logo-small
Features Prices
News 0
Latest News See All

Temporarily unavailable. Please come back later.

See All
Webinars 0
Upcoming Webinars See All
Upcoming Webinars

Sorry, we could not find any upcoming webinars.

See recorded webinars
Blog 0
Recent Posts See All

Temporarily unavailable. Please come back later.

See All
Joydeep Bhattacharya

How to Use Cohort Analysis in Google Analytics to Drive More Business?

Joydeep Bhattacharya

Cohort analysis is a great way to increase user engagement for your business and generate more revenue. Earlier, paid tools like KISSmetrics gave us the facility to use Cohort Analysis as a way to measure data but now, Google Analytics has launched a beta version of Cohort Analysis that allows webmasters to determine which strategies are working for them.

What is Cohort Analysis?

You can think of Cohort Analysis as a study of the activities of a particular cohort over a period of time. A cohort is a group of people who are segmented based on a defined time period and share a common set of characteristics.

This study takes into account actionable metrics rather than vanity metrics. Actionable metrics are those that help to improve a product or sales like the rate of conversion of a particular product during a period of time or the rate of retention of customers who signed up when the sale was going on.

On the contrary, vanity metrics only provide data about what is going on in the site, like number of visits or number of downloads which offers no insights regarding the improvement of sales. As per Tommy Griffith, CEO of ClickMinded, “Cohort Analysis gives us a clear and accurate idea of the activity of the visitors in order to improve the flow of sign ups. We have analyzed this report in order to understand the specific days the students are more interested in signing up for the course.”

Why Do I Need Cohort Analysis?

  • Customize products or services based on specific cohorts that behave differently.
  • Determine the time taken by a customer for the first purchase.
  • Determine the time taken by a customer for the next purchase.
  • Know the retention rate of customers who signed up when a sale was going on.
  • Get the conversion rate of customers who signed up during festive season like Christmas.
  • Uncover the repeat rate of new orders over a specific period of time like in the first two weeks of each month compared to the last two weeks.
  • See the average retention rate of customers who signed up through a particular digital marketing channel during a defined period of time.
  • Realize the average spends of customers who are acquired at the starting of the year as compared to those acquired at the end of the year.

The AARRR Metrics for Businesses

Every business wants to know who are their customers, what do they do when they visit their site, what they expect when they visit the site and what causes them to visit the site repeatedly. In total, CA helps us to answers the 5 main metrics that are useful for every business as defined below:

  • Acquisition: the cost of acquiring first-time users
  • Activation: what causes the audience to remain active on the site
  • Retention: the retention rate of the customers and identification of the factors that causes the customers to repeat their orders
  • Referral: rate at which people are inviting others to the site
  • Revenue: determining the first point of purchase

Deciding Your Cohorts Based On Your Business Objectives

You should be extremely clear regarding your business objectives and what you want to achieve for your business with the help of the website. Now, decide Cohorts after segmenting your users based on a specific time period and examine their characteristics. Here is a sample objective and creation of a cohort based on the objective:

Objective                    

Your objective is to increase revenue from PPC campaigns.

Example Cohort:

Create several different cohorts based on the data from the past 12 months, past 6 months and past 3 months and analyze when did you generated the maximum average revenue per order as well as the highest retention rate. If the last 6 months' cohorts generated the maximum average revenue per order along with the retention rate then examine what did you offered to the users at that point of time which helped the users to order more products and also generated repeated visits.

You might have offered some discounts, you might have some product in your stock at that point of time that were liked by the users but currently you have stopped maintaining that stock or you may have targeted local cities in the past 6 months as compared to international market in the past 12 month campaigns, etc. These are some of the data that Cohort Analysis can easily tell you. Applying the results obtained from Cohort Analysis can improve your existing campaigns and help you generate more revenue.

How to Do Cohort Analysis Using Google Analytics – Some More Examples to Consider

There is nothing better than taking real life examples to understand the value of Cohort Analysis, so let's take a look at some more examples.

Example 1: I run an email campaign for 14 days offering 30% discount on my products. I want to track the behavior of users who signed up during this period.

