Product Analytics with Short Notes 4 #Cohorts
Cohort Analysis
Cohort analysis is a great way to analyze and compare behaviors of different groups of users. When used right, it can bring useful insights especially about activation, engagement and retention. Next we will discuss what are cohorts, how to frame the right questions to analyze cohorts and how to analyze them.
Cohorts
A cohort is a group of people that are bound by one common trait.
One can create cohorts based on any information or factors.
Framing Right Questions
Questions should be SAD:
Segmented (Segments should be defined.)
Actionable (The answer should serve a decision/action.)
Detailed (Define the question as clear as possible with enough specifications about dimensions and metrics.)An example could be:
Did our
monthly retention rates increase with a significance level of 5% (detailed)
after the release of the last feature (actionable)
if we measure retention with cohorts by activation month (segmented)?
Cohort Analysis
Companies use cohort analysis to analyze customer behavior across the life cycle of each cohort.
Common types of cohorts: time-based, segment-based, size-based
1. Select the basis of the cohort. (e.g. Activation Date)
2. Select cohort size. (e.g. Cohorts by Activation Month)
3. Select / define the metrics. (e.g. Retention Rate)
4. Select date range to analyze. (e.g. from activation to 12 months after activation)
5. Compare the results.
Example:
Retention Rate of Cohorts By Signup Month over Time
Previous Part:
