content, that is tailored to the actions they previously took on the site, will certainly find an audience more likely to react to the content and convert.

Let’s look at how this works in practice:

This is real site data from the Google demo account, which I’ve set up to mimic an e-commerce store. I created two custom segments – female and male visitors.

Google Analytics e-Commerce Overview
Overview of an eCommerce report with visitors segmented by gender

According to the data, we see that female visitors are more likely to convert with a conversion rate of 1.94% vs 1.19% for male visitors. And, once they convert, they spend more money than their male counterparts, with average transaction value of $156.00 vs $102.83 that male visitors spend on average. Thus, 286 female visitors who made a purchase spent almost as much as 490 male visitors. However, only 14.733 of the 81.156 registered sessions (or 18%) were initiated by female visitors, which we can see in the audience overview, with the All Users segment added:

Google Analytics Male vs Female Audience Report
Audience report with both segments

If 18% of the audience accounts for close to 50% of all revenue, we can conclude that this store should try to attract more female visitors!

With 50,000 sessions and an average conversion rate of 2% it would mean some 1,000 new customers. This would net an increase in revenue of $150,000, if the average value of transactions for female customers remained the same.

This is one example of how to use Google Analytics segmentation analysis to inform and direct marketing decisions. And this is why segmentation is one of the most valuable tools in Google Analytics.

FREE Download: Save yourself from the missed steps and trial + errors and get the checklist we follow and keep in mind when creating custom segments in Google Analytics.
Series Navigation<< Google Analytics for e-CommerceGoogle Analytics Funnels and How To Structure Them for e-Commerce >>

Published by

Edin is a Senior CRO Consultant. Edin is into Google Analytics and testing (any A/B testing tool really) and likes to write about it. You can follow Edin on Twitter.