Cohort Analysis for Growth Marketers
If you have read some of my previous posts here on Medium, you might already know that I strongly believe cohort analysis is one of the most valuable types of analysis for any growth marketeer.
Cohort analysis is a powerful tool for growth marketers looking to understand and improve the performance of their products. By analyzing the behavior and outcomes of a specific group of users over time, growth marketers can identify trends and patterns that inform their strategy and help drive growth.
In this post, we’ll cover the basics of cohort analysis, including how to define your cohort, gather data, and analyze the results. We’ll also discuss how to use cohort analysis to inform your growth strategy and provide some real-world examples of how companies have successfully used this approach.
Introduction to cohort analysis
At its core, with cohort analysis, you analyze how a certain metric changes over time for a specific group of users. This allows you to understand how different groups of users are interacting with your business over time.
For example, if you are a growth marketer for a SaaS company, you might use cohort analysis to understand how different groups of customers are retained over time or how much they are spending. This information can help you identify trends and patterns that inform your growth strategy, such as which marketing channels are most effective for different types of customers, when to follow up with leads, or which features are most popular with certain groups of users.
Still lost?
Check out the SAAS CFO explaining it below:
Defining your cohort
The first step in conducting a cohort analysis is to define the group of users that you want to analyze. This could be a group of customers who signed up for your product during a specific time period or a group of users who have reached a certain milestone.
For example, you might want to analyze a cohort of customers who signed up for your service in the first quarter of the year, or a cohort of users who have made at least one purchase in the past six months. The specific criteria for defining your cohort will depend on your goals and the questions you are trying to answer.
Gathering data
Once you have defined your cohort(s), the next step is to gather the data that you will need to analyze their behavior over time. This will typically include metrics such as user acquisition, engagement, retention, or conversion.
For example, if you are analyzing a cohort of customers who signed up for your service in the first quarter of the year, you might track metrics such as the number of new users, the number of active users, the number of users who have made a purchase, or the average revenue per user.
It’s important to track a range of metrics in order to get a comprehensive understanding of your cohort’s behavior and outcomes. You may also want to segment your data by different factors, such as geographical location or device type, to gain deeper insights.
Two types of cohort analysis:
In general, there are two types of cohorts you can look at:
- Segment-based cohorts: The shared characteristics of these types of cohorts could be what browser they are using, which product they have bought, or what country they are from.
- Time-based cohorts: These cohorts are grouped by when a certain action happened. This could be the first sign up, first purchase or when they tried out a certain feature.
This is also what is called the cohort inclusion metric if you are using GA4 Cohort Explorer for instance. Whether it’s segments or time-based cohorts, then these types of cohorts are what you will have on the 1st column in your cohort table.
Below you can see an example of a segment-based cohort table exploring how different browsers are performing.
Below you see an example of a time-based cohort table.
Analyzing the data
Once you have gathered the necessary data, the next step is to analyze it in order to identify trends and patterns. There are several ways to do this, including using tools like Excel or Google Sheets to create graphs and charts, or using specialized cohort analysis software. These types of cohort analysis software are typically BI tools.
One common approach is to create a cohort table, which shows the progress of different groups of users over time. For example, you might create a table that shows the number of active users in each cohort at the end of each month. This can help you identify trends such as whether certain groups of users are more or less likely to remain active over time.
Another approach is to create a retention curve, which shows the percentage of users in each cohort who are still active at different points in time. This can help you understand the long-term retention rates for different groups of users and identify opportunities for improving retention.
You may also want to segment your data by different factors, such as geographical location or device type, to gain deeper insights. For example, you might compare the retention rates of users in different regions to see if there are any significant differences.
Overall, the key to effective cohort analysis is to be systematic and thorough in your analysis. By carefully organizing and analyzing your data, you can gain valuable insights that inform your growth strategy and help drive growth.
Using cohort analysis to inform growth strategy
Once you have analyzed your cohort data and identified trends and patterns, the next step is to use this information to inform your growth strategy. This could involve making changes to your product based on feedback from certain groups of users, or adjusting your marketing efforts based on what has worked well for different cohorts.
