How to use custom segments in Google Analytics to boost conversions
A Step-By-Step Guide to Custom Segments in Google Analytics:
Custom segments are one of my favourite features of Google Analytics because unlike goals they work retrospectively, and they offer users a multitude of options, including creating audiences for remarketing purposes. This flexibility means custom segments can be incredibly powerful in uncovering barriers to conversion, identifying new opportunities and targeting valuable visitors via remarketing.
In this post I will show you how to get the most out of custom segments so that you can better understand your visitors, improve targeting, reduce barriers to conversion and automate funnel visualisations. Custom segments are a must for anyone wanting to optimise their site because they are a simple, but effective means for gaining more insights from Google Analytics.
1. Plan your segments:
Before creating any custom segments in Google Analytics it’s worth considering why we segment visitors.
Improve ad and content targeting:
Segmentation helps us better understand our target audience by breaking them down into smaller, more homogenous groups. You can focus on the specific needs or interests of valuable groups of customers who are connected by some common characteristic or behaviour. This allows us to refine our message or the content we generate to better engage with key groups of users and improve our conversion rate.
In Google Analytics, custom segments enable you to create audiences for remarketing purposes by integrating with Google Ads or Google Optimize360. For PPC campaigns this is an important strategy for targeting users who don’t convert.
More effective retention strategies:
Custom segments enable us to create more focussed strategies for customer retention. It allows us to group customers or visitors according to how frequent they visit our site, purchase intent, average basket value and life-time value.
For example, we may wish to personalise content for frequent purchasers to reward them for their loyalty or re-engage with customers who have not purchased for some time with a time-limited promotion.
Offer a superior customer experience:
Personalisation based upon custom segments can be used to deliver product recommendations, abandoned basket emails, wish list reminders and other targeted messages to improve the user experience. Such interventions based upon segmentation can enhance the user experience by helping the visitor achieve their goals more quickly and with less friction.
The use of buyer personas is a common method for bringing customer segments to life by creating a semi-fictional customer to consider when making decisions about changes to the user experience. Buyer persona templates can be very effective at communicating key aspects of the user to IT and operational teams and help to put the user at the heart of decision making. I have previously outlined a process for using buyer personas to improve conversions.
Identify pain points in the user journey:
Google Analytics is especially good at creating custom segments based upon behaviour so that we can identify potential pain points in the user journey. By creating segments based upon steps in a user journey we can identify drop-off rates and focus attention on reducing friction and pain at these key points in the user experience.
Identifying potential new opportunities:
Analysing conversions and other key metrics by custom segments may also identify new opportunities for meeting unmet needs. For example, finding a substantial segment of users who are frequently searching for a product that is not currently available on an ecommerce web site may provide an opportunity to grow your product range.
3. System Segments:
Google Analytics comes with numerous system segments as standard. These cover things like device category (i.e. Mobile, Desktop and Tablet), purchasers (if you have ecommerce enabled), traffic source (e.g. Organic, Direct and Paid), new visitors and multi-session users.
System segments can be very useful when you want to break down users into more meaningful groups. However, they don’t reflect the unique nature of your site or your business model. That’s why custom segments are so valuable because you can tailor this to your specific requirements.
4. How to create a custom segment:
When logging into Google Analytics go to a standard report, such as Behaviour > Site Content. Select a suitable time period to gather enough visitors to make it worthwhile segmenting your data, say a month or less depending upon your traffic volumes.
Near the top of your screen and opposite ‘All users’ you will see ‘+Add Segment’.
Click on ‘+Add Segment’ and when this opens the list of existing segments click on ’+NEW SEGMENT’.
You are now presented with a menu of choices to define your segment.
The areas to note are:
- A: Segment name – make it descriptive and consider a prefix if you work for different clients.
- B: Standard segments include demographics, technology and enhanced ecommerce.
- C: Advanced segments cover conditions (e.g. an event) and a sequence (i.e. a funnel). You can combine a standard segment with an advanced segment criterion.
