What you need to know
- Activation Rate shows how many users reach the "aha moment", for good apps, it's 20-40%
- Day 1 Retention below 25%? Something's fundamentally wrong
- The first 7 days determine a user's entire lifecycle
- DAU/MAU ratio (Stickiness) should be at least 20%
- If your activation event doesn't correlate with better retention, you're tracking the wrong thing
Every morning at 8:15, I open my analytics dashboard. Coffee in hand, Slack muted. The next 10 minutes determine whether I start my day relaxed or immediately go into fire-fighting mode.
After three years as a self-employed developer tracking apps for startups and scale-ups, I've learned: there are exactly 5 numbers that really matter. Everything else is vanity, nice for investor decks but useless for product decisions.
This guide shows you which activation metrics you should track, what benchmarks are realistic, and which tools work best for different use cases. How Duolingo uses retention at scale is in how Duolingo scales monetization.
What Are Activation Metrics Anyway?
Activation metrics measure whether a new user understands and uses your app's value. The difference from pure engagement metrics: it's not about how often someone opens the app, but whether they've done the one thing that turns them into an active user.
PostHog defines it well:
"Activation metrics are a set of actions a new user performs that correlate with greater retention."
That's the key: correlation with retention. If your activation event doesn't lead to better retention, you're tracking the wrong thing.
Activation vs. Engagement vs. Retention
- Activation: Has the user experienced the core value? (e.g., first photo posted, first order placed)
- Engagement: How intensively do they use the app? (Sessions, screen views, feature usage)
- Retention: Do they come back? (Day 1, Day 7, Day 30 retention)
All three are connected, but activation is the lever. Improve activation and retention follows automatically. How to get your first users without a budget is in app growth without a marketing budget.
The 5 Numbers I Check Every Morning
1. Activation Rate
What it measures: Percentage of users who reach the activation event within a defined time window (usually 7 days).
Formula: ``` Activation Rate = (Users with Activation Event / New Users) × 100 ```
Benchmarks by app category:
- SaaS B2B: 20-40%
- E-Commerce: 15-25%
- Social/Community: 10-20%
- Gaming: 15-30%
- Fintech: 25-35%
My take: Below 15% is a red flag. Either your onboarding is broken or your activation event is wrongly defined.
2. Day 1 Retention
What it measures: How many users come back the day after installation?
Formula: ``` Day 1 Retention = (Users active on Day 1 / Installs) × 100 ```
Benchmarks 2024/2025 (per AppsFlyer Retention Report):
- Average all apps: 21-25%
- Top 10% apps: 35-45%
- Gaming: 25-30%
- Social: 20-25%
- E-Commerce: 15-20%
- Finance: 22-28%
My take: Day 1 retention is your canary in the coal mine. If it suddenly drops by 5+ percentage points, something is fundamentally wrong, new bug, bad campaign, or a store update broke something.
3. Day 7 Retention
What it measures: The most important mid-term indicator, how many users stay after a week?
Benchmarks:
- Average: 10-15%
- Top 10%: 20-30%
- Gaming: 8-12%
- Social: 15-20%
- Utilities: 18-25%
My take: Day 7 is the reality check. Day 1 can be artificially boosted by push notifications. Day 7 shows whether there's genuine interest.
4. DAU/MAU Ratio (Stickiness)
What it measures: How often do active users come back? A proxy for habit formation.
Formula: ``` Stickiness = (Daily Active Users / Monthly Active Users) × 100 ```
Benchmarks:
- Excellent: >25%
- Good: 20-25%
- Average: 10-20%
- Concerning: <10%
Context:
- Facebook: ~50% (Outlier, not realistic for 99% of apps)
- WhatsApp: ~70%
- Average app: 12-15%
My take: 20%+ is the goal. Below 10% means your users forget you between sessions. Time for re-engagement strategies.
5. Time to Activation (TTA)
What it measures: How long does it take a user from install to activation event?
Benchmarks:
- Ideal: <24 hours
- Acceptable: <72 hours
- Problematic: >7 days
My take: Shorter is better. Long TTA correlates strongly with higher churn. If your average TTA is over 3 days, your onboarding is too complex or your activation event too hard to reach.
