Что такое вовлечение?

Engagement (вовлечение): определение, метрики, стратегия

Краткий ответ

Engagement = "Как часто и насколько глубоко пользователи взаимодействуют с продуктом."

Это не одна метрика. Это combination:

• Frequency (как часто заходит)
• Duration (как долго использует)
• Depth (глубину features uses)
• Satisfaction (нравится ли им)

Почему это важно

Диагональ успеха: Signup → Activation → Engagement → Retention → Revenue

Если engagement низкий, все падает:

• Низкий engagement → Churn высокий
• Чем меньше engagement → Меньше chance upgrade
• Меньше chance refer (не в любви с product)

Правило: Более engaged users = выше LTV = выше profitability.

Типы engagement

1. Daily active (DAU) engagement

Метрика: % пользователей активных в день

Пример:

• 10k monthly users
• 3k daily active (DAU)
• DAU ratio: 30% (healthy is 20-40%)

Means: 3 из 10 users open app every day.

2. Session duration

Метрика: Как долго user в приложении per visit

Пример:

• App A: Average 5 minutes per session
• App B: Average 15 minutes per session
• App B has 3x better engagement

Means: Users more absorbed in App B.

3. Feature adoption

Метрика: % users using key feature

Пример:

• Core feature adoption: 80% (most use)
• Secondary feature: 40% (many ignore)
• Advanced feature: 5% (power users only)

Means: Core feature sticky, advanced underused.

4. Return rate

Метрика: % who come back after first visit

Пример:

• Day 1 retention: 70% (70% return day 2)
• Day 7 retention: 30% (30% still active day 7)
• Day 30 retention: 15% (15% still active day 30)

Means: Healthy retention for typical app. But could improve.

5. Actions per session

Метрика: Average actions (clicks, searches, posts) per session

Пример:

• Twitter: Average 10-15 interactions per session
• LinkedIn: Average 3-5 interactions per session
• Why? Twitter designed for more interaction

Means: More actions = more invested in product.

6. Emotional engagement

Не metric, but important.

Measured by:

• NPS (would you recommend?)
• Sentiment (app reviews)
• Love (does user love product, or tolerate it?)

Pример:

• App with high DAU but low NPS: Users use it but don't love it
• App with lower DAU but high NPS: Users love it, but infrequent use case

Means: Engagement isn't just frequency, but emotion too.

Как я измеряю engagement

1. Engagement score (custom metric)

Formula:

Engagement Score = (DAU rate × 30%) + (Session duration × 25%) + 
                   (Feature adoption × 30%) + (NPS × 15%)

Example:
- DAU rate: 30% (target 20%, so 150%) → Weight 30% → 0.45 point
- Session duration: 10 min (target 8 min, so 125%) → Weight 25% → 0.3125 point
- Feature adoption: 60% (target 60%, so 100%) → Weight 30% → 0.30 point
- NPS: 45 (target 50, so 90%) → Weight 15% → 0.135 point
- Total: 1.1925 (can compare month to month)

2. Cohort analysis

Question: Do new users more engaged than old users?

Cohort analysis:
Jan cohort: DAU 45% (first week), 30% (week 2), 20% (week 4)
Feb cohort: DAU 42% (first week), 28% (week 2), 18% (week 4)

Pattern: Jan cohort slightly more engaged. Why?
- Jan had better onboarding? 
- Better product features?
- Or better marketing (cohort quality)

Find root cause → improve for future cohorts.

3. Feature engagement

By feature:

Feature | % users | Avg uses/month | Stickiness
---|---|---|---
Core search | 95% | 45 | High
Filters | 60% | 12 | Medium
Saved items | 40% | 8 | Medium
Sharing | 15% | 2 | Low

Conclusion:
- Search core engagement driver
- Sharing not working (15% adoption, 2 uses/month)
- Should improve sharing or remove it

Стратегия улучшения engagement

1. Fix onboarding

Why: Poor onboarding = users don't see value → low engagement

Example from my work: Онбординг был скучный (10 шагов). Результат: 30% activation.

Оптимизировали:

• Сократили до 3 шагов
• Добавили интерактивные элементы
• Показали value immediately

Результат: 70% activation (+140%!)

