How to Use Google Analytics Advanced Segments

Google Analytics provides powerful but under-advertised features for grouping your visitors into buckets based on features of their visit. If you don’t know much about these yet or how to use them, read on.

What are Advanced Segments?

Google Analytics provides some very intuitive default reports. From simple things like being able to see your total number of monthly visits, to being able to track your conversion rate for each mobile device being used to access your site. It’s that latter ability to compare Scenario A versus Scenario B that provides the power of an analytics package, and Google Analytics lets you do this in custom ways.

A few definitions:

  • Dimension: A way that data might be broken down. Think of rows in a table. Dimensions might include day of the week, device, screen size or landing page URL.
  • Metric: The columns in your table. The numbers you’re interested in or want to track. Visits, page views, conversions and sales value are all typical metrics.
  • Segment: A grouping of visits based on criteria of those visits. This acts much like a filter, so your metrics will only include those visits for which your criteria evaluated to true. For example: only visitors who landed on the homepage, or visitors who spent less than 2 minutes on the site.

Google Analytics provides a tool called Advanced Segments that will let you pick and choose your own criteria from a mix of dimensions and metrics. You can apply more than one segment at a time to compare and contrast metrics between multiple groups of people.

Why are Advanced Segments Important?

It’s difficult to take actionable insights by looking at your data on aggregate. You’ll see a lot of totals and averages without understanding what separates your best customers from your worst.

Consider trying to work out the paths through a small B2B website. You want to know which pages are important to helping persuade your visitors to convert.

By applying two segments – visitors who converted and visitors who didn’t – you can compare the typical visits to each page of the site by the two groups. If you see a page that is more frequently visited by people who go on to contact you, then that gives you a starting point.

Note that correlation doesn’t imply causality, but at least you know which page to start looking at to try to figure out what’s going on.

Creating an Advanced Segment


Hit the “Advanced Segments” link at the top of the page and choose “+New Custom Segment” You will see a selection of options like below:


First name your segment. A segment won’t save without a unique name, and being really descriptive will help you to know what it is when you come back to it 12 months later.

You have a drop-down containing almost every dimension and metric available within Google Analytics. Dimensions are in green and metrics are in blue. You can search for the item you want here. You can choose whether this field should include or exclude matching visits.

You will have a drop down menu to the right of the dimension/metric that will have a variety of text matching options (including regular expressions) if you’re segmenting on a dimension, or number matching options if you’re segmenting on a metric.

You can create several of these individual fields, and put them together with OR or AND statements. Be careful with the logic here, and keep in mind that if you choose to exclude items, then your OR and AND statements may need to be switched from what you expect.

The best way to construct a complex segment is to draw it as a Venn diagram first. Because each field applies totally separately to your dataset, you can use OR to tell Google Analytics to get any visits that match anything in your diagram, or AND to get only those in the overlapping sections.

Let’s look at a couple of important segments to get you started.

Segment Example 1: New vs. Returning


This segment filters on dimensions only to restrict your view to only visitors who have never come across your site before. As with all web analytics this is inexact, but it’s much closer than the built-in New vs Returning dimension.

  • Include visitors who are marked as New. These people have no cookie on their machine to indicate that they have been to your site before.
  • Exclude visitors who have come direct to your site. These people know your URL already and should be treated the same as returning visitors, when analyzing your online marketing efforts.
  • Exclude visitors who have searched for your brand. Brand searches are not new visitor acquisition. Whether SEO or PPC traffic it should not be factored in.

Now create the exact opposite segment – one that excludes new visitors and includes direct traffic and brand searches. Compare the two groups. How/where do they behave differently? How can your site accommodate both sets of users?

Segment Example 2: Converted vs. Non-Converted


The behavior of visitors who purchased versus those who didn’t. This simple pair of segments will need to be slightly different if your site is e-commerce or not, because it will depend if you’re using goals or e-commerce tracking. For the former, your segment should be based on goal completions > 0. For the latter, use Transactions instead. Make sure you create the opposing segment also. Goal completions = 0.

Look at the two groups in your data and look for patterns and key differences. Specifically, you’re looking for differences you can’t easily explain:

  • Higher pages per visit for converted users? That’s because there are several pages in your checkout.
  • High proportion of traffic from a specific referral in the converted segment? That’s worth investigating.

What Can Advanced Segments do for You?

Expand your horizons. Blow your mind. Et cetera. No really, they’re good.

Advanced segments are the tool that expands the repertoire of possible analysis within Google Analytics up to near-unbelievable levels. If you want to know anything about how different types of users behave on your site (e.g. from different traffic sources, based on user personas, etc) then segments are the way to do it.

Become comfortable with setting them up and applying them to your data. Learn regular expressions (advanced text matching) for even more power. Most importantly, just dive in.

Related reading

guide to call tracking
Converting custom Using analytics to optimize sales funnels for new and returning customers
set up google analytics annotations for google updates
Three tools providing actionable competitive research insight