My first instinct was to jump straight to 'good thing, ship it' which is obviously wrong.
Start by clarifying the metric definition and time frame, then segment the 25% increase across dimensions like platform, geography, and acquisition channel to isolate where the change occurred. Finally, correlate with internal releases and external events, and validate causality with experiments or holdout groups.
Pro tip: Always distinguish between correlation and causation—a spike may coincide with a marketing campaign, but only a controlled test or a natural experiment can prove causality. Also, consider whether the increase is sustainable or a one-time bump from a viral event.
Define what 'app installs' means (new installs, re-installs, unique devices?) and confirm the time period and comparison baseline (e.g., week-over-week, year-over-year).
Break down the 25% increase by dimensions such as platform (iOS/Android), geography, acquisition channel (organic, paid, referral), and user cohort to identify where the change is concentrated.
List internal factors (product changes, marketing campaigns, pricing) and external factors (competitor actions, seasonality, press coverage) that could explain the increase in the affected segments.
Use data to correlate the timing of the increase with potential causes, and check for statistical significance. Look for leading indicators or anomalies in related metrics (e.g., app store views, sign-up rates).
If possible, design an experiment (e.g., A/B test, holdout group) or use quasi-experimental methods (difference-in-differences) to confirm that the identified cause actually drove the increase.
AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.