Retention Optimization

Retention Framework

1. Activation (Day 0-1)

The first session determines everything. Users who don't reach the "aha moment" in session 1 rarely return.

Diagnose:

• What % of users complete onboarding?
• How long until the first value moment?
• What's the drop-off point in the first session?

Optimize:

• Reduce time-to-value (show core value in < 60 seconds)
• Remove unnecessary onboarding steps
• Defer account creation until after value delivery
• Use progressive disclosure (don't overwhelm)
• Show a "quick win" in the first session

2. Habit Formation (Day 1-7)

Diagnose:

• What triggers bring users back?
• Is there a natural usage frequency?
• What do retained users do that churned users don't?

Optimize:

• Push notifications — Personalized, value-driven, not spammy
  • Day 1: "Welcome back — here's what you missed"
  • Day 3: "[Specific value] is waiting for you"
  • Day 7: "You're on a [N]-day streak!"
• Streaks & progress — Visual progress indicators
• Daily content — New content, challenges, or recommendations
• Social hooks — Friends, leaderboards, sharing

3. Engagement Deepening (Day 7-30)

Diagnose:

• Which features do power users use that casual users don't?
• What's the engagement cliff (when do users stop exploring)?

Optimize:

• Feature discovery prompts (introduce advanced features gradually)
• Personalization (adapt content/recommendations to usage patterns)
• Community features (forums, social, user-generated content)
• Achievement system (badges, milestones, rewards)

4. Long-term Retention (Day 30+)

Diagnose:

• What causes late-stage churn?
• Are there seasonal patterns?
• Do updates improve or hurt retention?

Optimize:

• Regular content updates
• Feature launches that re-engage dormant users
• Win-back campaigns for churned users
• Loyalty rewards for long-term users

Churn Prevention Tactics

Push Notification Strategy

Rules:

• Max 3-5 notifications per week
• Always provide value, never just "Come back!"
• Personalize based on user behavior
• Allow granular notification preferences
• A/B test timing and copy

Win-back Campaigns

For users who haven't opened the app in 7+ days:

1. Email (if you have it) — "We've added [feature] since you last visited"
2. Push notification — "[Specific value] is waiting for you"
3. In-app message (on return) — "Welcome back! Here's what's new"

Cancellation Flow (Subscriptions)

When a user tries to cancel:

1. Ask why (multiple choice)
2. Offer alternatives based on reason:
   • "Too expensive" → Offer discount or downgrade
   • "Don't use enough" → Show usage stats, suggest features
   • "Missing feature" → Share roadmap, offer to notify
   • "Found alternative" → Highlight unique value
3. Offer pause instead of cancel
4. Make it easy to cancel (forced retention backfires)

Output Format

Retention Diagnostic

Current State:
- Day 1: [X]% (benchmark: [Y]%) [above/below]
- Day 7: [X]% (benchmark: [Y]%) [above/below]
- Day 30: [X]% (benchmark: [Y]%) [above/below]

Biggest Drop-off: Day [N] to Day [N]
Estimated Impact: [X]% improvement = [Y] additional monthly users

Action Plan

Week 1 (Quick Wins):

1. [specific tactic with expected impact]
2. [specific tactic with expected impact]

Month 1 (High Impact):

1. [specific tactic with expected impact]
2. [specific tactic with expected impact]

Quarter 1 (Strategic):

1. [specific tactic with expected impact]
2. [specific tactic with expected impact]

Related Skills

• app-analytics — Set up retention tracking
• monetization-strategy — Retention's impact on revenue
• review-management — Retention issues surface in reviews
• app-launch — First-time user experience