If you’ve ever found yourself skimming past the cohort report, disoriented, or unsure of its potential, today is the day to dive in and find out what you’re missing.
Cohort analysis allows you to group users who share a common characteristic (more clarity on this vague description later), such as their first visit date, to analyze their behaviors and trends over time.
You’ll be able to gain invaluable insights into user behavior and address declines in your business by making sense of this zigzag report.
Here is an overview of how you can go from decline to growth using Google Analytics 4 cohort analysis to reengage your users:
- What Is Cohort Analysis in GA4?
- Google Analytics 4 Cohort Analysis Examples
- Where Can You Find GA4 Cohort Reports
- How to Create a Custom Cohort Exploration
- How Is Cohort Analysis Done?
Let’s dive in!
What Is Cohort Analysis in GA4?
According to the Analytics Help documentation, a cohort is:
“[…] a group of users who share a common characteristic that is identified in this report by an Analytics dimension. For example, all users with the same Acquisition Date belong to the same cohort. The Cohort Analysis report lets you isolate and analyze cohort behavior.”
Based on this definition, it’d be reasonable to wonder how a cohort is any different from a GA4 comparison, filter, audience, or segment.
The difference here is that with cohorts, you can analyze specific behaviors for a particular group over time.
Cohort analysis is most commonly used to assess user retention.
For instance, it can help you determine the effectiveness of your user acquisition strategy by tracking the number of users acquired on a specific day who return to your website in the following days.
The higher the return rate, the better your user retention.
However, there is so much more you can do with Google Analytics 4 cohort analysis.
You can decide the criteria for creating your group of users (cohorts) and the return criteria. We’ll look at examples to make this even more straightforward.
Google Analytics 4 Cohort Analysis Examples
Let’s look at a few examples to learn what’s possible with Google Analytics 4 cohort analysis.
User Retention
You can track the number of people who revisit your site after their first visit within your chosen period. To do this, you’ll have to start by grouping users into cohorts based on when they first landed on your site.
For instance, you might focus on everyone who visited during January.
You’ll have to check how often these newcomers returned throughout the month and drill down into daily or weekly details.
To illustrate: If 1,000 users visited on January 1st, you can see how many returned on the 2nd, 3rd, and so on through the 31st. If another 500 users showed up on January 2nd, you can track their return visits, too.
While it is possible to analyze this data on a daily level, it is expected to be examined weekly. This choice involves only five rows for the month, which makes it easier to manage.
Product Quality
You can see how often users uninstall your app after experiencing errors.
First, you’ll need to define the cohort, which can be a custom error event that you’ve defined in your analytics. This means you’re including users who have encountered a specific error in your app.
Next, you set the return criteria. The return criteria indicate the action that shows users have left the app. For this, choose the app_remove event, meaning you’re tracking users who have uninstalled your app.
By examining these groups, you can determine the impact the error had on app uninstalls over time.
Micro Trends
An increase in revenue can hide underlying flaws or areas for improvement in your business strategy. Let’s take the example of a company that saw a steady increase in revenue over the last three months. The trend appears positive overall.
However, when you examine the weekly data within that quarter, you see that new users are mainly responsible for increased transactions. You also notice a significant drop in transactions after five weeks.
The new users play a significant role in the initial boost of your transactions, but they also play a part in the subsequent drop-off as they tend to disengage after a few weeks.
The key to sustaining growth lies in the timing of your re-engagement efforts. By targeting them at the right time, you can effectively counter this trend and ensure a longer period of growth.
Reengage Users with Campaigns at the Right Time
This is similar to micro trends, but the difference is the focus on dealing with user disengagements, which can take many forms such as fewer sessions, fewer page views, or reduced revenue.
It’s about how you can compensate for the loss caused by attrition.
For example, if revenue drops during specific weeks after acquiring new users, you could reengage them with a remarketing or email campaign offering discounts or showcasing new products.
You can also use dynamic remarketing to show them ads for products related to what they bought before.
How Your Short-Term Campaigns Impact User Behavior
Imagine rolling out a series of one-day email promotions, like 30% off, 25% off, and 20% off deals leading up to a holiday.
With Google Analytics 4 cohort analysis, you can compare Revenue per User and Transactions per User metrics for each group of users attracted by these different discount campaigns.
