Segment Overlap Report
Learn to use the GA4 Segment Overlap report to isolate complex user groups and build better retargeting lists
Cohort Reports
January 10, 2026
- First Published: January 10, 2026
- Last Modified: January 14, 2026
Getting traffic to your site is one thing. Keeping those visitors coming back?
That is an entirely different challenge.
We spend a significant amount of budget on acquisition, but if we don’t understand what happens after the first click, we are flying blind.
Are those users sticking around?
Do they come back to buy, or do they vanish after a single session?
This is where the Cohort Exploration comes in. It is a powerful technique for isolating groups of users based on shared traits and watching their behaviour evolve over time.
If you want to get serious about retention and understand the long-term value of your channels, this is the tool you need.
Table of Contents
Key Takeaways
What it is: A cohort is a group of users who share a common characteristic (like acquisition date) within a specific timeframe.
Why use it: It is the most effective way to measure user retention and see how specific groups behave weeks or months after their first visit.
The nuance: You can adjust how retention is calculated (Standard, Rolling, or Cumulative) to answer different data questions.
The strategy: Use segmentation (e.g., users who landed on a blog post) to spot which content drives loyalty.
What Is a Cohort Exploration?
In simple terms, a cohort is just a fancy word for a group of people who did the same thing at the same time.
The Cohort Exploration report allows you to isolate these groups based on specific attributes and track them.
The most common example is grouping users by their acquisition date.
For instance, everyone who visited your site for the first time in January is one cohort. Everyone from February is another.
By looking at these groups separately, you can see if your site is getting better (or worse) at retaining people over time.
How to Set a cohort exploration
Getting this report running is straightforward, but the terminology can be a bit dry. Let’s break it down so it makes sense.
To start, navigate to the Explore section in GA4.
You can select the Cohort Exploration template directly, or choose a Blank Exploration and select “Cohort exploration” as your technique.
Once you are in, you need to configure three main levers:
Inclusion Criteria
This defines who gets into the cohort.
By default, this is usually First Touch (when they first visited). However, you aren’t limited to that.
You could define a cohort based on a transaction or a specific event like add_to_cart.
Return Criteria
This defines what counts as “coming back.” Are you checking to see if they just visited the site again (session_start)? Or do you only care if they came back and bought something (purchase)?
Cohort Granularity
This determines the time blocks you are analysing: daily, weekly, or monthly.
Understanding the Calculation Types
GA4 offers three ways to calculate the results, and choosing the right one is critical for getting the correct answer.
- Standard Calculation: This looks at each specific time period in isolation. If you are looking at Week 1, it shows users who came back specifically during Week 1.
- Rolling Calculation: This is broader. It includes users who met the return criteria in that period as well as all previous periods. It’s useful for understanding sustained engagement.
- Cumulative Calculation: This is the most generous metric. A user counts in the cell if they met the return criteria in any period within the exploration.
My advice? Stick to Standard when you are first starting out.
Breakdown
Adding a breakdown dimension allows you to get more granular with the information. You can easily compare how a cohort differs along that dimension. Some examples include:
- First User Source: How are different users who arrive by different sources interacting with your site.
- Device: Is there a huge disparity between mobile and desktop.
Values & Metric Type
You can also add / amend the values and metric types, so you could perhaps go from active users to average session duration or purchase revenue to understand the cohort that way.
Hovering over the chart also gives you an explanation of sum and per cohort user.
- Sum – display the total values for the selected metric.
- Per user – display the metric value divided by the number of cohort users. This is shown as a percentage.
Use Cases
Knowing how to build the report is fine, but knowing why to build it is better. Here are a few ways to use this data to improve your marketing.
Spotting Retention Issues
The most obvious use is checking user retention. If you see that users acquired in March dropped off significantly faster than users acquired in January, you need to investigate. Did you change your onboarding flow? Did you run a low-quality ad campaign that brought in irrelevant traffic?
Comparing Segments
You can apply Breakdowns and Segments to your cohorts. This is where it gets interesting.
For example, try breaking down your report by Device Category.
You might find that mobile users have a terrible retention rate compared to desktop users. That is a clear signal that your mobile experience might be broken or frustrating.
Content-Led Retention
One strategy I particularly like involves segmenting by the first page a user saw.
By creating a segment for “Landed on Blog,” you can see if your educational content drives better long-term loyalty than your product pages. If blog readers return more often, you have a strong business case for increasing your content marketing budget.
The Caveats
Before you go and build a hundred reports, there are limits you need to know.
The 60 Cohort Limit: You can only show a maximum of 60 cohorts in a single visualisation.
Breakdown Limits: If you apply a breakdown (like Country or Device), GA4 will only show the top 15 values.
User: The information is using device based data only, so even if your reporting identity is blended and your sending user_id, it won’t be taken into account.
Thresholding: As with all things GA4, if your user count is too small, Google might hide data to protect user privacy. If you see lots of zeroes where there should be numbers, this is likely the culprit.
Summary
Cohort Explorations allow you to move beyond “vanity metrics” like total hits and start measuring actual loyalty.
By understanding who stays and why, you can stop wasting budget on channels that bring in fleeting traffic and double down on the strategies that build lasting customer relationships.
Don’t just stare at the table.
Define a clear objective, pick a meaningful cohort (like purchasers vs. browsers), and use the data to make a change.
Frequently Asked Questions
Can I use Cohort Exploration for specific events?
Yes. While the default return criteria is usually any_event (any visit), you can change this to specific actions like purchase or sign_up. This allows you to track “repeat purchasers” rather than just “repeat visitors.”
Why does my Cohort report show low numbers?
This could be due to thresholding. If the group of users is small, GA4 may withhold data to prevent individual identification. Try expanding your date range or removing complex breakdowns to increase the sample size.
What is the difference between Rolling and Standard calculation?
Standard counts users who return in that specific time block (e.g., Week 2). Rolling counts users who returned in that block or any previous block, giving you a broader view of engagement.
How to Fix the China & Singapore Bot Issue
Seeing a sudden surge in traffic from China or Singapore? Learn more about it here
Tracking Emails in GA4
Learn the correct approach to track email campaigns accurately in Google Analytics 4.
Author
Hello, I'm Kyle Rushton McGregor!
I’m an experienced GA4 Specialist with a demonstrated history of working with Google Tag Manager and Looker Studio. I’m an international speaker who has trained 1000s of people on all things analytics.