Simply run a Cohort report after selecting "Acquisition date" under Cohort type, "by day" under cohort size, "user retention" under metric and "last 14 days" under date range.

Also, create a new segment named "email" and choose "email" from the "medium" menu under the "traffic sources" as shown in the screenshot below:

Cohort Analysis

Example 2: I am selling online clothes and I want to see how long the users remain interested in my offers and deals.

Run a cohort report for the past 6 weeks. Choose "Acquisition date" under Cohort type, "by week" under cohort size, "user retention" under metric and "last 6 weeks" under date range.

Cohort Analysis

You can easily track the time when the engagement starts to decline, if suppose, it happens after 4 weeks then you can easily opt for a remarketing or email marketing campaign with some discounted offers in order to reengage users.

Bonus tip: If you have an email database of users you wish to target then you can use FB remarketing to target only those users who are already your customers. This way you can make more money by investing less.

Example 3: Compare behavior of users who are acquired by organic channels vs PPC ones

Run a cohort report for the past 6 weeks. Choose "Acquisition date" under Cohort type, "by week" under cohort size, "user retention" under metric and "last 6 weeks" under date range.

Also, segment the users based on "organic" and "paid" traffic as shown in the screenshot below:

Cohort Analysis

Now, you will get the data of the past 6 weeks for both organic and paid traffic and you can easily identify the retention rate of users obtained via organic or paid traffic. You can adjust your spend on these channels accordingly. (For more examples, please refer to this document).

Understand the Choices Available - What is Displayed in the Cohort Chart?

The horizontal column displays the time and the vertical column displays the number of sessions. The vertical columns displaying the sessions are the cohorts. Have a look at the technical terms of the elements included in the chart:

Cohort Analysis

The menus that are used to configure the Cohort Analysis report:

  • Dimension characterizes the Cohort Type. By default only 1 value is available as of now i.e. the Acquisition date. In future, Google will expand it with other options. The Acquisition date is the date of the first session when the user first arrives on your site.
  • Next, you choose the Cohort size. 3 options are available, “by day,” “by week” and “by month.” Selecting “by day” will help you to see the data of customers who were acquired the same day. Similarly, selecting “by week” or “by month” will help you to see the data of customers who were acquired in the same week of the same month respectively.
  • Next, you choose the metric that you want to evaluate. The following options are available:

Per User A- Goal Completions per user  B- Page Views per user  C- Revenue per user  D- Session Duration per user  E- Transactions per user 

Retention

A- User Retention

Total

A- Goal Completions B- Page Views C- Revenue D- Session Duration E- Sessions F- Transations G- Users

  • Next, you choose the relative date range of the data. The “date range ”and “cohort size” columns are interrelated. This means that the options displayed under the “date range” changes as per the value selected under the “cohort size” column.

Understand the Data

Cohort Analysis

Report configured to show Acquisition Date cohorts by the User Retention metric.

Chart

The chart displays the cumulative metric values for all cohorts.

Columns

The column identifies the Cohorts based on the menu values that are selected. By default it shows 13 columns. For example, if you chose to display Cohorts by Acquisition Date then the column lists the acquisition date for each cohort.

Rows

The first row displays the total metric value for all cohorts for each column. For example, if the metric is unique pageviews and the columns are daily data, then the first row shows the unique pageviews for the day.

Cells

The cells are used for holding the relevant metric values. For example, if you are using the unique pageviews metric, then each cell contains the number of pageviews per Cohort per time increment.

Hence, Cohort Analysis is a great way to measure the user engagement and user retention on your site. It helps you to analyze the metrics that matters the most for your business. Have you been benefitted by the Cohort Analysis report? Please share with us in the comments.

Joydeep Bhattacharya has over a decade of experience in digital marketing. He maintains a personal SEO blog, Seosandwitch.com where he shares stuff related to SEO, PPC, Social Media and Conversion Optimization.

Comments

2000 symbols remain
Great description from this tool! Any idea why our account is stuck on 3 months max only ?
Marian Popovych
Marian Popovych
Thank you for the great post, Joydeep. Starts to use the Cohorts :)
Have a Suggestion?