For example, if you notice that a certain group of users is particularly engaged with your product, you might want to focus on acquiring more users like them through targeted marketing campaigns. On the other hand, if you see that a certain group of users is not retaining as well as others, you might want to investigate the reasons for this and implement changes to improve retention.
Techcrunch has written an excellent article on cohort analysis, where they are showing an example of what looks like a successful marketing campaign. However, when following the cohort over time, the customers from that campaign have really bad engagement.
When zooming in on the November cohort, then it appears that engagement has been really low for that cohort.
Whatever marketing channel was used, this might not be worth doing again.
Case studies: SaaS, E-Commerce and Mobile Gaming
To give you an idea of how other companies have successfully used cohort analysis to drive growth, here are a few examples:
- A SaaS company used cohort analysis to identify a group of users who were particularly engaged with their product. They then used this information to create targeted marketing campaigns and saw a significant increase in customer acquisition and retention.
- An e-commerce company used cohort analysis to identify trends in customer behavior and spending. They noticed that certain groups of customers were more likely to make repeat purchases, so they focused on acquiring more users like these through targeted marketing efforts. This resulted in a significant increase in revenue.
- A mobile gaming company used cohort analysis to identify trends in user behavior and engagement. They noticed that certain groups of users were more likely to make in-app purchases, so they focused on acquiring more users like these through targeted marketing campaigns. This resulted in a significant increase in in-app revenue.
Here is a quick example of how a few cohorts where retained over time. Watch from 11:57–12:30.
I can really recommend the rest of the video as well!
Cohort Analysis for SaaS Businesses
As a growth marketer in a SaaS startup, there are several growth metrics that would make sense to track using cohort analysis:
- User acquisition: Track the number of new users acquired in each cohort, as well as the acquisition channels that are most effective for each group.
- Engagement: Monitor the level of engagement for each cohort, including metrics such as the number of active users, the number of sessions per user, and the average session duration.
- Retention: Analyze the retention rates for each cohort over time to understand which groups of users are most likely to continue using your product.
- Conversion: Track the conversion rates for each cohort, including the percentage of users who complete key actions such as signing up for a trial or making a purchase.
- Revenue: Analyze the revenue generated by each cohort to understand which groups of users are most valuable to your business.
By tracking these metrics using cohort analysis, you can gain a better understanding of how different groups of users are interacting with your product, and how this is changing over time.
Challenges of cohort analysis
There are a few challenges that growth marketers in SaaS startups (or anyone performing cohort analysis, basically) may face when conducting cohort analysis:
- Defining the cohort: Identifying the right group of users to analyze can be challenging, as you need to define your cohort in a way that is meaningful and relevant to your business goals.
- Gathering data: Ensuring that you have the right data points and metrics to track can be difficult, especially if you are working with a large, complex dataset or if data from different systems need to be merged.
- Analyzing the data: Interpreting the results of your cohort analysis and identifying trends and patterns can be challenging, especially if you are dealing with a large amount of data.
- Acting on the insights: Once you have identified trends and patterns through cohort analysis, it can be difficult to determine the best course of action based on this information.
- Limited historical data: If your SaaS startup is relatively new, you may not have a lot of historical data to work with, which can make it difficult to conduct cohort analysis.
Despite these challenges, cohort analysis can be a valuable tool for growth marketers in SaaS startups, as it can provide insights into user behavior and outcomes that can inform your growth strategy. By carefully defining your cohort, gathering the right data, and analyzing the results, you can gain valuable insights that can help drive growth.
A few other valuable resources:
Here are a few resources on cohort analysis that may be helpful for growth marketers in SaaS startups:
Please share if you have other valuable resources to share!
Conclusion
Cohort analysis is a valuable tool for growth marketers looking to understand and improve the performance of their products. By analyzing the behavior and outcomes of specific groups of users over time, you can identify trends and patterns that inform your growth strategy and help drive growth. Try using cohort analysis in your own growth marketing efforts and see the positive impact it can have on your business.
