The Instagram framing in a PayPal interview is a little weird and I spent the first 30 seconds just mentally adjusting.
Start by clarifying the metric definition and the 20% drop (e.g., time period, cohort, platform). Then systematically break down potential causes across internal (product changes, bugs) and external (seasonality, competition) factors, using a data-driven approach to prioritize hypotheses.
Pro tip: Demonstrate maturity by acknowledging that retention is a lagging indicator and that you would first check for data integrity issues (e.g., tracking bugs, logging errors) before jumping to product conclusions.
Define what 'month-over-month retention' means (e.g., D30 retention, monthly active users) and the exact drop (e.g., from 40% to 32%). Ask about the time frame, user segments, and platforms affected.
Rule out measurement errors: verify tracking implementation, data pipeline issues, or changes in logging that could cause artificial drops.
Break down retention by user cohorts (new vs. existing), demographics, geography, platform (iOS/Android), and acquisition channels to identify where the drop is concentrated.
Investigate recent product changes (e.g., algorithm updates, UI changes), bugs, performance issues, or marketing campaigns that could negatively impact retention.
Evaluate seasonality, competitive actions (e.g., new features from TikTok), platform policy changes, or macroeconomic trends that might affect user behavior.
AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.