- D: Summary – this displays the percentage and number of users and sessions accounted for by the segment. Be careful not to be so selective that your segment becomes too small to be of any use.
Before creating a demographic custom segment do some research about your target audience and check out the demographics report in Google Analytics. For example, GA indicates that around two thirds of our site’s visitors are in the three youngest age groups. Some of the most engaged users are in the technology/technophile’s affinity category. The in-market segment which appears most aligned to our recent new content would be business services/web services and business software/productivity.
As a result, I decided to create two demographic custom segments . The first one we called Young Tech Savvy Business Services/Tech and the second Young tech/software productivity. For the first group, I selected:
- Age: 18-24, 25-34 and 35-44.
- Affinity Category (reach): Technology/Technophiles.
- In-Market Segment (Intent): Business Service/Business technology/web services.
You won’t reduce the number of visitors by not selecting an attribute. It’s worth selecting and un-selecting some demographic attributes to see what effect it has on the size of your custom segment. I then created a second segment by selecting the in-market segment of business software/productivity rather than the business technology/web services one.
If you have the traffic volume, you might also consider selecting location for a more geographically targeted audience. This allows you to start at the continent level or be as specific as region of a country (e.g. New York or California) or target an individual city. In the UK, the region only goes down to country level (e.g. England and Wales). You can also set a condition from the advanced segments, such as an event to target users who added to basket or interacted with certain content.
Once you have added your new custom segments to your tables, make sure you add the ‘All Users’ segment so that you have a benchmark to compare the performance of your new segments. Tip: always place the ‘All Users’ segment as the last segment on the right so that Google Analytics will automatically sort data by your first custom segment.
Validate your segment:
As a content publisher I wanted to see if these segments are more engaged and will they generate more income than the average user. To do this you can go to the Behaviour > Publisher > Publisher Pages report in Google Analytics. Here we can see that for revenue per 1,000 sessions our first our segment, and to a lesser extent our second segment has a much higher value.
Let’s look below at the individual pages to see what kind of content generates the most revenue. This shows that the blog post ‘How to create calculated metrics in Google Analytics’ generated a higher proportion of impressions for both the first (11%) and second (9%) segments compared to the All Users segment (4%). A similar pattern emerges with the post ‘How to create a complex funnel visualisation in Data Studio’.
This kind of analysis clearly indicates that users in the first segment are more engaged with certain content and generate more revenue than the average user. This can be extremely valuable when planning a content creation strategy to drive more engaged and high value visitors to a website.
If your website aims to generate leads or sales of some kind, you should look at the Conversions > Goal URLs report or the Ecommerce > Sales Performance report. Follow the same process to see how your segments perform against all users or other custom segments you may have created or used.
The technology custom segment allows you to analyse and target users according to the tech they are using to browse your site. The user experience of a site is very different according to whether you are on a smart phone or a desktop machine. If your website offers phone related merchandise or content, you might want to target different content according to the user’s operating system or brand of phone.
Frequent visitors and purchasers on your website are important users who you may want to target as an audience and better understand their behaviour and interests. Behaviour enables you to create custom segments according to the frequency of visits and transactions, as well as days since their last session. This means you could also target purchasers who have not returned to your site recently to re-engage them and entice them back to your site.
8. Date of First Session:
You may want to remarket to users who first visited your site during the holiday season or some other event of significance to your business. Date of first session enables you to define a custom segment based upon either a specific date or a range of dates to evaluate their behaviour or target such users more effectively.
9. Traffic Sources:
Traffic source can be useful for remarketing purposes where you want to follow up on a campaign. You might also want to analyse engagement or conversions for different email or SMS marketing campaigns.
10. Enhanced Ecommerce:
If you have implemented enhanced ecommerce on your website you can create individual custom segments for customers how have performed any action, added to card or purchased an item. This allows you to select by product and category, plus by brand and product variant if you need to go down to that level.