2. Reduce friction

Why: Friction kills engagement. Remove it.

Examples:

• "Sign up required before trying?": Remove (let try first)
• "Long form required?": Use shortcuts (ask more later)
• "Hard to find feature?": Make discoverable

3. Personalization

Why: Generic product for everyone = engagement okay. Personalized product = engagement great.

Example: Recommendation system:

• Before: Random content feed
• After: Personalized to user interest
• Engagement: Session duration 8 min → 15 min (+87%)

4. Habit loops

Why: Best engagement is habitual use (like checking email).

Framework:

1. Trigger (notification, email, cue)
2. Action (user opens app)
3. Variable reward (something interesting inside)
4. Investment (user puts in effort)

If loop works, user returns regularly (habit).

Example: App sent digest email (trigger). User opened app (action). Found personalized feed (reward). Rated content (investment). Result: User now checks daily (habit).

5. Progressive disclosure

Why: Showing all features at once = overwhelm. Show gradually.

Example: Onboarding shows only 1 core feature first. After user masters it, unlock 2nd feature. After that, unlock advanced features.

Result: Users not overwhelmed, learn progressively.

6. Community/social

Why: People engage with people, not features.

Examples:

• Twitter: Engaging with people (followers, followers)
• Reddit: Community (subreddits, discussion)
• LinkedIn: Network (connections, posts)

If you add social layer, engagement rises.

Engagement vs retention (важное различие)

Engagement: How much user interacts when they ARE in product. Retention: Whether user comes back.

Пример:

App A:
- DAU: 20% (low retention)
- Session duration: 30 min (high engagement)
→ Users engaged but don't come back
→ Maybe product solves temporary need

App B:
- DAU: 60% (high retention)
- Session duration: 3 min (low engagement)
→ Users return often but don't stay long
→ Maybe product solves quick need (check email)

Best case: High engagement + High retention

Red flags (low engagement)

❌ DAU dropped 20% in a month

• Product degradation? New bug?
• Competitor launched?
• Seasonal (expected)?
• Need investigation

❌ Session duration declining

• Users bored?
• Feature changed (worse UX)?
• Churn early warning

❌ Feature adoption flat

• Feature not discoverable?
• Feature not valuable?
• Need A/B test or redesign

❌ NPS declining while DAU stable

• Users tolerate product, don't love it
• Churn risk (will leave if alternative appears)

Моя engagement стратегия

Для продуктов которыми я управлял:

Phase 1: Measure engagement baseline

- Define key engagement metrics (not just DAU)
- Baseline current state
- Set targets (realistic improvement)

Phase 2: Diagnose engagement problems

- Cohort analysis (which users more engaged)
- Feature analysis (which features drive engagement)
- Churn analysis (when do users leave)
- Customer interviews (why do they leave?)

Phase 3: Prioritize fixes

- Biggest impact: Onboarding (affects all new users)
- Second: Core feature UX (most used feature)
- Third: Personalization (customized experience)
- Fourth: Community (social engagement)

Phase 4: Experiment

- A/B test onboarding
- Measure impact on engagement
- If win, roll out
- If loss, try different approach

Phase 5: Monitor & iterate

- Track engagement metrics weekly
- Report progress to stakeholders
- Continue improving

Engagement по типам продуктов

High engagement expected:

• Social (Twitter, TikTok): 60%+ DAU
• Gaming: 40%+ DAU
• Communication (WhatsApp): 70%+ DAU

Medium engagement expected:

• Productivity (Notion): 30-40% DAU
• E-commerce: 10-20% DAU
• News: 20-30% DAU

Low engagement expected (but should maximize):

• Tax software: 2% DAU (seasonal)
• Banking: 5-10% DAU (used as needed)
• Healthcare: 5% DAU (infrequent)

Итоговый принцип

Engagement не is end-goal, it's means to end-goal (retention, revenue).

Better engaged users = stick around = pay more = refer others.

My job as PM: Make product so engaging that users can't resist coming back.

Not through addiction (Facebook/TikTok way), but through genuine value (Notion way).

That's sustainable engagement.