Practice real data science interviews tuned by the interviewers who actually run them.

Spotify's app installs dropped 25% month-over-month across all platforms. The CMO flagged this in the weekly business review and wants a root cause analysis within 48 hours. Multiple factors could be at play — marketing spend changes, competitive moves, platform-level shifts, and seasonality. You need to structure the investigation, identify the most likely causes, and size their impact.

Spotify's Trust & Safety team has flagged a spike in anomalous account behavior — accounts with listening patterns consistent with credential sharing, bot-driven stream farming, or compromised accounts. These accounts inflate streaming metrics, distort royalty payouts, and degrade recommendation quality. You need to design signals for detecting different types of anomalous behavior, set appropriate thresholds, reason about false positive tolerance, and define success metrics for a fraud detection system.

Spotify's Discover Weekly playlist has been live for years, but leadership wants a rigorous success measurement framework. The playlist generates high play counts, but the team debates whether play-through rate or save-to-library rate is the better signal of genuine discovery. You need to define success metrics, segment by user type, and propose an experiment design for testing algorithm changes.

Spotify Wrapped is the company's most viral annual feature. Leadership wants a comprehensive measurement framework that covers engagement, virality, retention, and brand impact. The challenge is that Wrapped is a once-a-year event, which makes traditional A/B testing nearly impossible. You need to define success metrics, measure social sharing impact, address the December DAU surge that doesn't stick, and propose evaluation methods for a feature you can only 'launch' once per year.

Spotify's recommendation system drives 35% of all listening hours. The ML team has a new model that improves click-through rate by 8% but reduces the diversity of recommended artists by 22%. The team is debating whether to ship it. You need to evaluate tradeoffs beyond CTR, define what 'good recommendations' means for a music platform, and propose a framework for balancing exploitation (playing what users already like) with exploration (helping users discover new music).

Spotify has invested heavily in podcasts. Early data shows podcast listeners retain 25% longer than music-only users, but their music streaming hours declined 18%. Leadership wants to understand the true impact of podcasts on customer lifetime value. There's a debate about whether podcast engagement genuinely increases CLV or if podcast adopters were already higher-value users (selection bias). You need to design an approach to isolate the causal effect.

Spotify's 1-month free Premium trial converts at 46%, but 54% of trial users revert to the ad-supported tier. The Growth team wants to understand which behaviors during the trial predict conversion, what nudges could improve conversion, and whether testing different trial lengths (7-day, 14-day, 3-month) could improve outcomes. You need to build a framework for diagnosing non-conversion and designing interventions.