Understand the critical distinctions between audiences and segments in Google Analytics 4 and learn how to effectively leverage them for advertising and analysis. This article walks through the creation of custom, template-based, and suggested audiences, and guides you on using segments for in-depth data exploration and user behavior insights.
Key Insights
- Audiences in GA4 are used primarily for advertising and remarketing purposes and can be created from scratch, template modifications, or Google’s suggested options; they target users in real-time and are shared with platforms like Google Ads.
- Segments, on the other hand, are used exclusively for analysis within GA4’s Explore section, allowing for retrospective data review up to 14 months to evaluate user behavior and sales paths.
- Use cases include creating remarketing audiences for abandoned cart users, generating lookalike audiences based on buyer behavior, and using segments to compare mobile vs. desktop engagement or analyze purchasing behavior within specific date ranges.
This lesson is a preview from our Digital Marketing Certificate Online (includes software). Enroll in a course for detailed lessons, live instructor support, and project-based training.
In section six, we're going to be discussing how to create an audience, why do we want to create an audience, what's the difference between audiences and segments, how to create an audience using the audience template versus using a selected audience who Google Analytics recommends, as well as creating new audience from scratch, something known as custom audiences. And then finally, how to use remarketing audiences in your Google Ads campaigns. So, why use audiences? Audiences let you segment your users in the ways that are important to your business.
You can segment by dimensions, metrics, and events. And we'll discuss events in the next section. To include practically any subset of users, you have three options for creating audiences.
You can create a new audience by defining all the parameters yourself, which would be a custom audience. You could use a template and modify, you know, the existing parameters of that template, customize it a little bit for you. Or you could select a suggested audience and use it as is, or modify the suggested audience to your needs.
We'll go into that in a moment. But first, I want to introduce another term, which is segments, right? So, audience segments are two different things, but they do have some relationship. The purpose of an audience is for advertising, targeting, and remarketing.
The purpose of a segment is for analysis, reports, and explorations. What do I mean by explorations, right? If you go to your home page and look at your menu here, you see this menu item Explore. Explore is where you can do deeper analysis and gain further insight by doing explorations, just looking at data, and analyzing the data.
And some of the things you can do here is create, you know, and we'll go into detail on explorations. We will be discussing audiences more, but you can create a sales funnel, essentially track your users through your sales process. Think of an ecommerce site where a customer may first go to a product page, add something to the cart, and then provide their shipping and billing information.
You can see the whole process and see if or where customers are falling off to help you optimize that process. Path exploration, what user journeys can you, you know, you see what your users are doing on your website, the journeys they're taking to optimize the user experience. So all of this is done in the explore section, and this is where segments are created.
Okay, so back to the, you know, the comparison. Audiences are shared with Google Ads for remarketing or retargeting purposes, and we'll look at that. BigQuery.
BigQuery is essentially exclusive for the largest Google Analytics account and provides, you know, a little bit more sophisticated data analysis capabilities. We won't be covering BigQuery here. And they're used for predictive metrics.
What do I mean by that? Lookalike audience would be one example of predictive metrics, saying, Hey, these are the people who have come to my website and purchased in the past. I want you to create an audience who looks like these people in some measurable way, so I have a better chance of finding new customers going forward with my marketing campaigns. Segments, on the other hand, are used only within Google GA4 explorations, which I was just showing you.
Audiences are evaluated in real time for audience inclusion. When you share an audience with the Google Ads, Google Ads as people are searching and using the keywords that you've identified, it will be serving ads in real time to, you know, on a basis of that to, you know, to the audience based upon the criteria that you set up as you define your audience. Whereas with segments, user sessions, any data is pulled from historical data only.
So you're not looking forward, you're looking back to see what has happened, right? Permanence, users are added going forward. There's no retroactive inclusion. It's just what's going to happen next.
I've identified this audience based on these criteria. Now I'm going to go out and look for these people in my marketing campaigns. Segments can analyze past data within a specified date range so you can look backwards, keeping in mind that there's a limit to how far back you can look and that's 14 months, but any period of time within that 14 months, you can look back and say, hey, last Christmas, I want to see all the people who visited my website.
I want to create a lookalike audience of those folks before they head into the next holiday season. Use cases. Well, you can use audiences to retarget visitors who abandoned a cart.
So I'm sure all of us at some point in time have received an email from a company or an ecommerce company or even an SMS text from Uber Eats because you didn't finish an order and say, hey, you know, if you want to complete your order, sometimes they offer you a discount or a bonus of some sort. Well, the reason they're able to do that is that they can retarget visitors who abandoned a cart. They can set up an audience of people within a certain period of time who abandoned a cart, and then, you know, find those folks.
Or create lookalike audiences. Create a lookalike audience of people who statistically look like other bases of demographic criteria, behavior criteria, and the other websites that person has visited within a period of time. They create these, they identify these patterns, and create lookalike audiences that you can use going forward with the expectation that you're going to reach people similar to the people who have already engaged in your website or already purchased.
Whereas with segments, some use cases for segments, you can analyze the behavior of users who completed the purchase, right? So people who completed the purchase in the past, like, well, what was their path to creating that purchase? What was the average length of time they were on a website before they purchased? So you understand a lot about what people who purchased were actually doing to give you better insight. What pages do they visit? Okay, this page didn't seem to be as important in the sales process, whereas this one did, right? So that's looking backwards and analyzing what has already happened to gain insight, as opposed to creating criteria of audiences that you want to target going forward.
Determine who, you know, what criteria should be included in that audience definition. Comparing mobile versus desktop users is another use case. Are mobile users engaging more or less than desktop users? Are they engaging with more pages? Are they converting at the same rate? And you might, you know, determine some issues between the two user experiences as you gain insight into the behavior of mobile and desktop users.