Where Can You Find GA4 Cohort Reports
You can find cohorts in the User activity by cohort card in the Report snapshot, the Retention report, and Cohort Explorations.
User Activity by Cohort Card
- Click Reports.
- Click Reports snapshot.
- Find the User activity by cohort card.
Retention Report
- Follow the same steps shown earlier in the Report snapshot.
- In the User activity by cohort card, click View retention.
You’ll land on the Retention overview report, and will see the following charts:
Cohort Explorations
You can create one from scratch or use a pre-made report here. You’ll be able to conduct a deeper analysis, as there are more options regarding cohort inclusion, return criteria, and breakdown dimensions.
To access a pre-built cohort report:
- Go to Explore in your left navigation panel.
- Select the Cohort Exploration card in the Template Gallery.
You should see a ready-made cohort exploration to determine user retention.
How to Create a Custom Cohort Exploration
Creating a Cohort Exploration is straightforward, but you need to know your inclusion and return criteria.
To get a better sense of this, we’ll create a cohort Exploration to look at Product Quality (similar to one of our previous examples) from scratch.
Or to put it another way, we want to see how often users who experience an error come back to our website. And we’ll use the Google Analytics demo account.
First, we need to select a template for using the Cohort technique. The easiest way to start is to follow the steps demonstrated in the previous Cohort Exploration section.
If you use another Exploration template and need your cohort report included, simply switch the technique. For example, here we are in a Free Form Exploration. To add your cohort report:
- Click the tab’s plus sign +.
- Select Cohort exploration in the dropdown list.
Let’s review our objective and criteria:
| Description | Details | Events Used |
| Objective | How often do users who experienced an error engage with our site again (started a session) | – |
| Date Range | The month of December 2023 | – |
| Cohort Inclusion Criteria | Users who experienced errors | errors |
| Cohort Retention Criteria | Users who engaged with our site which we’ll define as those who started a session | session_start |
| Granularity | Weekly | – |
| Values | Active users (metric) | – |
You can follow our configuration from the following image:
The last step is to analyze these cohorts.
How Is Cohort Analysis Done?
Understanding the Google Analytics 4 Cohort Analysis Chart
Each row in the chart applies to a different cohort, and each column represents a time interval based on your granularity selection which can be monthly, weekly, or daily.
The first column, Week 0, shows a cohort’s (measured) behavior in their first week. The subsequent columns represent the following weeks.
The values within the cells of this chart represent the metric being analyzed.
The Colors denote the metrics and their values. The darker the color, the higher the number, and the lighter the color, the lesser it is.
Reading the Cohort Analysis Chart
Left to Right
Reading the chart from left to right helps to understand a cohort’s behavior changes over time.
In our example, between May 21 and May 25, we acquired 8053 users. The same 8053 users had 93 transactions in the first week they were acquired (week 0). The same 8053 users then had 16 transactions the following week. And so forth.
Up and Down
Reading the chart up and down allows us to compare different cohorts at the same point in their lifecycle.
The start of each row applies to a different cohort.
Therefore, newly acquired users are one cohort between May 21 and May 25. Newly acquired users between May 26 and June 1 are another cohort. Newly acquired users between June 2 and June 8 are a different cohort. And so on.
An example of insight is that from June 9 to June 15, this cohort had 368 transactions, which was higher than the other cohorts.
Yet it had only 26 transactions on the first week (week 0) they were acquired, which is average for all the cohorts during that week.
They have no other transactions after their second week (week 1).
On the acquisition date, there might have been a short-lived flash promotion sale for the product. A possible strategy to encourage repeat purchases could involve offering promotions on other products simultaneously.
Breakdowns
You can use dimensions to break down your cohorts. We are breaking down this report by devices using the device category dimension:
Summary
Google Analytics cohorts can look unclear at first. However, we’ve uncovered how they can be the key to re-engaging users.
Now, you should understand what cohorts are, how to read and create them, and what you can do to sustain or increase your user engagement.
If you enjoyed our use of cohorts, you’ll find value in our Top 10 GA4 Exploration Reports with Use Cases.
Which results have you witnessed from leveraging cohorts? Share your experience with us.