The First 7 Days: The Critical Window
The first week decides the life or death of your user relationship. Here's what should happen in each time window:
Day 0-1: The First Impression
- Goal: Guide user to activation event
- Optimize: Onboarding flow, first user experience
- Measure: Onboarding completion rate, time to first value
Day 2-3: The Reality Check
- Goal: Trigger second and third visit
- Optimize: Push notification strategy, content freshness
- Measure: Return visit rate, feature discovery
Day 4-7: Habit Formation
- Goal: Establish routine
- Optimize: Demonstrate use cases, personalization
- Measure: Session frequency, feature depth
What You Should Track (Event List):
``` // Onboarding Events
- signup_started
- signup_completed
- onboarding_step_1_completed
- onboarding_step_2_completed
- onboarding_completed
// Activation Events (app-specific)
- first_core_action_completed
- profile_completed
- first_purchase
- first_content_created
// Engagement Events
- session_started
- feature_x_used
- notification_opened
- share_completed ```
Tool Comparison: PostHog vs. Mixpanel vs. Firebase
I've worked with all three. Here's my honest assessment:
PostHog
Strengths:
- Open source, self-hosting possible
- Autocapture for quick start
- Session recordings included
- Generous free tier (1M events/month)
- SQL queries for power users
Weaknesses:
- UI can be overwhelming
- Mobile SDKs slightly less mature
- Feature flags cost extra
Best for: Startups with technical founders who want control over their data
Price: Free up to 1M events, then ~$0.00045/event
Mixpanel
Strengths:
- Best cohort analyses
- Intuitive UI for non-technical users
- Strong funnel visualizations
- Excellent for A/B test analysis
Weaknesses:
- Expensive at scale
- Autocapture less robust
- Vendor lock-in
Best for: Product teams with budget who need clear dashboards for stakeholders
Price: Free up to 20M events (but limited features), then from $20/month
Firebase Analytics
Strengths:
- Free and unlimited
- Native Google Ads integration
- Easy setup for mobile apps
- Real-time analytics
Weaknesses:
- Limited retention analyses
- Max 500 event types
- Export only via BigQuery
- Less granular cohort analyses
Best for: Early-stage startups, apps with Google Ads focus
Price: Free (BigQuery export costs extra)
My Recommendation
| Stage | Tool |
|---|---|
| Pre-Launch / MVP | Firebase Analytics |
| Post-Launch / Traction | PostHog or Mixpanel |
| Scale (>100k MAU) | Mixpanel + Data Warehouse |
How to Find Your Own Activation Metric
The biggest trap: Choosing the activation event based on gut feeling. "Of course account creation is important!", but often it's not.
The PostHog Approach (which I use):
- Gather hypotheses: List 15-20 events that could be "activation"
- Form groups: Combine 3-5 events into groups
- Test retention: Which group correlates most strongly with 3-month retention?
- Validate: A/B test with optimized onboarding
Real-World Example:
For a fitness app, we had three hypotheses:
- H1: Profile fully completed
- H2: First workout logged
- H3: Workout + photo uploaded
The analysis showed: H3 had 2.4x better 90-day retention than H1. The "aha moment" wasn't the profile, but sharing the achievement.
Common Mistakes (All of Which I've Made)
1. Tracking Too Many Metrics
You don't need 50 KPIs. You need 5 that you truly understand.
2. Activation Event Too Early in the Funnel
"App opened" is not an activation event. "Account created" usually isn't either. Look for the moment where real value emerges.
3. Ignoring Cohort Analyses
Aggregate numbers lie. A new marketing campaign can ruin your retention while overall numbers look fine.
4. No Segmentation
iOS vs. Android, organic vs. paid, geo segments, they behave completely differently.
5. Tool Hopping
Stick with one tool. Historical data is gold. Every switch costs you 3-6 months of comparability.
Actionable Takeaways
This Week:
- Define your activation event (or validate the existing one)
- Set up a dashboard with the 5 core metrics
- Compare your numbers with the benchmarks above
This Month:
- Analyze cohorts from the last 3 months
- Identify the biggest drop-off in onboarding
- Test one hypothesis to improve Day 1 retention
This Quarter:
- Build an alerting system for critical metric changes
- Segment retention by acquisition channel
- Document learnings for the team
Conclusion
Activation metrics aren't rocket science, but they require discipline. Most apps don't fail because of bad products, but because of a lack of understanding of what makes users stay.
Start with the 5 metrics I described above. Build a morning routine where you check them. And most importantly: act based on data, not gut feeling.
Your users tell you every day what works and what doesn't. You just need to listen.
