Retention Analysis
Retention measures whether users return to your product over time. It’s one of the most important indicators of product health and product-market fit. If users don’t come back, nothing else matters.Understanding Retention
The Retention report in Mixpanel is designed to help you assess user engagement over a specified period of time. It answers critical questions:- Have you achieved product market fit?
- Are your users returning over time or churning?
- Which cohort of users come back and what drives them to?
- How does retention differ across user segments?
How Retention Works
A retention report measures the percentage of users who perform an action (the “return event”) after first performing another action (the “first event”). Basic structure:- First event: The action that starts the clock (e.g., Sign Up, First Login)
- Return event: The action that indicates a user has returned (e.g., Login, Key Feature Used)
- Time window: Daily, weekly, or monthly intervals
Understanding Retention Criteria
Define retention analysis to calculate users who come back on a specific time unit (e.g. day, week, month) or any time unit afterwards. You can also switch the retention mode between the default rolling time window and calendar defined time window.Rolling vs. Calendar Windows
Rolling retention measures retention in intervals from the first event. Day 1 = 24-48 hours after first event, Day 2 = 48-72 hours, etc. Calendar retention measures retention in fixed calendar periods. Week 1 = the calendar week following the first event, regardless of when in that week the user signed up.On vs. Return Retention
On retention (also called “Classic retention”) measures users who return in a specific time period. For example, users who return exactly on Day 7. Return retention (also called “Range retention” or “Return or Later”) measures users who return in a time period or any time after. For example, users who return on Day 7 or later.Reading a Retention Report
Retention reports typically show:- Cohorts (rows): Groups of users who performed the first event during a time period
- Time intervals (columns): Day 0, Day 1, Day 2… or Week 0, Week 1, Week 2…
- Retention percentages (cells): The % of each cohort that returned in each time interval
- Day 0/Week 0: Always 100% (users who performed the first event)
- Sharp drops: Large decreases between adjacent periods indicate friction
- Flattening curve: When the curve levels off, you’ve found your retained user base
- Cohort comparison: Do newer cohorts retain better than older ones?
Choose the event that indicates a user has returned (e.g., Any Event, or a specific high-value action).
Understanding Frequency Criteria
Define retention analysis to calculate users who did the event in at least X unique intervals or in exactly X unique intervals.Why Frequency Matters
Frequency criteria helps you identify power users and understand usage patterns:- At least 1 time: Standard retention (did they return at all?)
- At least 3 times: Shows consistent usage
- At least 5 times: Identifies highly engaged users
- Exactly 1 time: Flags one-time users who may need re-engagement
Segmenting Retention Analysis
Breaking down retention by user properties reveals which types of users stick around:Common Segmentation Dimensions
- Acquisition source: Organic, paid, referral
- User attributes: Plan type, company size, role
- Product experience: Completed onboarding, used specific features
- Geographic: Country, region, language
Questions to Answer with Segmentation
- Do users from paid channels retain better than organic?
- Does completing onboarding improve retention?
- Which features correlate with higher retention?
- Do retention patterns vary by user role or company size?
What Good Retention Looks Like
Retention benchmarks vary dramatically by product type:Consumer Products
- Social networks: 40-60% Day 7 retention, 20-30% Day 30 retention
- Utilities/Tools: 20-40% Day 7 retention, 10-20% Day 30 retention
- Gaming: 30-50% Day 1 retention, 15-25% Day 7 retention
B2B Products
- Collaboration tools: 60-80% Week 1 retention, 40-60% Week 8 retention
- Business intelligence: 50-70% Week 1 retention, 30-50% Week 8 retention
Improving Retention
Once you understand your retention patterns, take action:Onboarding Optimization
Users who complete onboarding typically retain better. Focus on:- Getting users to their “aha moment” faster
- Reducing time to first value
- Setting clear expectations about what to do next
Feature Adoption
Identify features that correlate with higher retention, then:- Promote those features in onboarding
- Send in-app messages or emails encouraging usage
- Make these features more discoverable
Re-Engagement Campaigns
Create cohorts of users at risk of churning:- Users who haven’t returned in X days
- Users who signed up but never completed onboarding
- Users who were active but have gone dormant
Product Changes
Use retention data to guide product decisions:- If Day 1 retention is low, focus on first-time experience
- If Week 4 retention drops, add features for ongoing value
- If retention varies by segment, optimize for your best users
Retention and Product-Market Fit
Retention is one of the best indicators of product-market fit: Early signs of PMF:- Retention curve flattens (doesn’t trend toward zero)
- 20-40% of users return consistently
- Highly engaged cohort forms organically
- Retention curve flattens quickly
- 40%+ of users become regular users
- Newer cohorts retain better than older ones
- Users organically tell others about your product
Advanced Retention Techniques
Unbounded Retention
Measure users who return at any point after the first event, regardless of time. This shows your “reactivation” potential.Multi-Event Retention
Measure retention of users who complete multiple key actions. Example: Users who complete onboarding AND use a core feature retain at 2x the rate.Cohort Comparison
Compare retention across:- Time periods (Q1 2024 vs Q2 2024)
- Product changes (before/after feature launch)
- Experiments (control vs variant)
Combining with Other Reports
Use retention insights in conjunction with:- Funnels: How does completing certain funnel steps affect retention?
- Insights: What actions do retained users take that churned users don’t?
- Flows: What paths do retained users follow?
Real-World Example
Scenario: A project management tool notices Week 1 retention is 55%, but Week 4 drops to 20%. Segmentation: Breaking down by “invited team members” shows:- Users who invited teammates: 72% Week 4 retention
- Users working solo: 8% Week 4 retention
- Redesign onboarding to emphasize team invites
- Add email prompts to invite teammates in Week 1
- Create solo-friendly templates for evaluation
Key Takeaways
- Retention measures whether users return to your product over time
- Use retention criteria (rolling vs calendar, on vs return) that match your analysis needs
- Frequency criteria helps identify highly engaged users
- Segment retention to understand which users stick around and why
- A flattening retention curve indicates product-market fit
- Improve retention through better onboarding, feature adoption, and re-engagement
- Combine retention insights with other reports for deeper understanding
Next Steps
- Build your first retention report tracking users from sign-up to key actions
- Identify where your retention curve drops most sharply
- Segment by key user properties to find patterns
- Create cohorts of at-risk users for re-engagement campaigns
- Set up alerts on retention metrics for key cohorts
- Review the User Engagement guide to understand what drives users to return