Measure and improve · Add-on
Understand and improve retention
Retention shows whether people keep getting value. Without it, growth leaks away.
About 2-5 daysEvergreenLast reviewed 5 November 2026
Who this applies to
- Skills it builds
- Data and measurement
- PM types
- Growth
- Product types
- All
- Starting points
- All
- Ways of working
- All
- Experience
- All
- Size of organisation
- All
Steps, by experience
Beginner (under 1 year)
- Pick the action that shows someone got value.
- Measure how many new users come back and do it again after a day, a week and a month.
- Compare groups who stay with groups who leave.
- Talk to people who left to learn why.
- Test changes that help people reach value sooner.
Intermediate (1 to 4 years)
- Build retention curves by cohort.
- Find behaviours linked to staying.
- Test improvements to early experience.
Experienced (5+ years)
- Separate new, current and returning users.
- Link retention to lifetime value.
- Set retention goals by segment.
Who can do this
Owner: Product manager
Can be delegated to
- Data Analyst or Data Scientist / ML Engineer: Build retention analysis
- User Researcher: Interview people who left
Reviewed by: Data Analyst
Tool examples
- Amplitude
- Use a retention analysis chart by sign-up cohort.
- Mixpanel
- Use retention reports to compare cohorts.
Definition of done
- Value action defined
- Retention curves built
- Improvement experiments planned
Common pitfalls
- Measuring logins instead of value
- Looking only at averages
- Focusing on acquisition while users leave
If you lead the team
What good looks like: Teams understand who keeps using the product, who leaves and why.
Ask your PM
- How does retention look for recent groups of new users, and why?
- Which actions predict that someone will stay?
Warning signs
- Growth reports show sign-ups but not retention.
- New users are acquired faster than they leave, hiding the problem.