The Fundamentals of Cohort Analysis

I think our earlier articles would have given you fundamental ideas of marketing analytics & would be sufficient to learn & map them in incoming articles.

Now, let me take you through one of the niche topics which is very new in the market & today being adopted by many companies to get better analysis of their businesses. 

Situation- Well, suppose if I put you in a hypothetical situation where “you” are an associate of a company who has launched a game application & now the task has been given to you in order to measure the growth & engagement of the application. 

Now, take some time & think what will be your parameters to assess the growth of the app?

I believe you would be on the same page, that is you are going to look for the Daily active users (DAU) or the Monthly Active users (MAU). It could also be the download counts to measure the growth. You are absolutely correct in your approach but there is something we missed out.  

Well, these are the basic growth metrics mentioned above which can be analyzed but what about the engagement metrics? How would we be able to get the understanding how users are getting engaged over the time. Therefore, I am going to take you beyond the conventional vanity metrics & will present you the new analysis known as “Cohort Analysis“. 

What are Cohorts?

A cohort is a group of users having similar characteristics or experience with a defined time span. Those characteristics could be users in a certain geographic location or the users who installed an app within the same period (daily, weekly & monthly) etc.

Cohort Analysis

Cohort analysis is a subset of behavioral analytics that takes the data from a given e-commerce platform, web application, or an online game and rather than looking at all users as one unit, it breaks them into related groups for analysis. Here, Cohort Analysis is achieved by analyzing Cohort Heat Map/Table.

Cohort Basic Terminology

Let us understand the Cohort Map mentioned below with the example of Netflix.

As mentioned above regarding the definition of cohorts, in this example we are considering daily cohorts starting from 25th Jan to 3rd Feb which means each date starting from 25th Jan will be a cohort group & number of users in each of these dates will be customers who had done their first sign up with the Netflix.

Now, below snapshot may increase your blood pressure but please hold your horses at this moment. 

[Note] :- Cohort table  gets automatically generated when we feed data in any software like Python, R etc. We don’t need to worry about it in order to prepare the table every time but yes intensive coding goes on at the back end from cleaning of the data to have the results on board.

Explained: Cohort Heat Map

What Does this Map tell us all about?

Retention over Product lifetime/Overall Retention (vertically) – As mentioned above in cohort table on 25th Jan, 1098 users signed up. Similarly, on 26th Jan – 1358 new users joined & so on. What does it mean?

It means folks signed up on Netflix and reached the home page but did not watch any films/episode or series – didn’t make use of the product/service. From this table, we can exactly determine each day retention of all cohorts as mentioned above for day 1 (27%) & day 2(19.2%) & so on.

Product lifetime depends on the factors like: – Product on-boarding experience, operations, customer support, visuals etc.

User lifetime (horizontally) – As mentioned above in cohort table on 26th Jan, 1358 users signed up on Netflix. Day 1 retention was 31.1%, day 7 retention was 12.9%, and day 9 retention was 11.3%.

It means it shows what % of people from each cohort comes on the app on each day giving us an estimate to get the retention rates of each cohort group.

User lifetime depends on the factors like: – Quality of the content, App exclusivity etc. 

This kind of an engagement analysis can be done through cohort & it will be useful for the company to take some key decisions which can entirely take business to the next level. Cohort is used for improving customer retention as we can determine how long people continue to use your app from their start point, time period that they stayed with us & etc. 

Deeper insights will be given on cohort analysis in the upcoming articles which will enlighten your mind in the areas of behavioral & acquisition cohorts. Till then, stay with us and happy learning!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s