The advanced conditions segment can either be used as a rule for another custom segment (e.g. Event Action equals Add to basket) or as a segment in its own right. This enables you to target a very specific group of users based up almost any dimension, custom dimension or variable you decide to use. You must be careful not to make it so narrow that very few users meet the criteria.
This is my most frequently used custom segment because it allows me to automate complex funnel visualisations in Data Studio. Google Analytics goals rely on URLs to build funnels and doesn’t give you the ability to have multiple options at each step of a funnel. Sequences are much more flexible than goals because they include a ‘OR’, plus ‘AND’ for each step. This means you can create complex funnels and use a combination of different dimensions, rather than solely URLs, to define a step in the user journey.
This is especially useful when you have dynamic content or a single page application (SPA) where the URL does not change when different content is served to the user. Sequences allows you to use the ‘AND’ function to qualify which content is being served (e.g. via an element visibility event), and you can also include other micro-conversions, such as click events (e.g. input fields) in the journey.
Below is a an example of a complex funnel visualisation where both step 4 and step 8 had an alternative page in the journey. There were also three different final pages to the journey. Sequences are able to manage this complexity and use the ‘AND’ function to further qualify which journey the user is on.
Automated Complex Funnel Visualisation in Data Studio
Unlike goals, sequences also work retrospectively, and so you can analyse historic data if you have not already set up a funnel. The flexibility of sequences means that you combine lots of different dimensions, such as an element visibility event (e.g. a call to action or banner being visible on the screen), with page path and other dimensions to create more realistic funnels for complex user journeys.
The main downside to sequences is that you are limited to ten steps and so there are times when you may have to use goals instead. However, you can get around this problem by breaking up your long funnel into a series of shorter funnels. I’ve also found that if I selectively leave out some of the early steps, I have been able to successfully build conversion funnel visualisations which have more than ten steps.
To build a funnel with sequences I create a segment for each step. Start with the first step, such as the example below. Then copy the first segment, name it, and add the next step in the funnel.
Copy a segment
Click ‘Add Step’ and define the rules for the next step. You have the option of either ‘is followed by’ or ‘is immediately followed by’. I normally use ‘is followed by’ because I find ‘is immediately followed by’ on many websites with pages that frequently refresh is too limiting and may not reflect a normal user journey. Experiment with it if you are unsure.
Once you have created a sequences segment for each step in the funnel you will be able to use these in Data Studio to show conversion rates at each stage in the user journey. Sequences can also be powerful audiences for remarketing purposes to target users who showed a certain level of engagement towards a product or service by navigating to related content. This a common strategy when sites want to follow-up on pre-launch activity.
To create a new audience for remarketing, simply click on the down arrow on the segment you want to use.
This will then up a menu for building your new audience. Here you can define the duration of the segment for remarketing purposes (from 1 day to 540 days). At the next step you add your destination (e.g. Ad Words account or Google Optimize360). Unfortunately, the free version of Google Optimize does not allow you to use a custom segment for targeting purposes. Hopefully this may change soon.
Plan your segments based upon your objectives in terms of analysis, reporting and remarketing. Custom segments are a flexible and powerful feature of Google Analytics that can help you better understand and target important groups of visitors. They can help you improve your strategies for acquisition, optimising the user experience and retention.
For reporting, custom segments will enable you to automate complex funnel visualisations and instantly calculate your conversion rate . This will help reduce the amount of time you spend running Google Analytics reports so that you can concentrate on deeper analysis and enriching the data you collect in GA.
- About the author: Neal (@northresearch) provides web analytics and CRO consultancy services and has worked in many sectors including financial services, online gaming and retail. He has helped brands such Hastings Direct, Manchester Airport Group Online and Assurant Solutions Ltd to improve their digital marketing measurement and performance.
- Neal has had articles published on website optimisation on CXL and Usabilla.com. As an ex-market research and insight manager he also had posts published on the GreenBook Blog research website. If you wish to contact us please send an email to firstname.lastname@example.org. You can follow us on Twitter @conversionupl, see Neal’s LinkedIn profile or connect on Facebook.