What are segments?
If you’re just getting started, think of segments as a temporary lens that you can apply to your Google Analytics reports. They allow you to focus on the important sections of your audience in order to perform in-depth analysis and make comparisons between your different segments.
For example, if most of your business comes from people located in Australia, then you can start by creating a segment that just includes people who are physically located in Australia. This way when you look at your Google Analytics reports you will only see data for your Australian audience members.
Updates to segments
Segments in Google Analytics have now been updated to give you more flexibility in the way you create them. You can continue to create visit-based segments, like the Australian example above, but now you can also create segments based on users. This means you can create a segment that contains people who have engaged with you and even people who have engaged in a particular order, for example, if someone has viewed a product page and then viewed their shopping cart.
The new segments are also ‘Universal Analytics ready’. This means that once you start measuring people across multiple devices you will also be able to include them in your segment. For example, you could create a segment for anybody that has viewed your website on the desktop and then engaged with you on their mobile.
Key changes to segments include:
- Segments are now represented by donut graphs
- Ability to navigate between ‘All’, ‘Built-in’, ‘Custom’ and ‘Starred’ segments
- Options for changing the way you view segments
- Ability to search for particular segments
- Drag-and-drop interface
- Pre-defined custom segment options (including ‘Demographics’, ‘Technology’, ‘Behavior’, ‘Date of First Visit’, ‘Traffic Sources’ and ‘Ecommerce’)
- Visit and user-based segmentation options
- Condition and sequence-based configuration
Visit-based segments will continue to be available within the new segmentation options. They allow you to define particular sections of your traffic based on the single visit that has occurred.
The Australian traffic example that we looked at previously is a good example of a visit-based segment. We are including all visits where the person was physically located in Australia when they accessed our website or app.
We could also make a visit-based segment that only included visits where a conversion occurred. This allows us to focus on people who have come into the website and converted, regardless of whether they have been to the website before or not.
User-based segments allows you to go beyond any single visit to your website in order to define your segments. This means you can combine particular interactions or audience attributes even if they occur in different website visits.
For example, you could see people who viewed your services page and then your contact form even if these pages were viewed by someone on different days. This enables you to better understand the true behaviour of people on your website.
Picking up on the first segment we looked at where we included our Australian visits, you could now create an Australian user segment. This would mean that anybody who visited your website while located in Australia would continue to be included in the segment even if they later traveled to Canada. This allows you to be more flexible in the way you define your target audience. It also takes into account the fact that most of our visitors will probably come back to our website a few times before converting. Let’s consider a visitor segment that focuses on conversions:
- A visit-based segment focusing on conversions will only look at each user’s final visit where they actually completed the conversion. You will therefore not be able to easily see, for example, what traffic sources these conversions ultimately originated from.
- A user-based segment focusing on conversions will look at all the visits from users who ended up converting. This would make the analysis of which traffic sources are sending qualified visitors much easier.
It is important to remember that in order to take advantage of this, Google Analytics still needs some way of tying together a person’s multiple visits to know that they’re coming from the same person. This is accomplished through cookies so if a person switches devices (or even browsers on the same device), Google Analytics will not be able to know that the separate visits came from the same person unless you implement additional user tracking with Universal Analytics.
Segmentation using Conditions
Conditions allow you to define criteria based on either visits or users. Conditions allow you to exclude or include traffic based on either visit-based or user-based criteria. You can also layer multiple conditions that need to be met for the advanced segment. For example you could create a segment where people need to be physically located in Australia and have converted in order to have them included in the segment.
Conditions allow you to apply ‘OR’ and ‘AND’ statements to the way you match sections of your traffic. In the example above we would match people who were from Australia AND converted on our site. Conditions, unlike sequences, which we will look at in a moment, don’t require interactions to have occurred in a particular order. This is where sequences fit in.
Segmentation using Sequences
Sequences allow you to match users based on a particular order of interactions. For example if you wanted to see people who came to your website from your email newsletter and then later converted. You could create a user sequence that show people who meet this criteria even if there were days (or weeks) between clicking the link in your email newsletter before they completed your goal action.
Another example of a sequenced filter is one that isolates users who went from your home page to a key selling page. If the conversion rate for those users is higher, it is worth trying to drive more visitors from your home page to this key page since seeing it after the home page has a positive effect on conversions.
If you have been using segmentation, then the updates to Google Analytics segments open up new opportunities for understanding the behaviour of people as they engage with you across visits – and potentially even devices, if you implement Universal Analytics.
For those of us just getting started, segmentation allows you to focus on the most important sections of your visitors. A great starting point is to think about your most valuable audience members or your target audience for your marketing campaigns and create segments that only include those people. You can then jump into the reports to get a more detailed view about how they are finding you, what they look at and if they convert.