Business Context

Duolingo is testing a more aggressive growth strategy for its mobile app: faster onboarding, more referral prompts, and more frequent win-back notifications. In the last month, weekly new user signups increased from 1.8M to 2.2M (+22%), and Day 0 lesson-start conversion improved from 62% to 68%. However, the Head of Growth is concerned that optimizing only for acquisition and activation may hurt long-term product health.

Metric Scenario

Leadership wants a clear framework for guardrail metrics to use alongside the primary growth KPI. The current proposal is to optimize for weekly activated users (new users who complete at least one lesson in their first 24 hours). Early results show: Day 7 retention fell from 28% to 25%, average lessons per retained user dropped from 5.4 to 4.9 per week, unsubscribe rate from push notifications rose from 1.9% to 3.1%, and premium subscription conversion stayed flat at 4.2%.

Requirements

1. Define the primary growth metric and explain why it is not sufficient on its own.
2. Propose 4-6 guardrail metrics that should be monitored during growth optimization.
3. For each guardrail, explain what bad tradeoff it protects against.
4. Show how you would decompose a decline in one guardrail metric to find the root cause.
5. Recommend which metrics should trigger a rollback, investigation, or continued rollout.

Data Available

• user_signup_funnel: signup_start, signup_complete, lesson_start, referral_source, country, platform
• lesson_events: user_id, lesson_id, start_time, complete_time, streak_status, session_duration
• notifications_log: notification_type, send_time, open_time, opt_out_flag
• subscriptions: trial_start, trial_convert, subscription_status, plan_type
• user_retention_daily: user_id, signup_date, active_date, days_since_signup