Cohort Analysis in Google Analytics
Posted: Wed Jan 29, 2025 9:36 am
Google now offers its basic cohort account for various studies. This analytical tool allows marketers and analysts to test content, ads, and other products more easily and effectively. Cohort analysis is a convenient way to compare two or more groups over a specific time period.
For example, let's take an advertising campaign of an online store, where marketers used email newsletters. The advertising was launched through three channels. The first was through outlook email list participation in a contest on the Facebook page, the second channel - here 50% of people who visited the site saw a call to action on a pop-up window, and the third option - the invitation was issued through a static link on the site page.
After some time, the marketers of the tested project will be able to analyze the dynamics of email views for each cohort: those who subscribed because of the contest, through a pop-up window or a permanent link. And finally, after collecting the information, the specialists will be able to conclude which group of people acted more consciously. As a rule, subscribers who came via the link view all emails, unlike the first and second cohorts. Moreover, this group demonstrates the best indicators for the entire testing period. Although this is just a hypothesis, you probably get the main idea.
Let's now move directly to cohort analysis in Google Analytics. It is performed in the Audience section. Previously, more experienced Google Analytics users used the segmentation tools available there to create their own cohort reports. But now more ordinary people will be able to understand the service.
Cohort Analysis in Google Analytics
Source: WDnet Creation / shutterstock.com
Google Analytics service allows you to work with a cohort report, using different types of groups, sizes, metrics, time frames:
Types of cohorts . When we were working on this material, the only criterion for combining people into a cohort was the registration date, which made it possible to observe their actions on the site over a certain period of time.
Cohort sizes . This indicator depends on the period selected for the report, for example, you can include everyone who came in a day, 7 days or during a month. Above, we looked at the example with email newsletters and in each cohort we combined visitors who came in one month. Let's hope that in the near future Google Analytics will offer additional cohort sizes.
Metrics . This refers to a wide variety of parameters that can be measured. Currently, the following can be used as metrics: conversion to user ratio, number of pages viewed by one person, number of visits to the site by one user, number of retained visitors, conversion rate, number of paid orders, etc.
Time period . This refers to the moment of viewing, i.e. the relative time range.
The service allows you to create segmented cohort reports. As an example, try to conduct a comparative analysis of the average duration of stay on the site when using a mobile application and a desktop computer. Or you can count the number of people who registered ten days before the New Year holidays.
These studies show how much time mobile and desktop users typically spend on the site, and how much traffic and session length increase before the New Year.
For example, let's take an advertising campaign of an online store, where marketers used email newsletters. The advertising was launched through three channels. The first was through outlook email list participation in a contest on the Facebook page, the second channel - here 50% of people who visited the site saw a call to action on a pop-up window, and the third option - the invitation was issued through a static link on the site page.
After some time, the marketers of the tested project will be able to analyze the dynamics of email views for each cohort: those who subscribed because of the contest, through a pop-up window or a permanent link. And finally, after collecting the information, the specialists will be able to conclude which group of people acted more consciously. As a rule, subscribers who came via the link view all emails, unlike the first and second cohorts. Moreover, this group demonstrates the best indicators for the entire testing period. Although this is just a hypothesis, you probably get the main idea.
Let's now move directly to cohort analysis in Google Analytics. It is performed in the Audience section. Previously, more experienced Google Analytics users used the segmentation tools available there to create their own cohort reports. But now more ordinary people will be able to understand the service.
Cohort Analysis in Google Analytics
Source: WDnet Creation / shutterstock.com
Google Analytics service allows you to work with a cohort report, using different types of groups, sizes, metrics, time frames:
Types of cohorts . When we were working on this material, the only criterion for combining people into a cohort was the registration date, which made it possible to observe their actions on the site over a certain period of time.
Cohort sizes . This indicator depends on the period selected for the report, for example, you can include everyone who came in a day, 7 days or during a month. Above, we looked at the example with email newsletters and in each cohort we combined visitors who came in one month. Let's hope that in the near future Google Analytics will offer additional cohort sizes.
Metrics . This refers to a wide variety of parameters that can be measured. Currently, the following can be used as metrics: conversion to user ratio, number of pages viewed by one person, number of visits to the site by one user, number of retained visitors, conversion rate, number of paid orders, etc.
Time period . This refers to the moment of viewing, i.e. the relative time range.
The service allows you to create segmented cohort reports. As an example, try to conduct a comparative analysis of the average duration of stay on the site when using a mobile application and a desktop computer. Or you can count the number of people who registered ten days before the New Year holidays.
These studies show how much time mobile and desktop users typically spend on the site, and how much traffic and session length increase before the New Year.