Subscription-based businesses’ success depends on building long-term, profitable and sticky relationships with clients. To get your customers to make repeat visits, you need a value proposition and go beyond vanity metrics - download counts, daily active users (DAU)/ monthly active users (MAU) that refer to growth and retention. The analysis has to focus on the behavior of your users and this is where the cohort analysis takes great importance.
Cohort analysis is a behavioral analysis that takes the data from a given eCommerce platform, web application, subscription, or SaaS business and breaks down all the users into related groups for analysis. These related groups, or cohorts, usually share common characteristics or experiences within a defined time span.
Cohort analysis is a tool to measure user engagement over time. It helps to know whether user engagement is actually getting better over time or if it is only appearing to improve because of growth.
Cohort analysis shows how the metrics are developing over the life of the customer, as well as over the life of the product.
Cohort analysis can be carried out in two different ways.
Users can be segmented based on when they first purchased or signed up to use a product. In this case, the cohorts could break down by day, week, or month and therefore track them on a daily, weekly, or monthly basis.
If we want to measure the retention of these cohorts, we would see how often they buy your product or how long they stay using your application.
Acquisition cohorts allow us to identify trends and define when our users leave us. However, if we want to understand why they leave, we must analyze the behavioral cohorts.
On the other hand, users can be organized by the behaviors or actions they perform during a certain period: install an app or uninstall it, purchase or return, etc.
The goal here would be to measure how long cohorts remain active after taking certain actions.
An example of a behavior cohort can be the action of users reading the reviews on a website before buying a product or service. Here the point is to analyze whether those who read reviews have a higher conversion rate than those who don't, for example.
In order to explain how a cohort analysis works, we are going to share with you the following triangle chart example. The image below shows a cohort analysis chart of new users acquired per day and how many days the customers are retained. The key point is checking when users subscribe to the SaaS and how long these cohorts remain to be users of the service.
What this particular triangle chart shows us is how many new users subscribed to the newly launched application and when on the Y-Axis - in January 2021, there were 196 new users subscribed to the app. These new users form what would be known as the “January 2021 cohort”. Every day the information being collected on the numbers of new users is tallied, up until the most recent day and a cohort of October 2021.
Next, we look at the X-Axis to see just how many users stay subscribed to the application. This gives us an idea of the user lifetime to see how users from a cohort are interacting with the app. On month 0, the subscription date, 100% of the users remain using the application. However, you will see that as the days go by, this figure decreases. By month 1, only 84% of the users forming the January 2021 cohort (196 users) remain as a subscriber of the app and this trend continues downward over time. This is expected as you do expect users to stop using the app over time but what you want to look out for is how fast
As you move along the triangle analysis, you can see how other cohorts are performing in terms of their churn rate from the app and how the retention rate per period for the app is changing over time. This represents the product lifetime and allows for comparisons between cohorts to be made at the same stages in the customer life cycle. This is useful when looking back to understand the performance of the company during certain periods of time or life cycles of the customers to make the necessary adjustments to your businesses’ strategy.
Cohort analysis can get answers to the questions like how many new users subscribed to the newly launched application and how many users stay subscribed to the application. A triangle analysis gives us other performance indicators such as the churn rate from the app and how the retention rate per period for the app is changing over time.
From cohort analysis, you can develop a quantitative approach to know how your users can fall in love with your app – and then make it happen again and again.
Having as much information as possible about what is happening in your business will allow you to identify errors and solve them, as well as verify what things are working well and enhance them.
Cohort analysis makes it possible to know which customers leave you and when they leave, but also why they are leaving your app. In this way, you can see how well your users’ experience is, how well your users are being retained and how your business’ growth, engagement, and revenue is working.