Debug a spike in complaints from students on Discover Weekly

Problem Statement Description

Product context: Spotify is an audio streaming company; its products include music, podcasts, audiobooks, playlists, recommendations, creator tools, subscriptions, and ads.

Spotify has seen a sudden spike in complaints from students about Discover Weekly, the personalized playlist experience that refreshes weekly with recommended tracks. The issue appears concentrated in the student segment, but the underlying cause is unknown: it could relate to recommendation quality, playlist freshness, availability, playback behavior, app experience, expectations, or how complaints are being captured.

Your task is to frame and investigate the anomaly as a product manager. Focus on how you would determine whether this is a real user-experience regression, a measurement or reporting artifact, or a segment-specific behavior change. Consider that students may have distinct listening patterns, subscription status, devices, geographies, school schedules, social trends, and price sensitivity compared with the broader Spotify population.

You should not jump directly to a fix. Instead, describe how you would structure the investigation, what data you would inspect, how you would segment the problem, what hypotheses you would test, and how you would decide whether immediate mitigation is needed.

The experience should consider:

- How to define the complaint spike, including baseline period, complaint rate denominator, severity, and channels such as in-app feedback, support tickets, app-store reviews, or social media.

- Segmentation by student plan status, age range where available, geography, campus regions, device, platform version, free vs paid tier, and new vs existing Discover Weekly users.

- Discover Weekly-specific signals such as playlist generation success, refresh timing, recommendation relevance, skips, saves, hides, repeat listening, completion rate, and “not interested” actions.

- Instrumentation checks to confirm whether complaint volume, tagging, routing, or classification changed before assuming a product issue.

- Hypotheses across recommendation model changes, catalog availability, licensing gaps, playlist refresh failures, UI changes, ads experience, latency, or student-specific marketing campaigns.

- Evidence needed to distinguish a student-only issue from a broader Discover Weekly regression that is merely more visible among students.

- Short-term mitigation options, user communication needs, escalation paths, and criteria for rolling back or pausing recent changes.

- Longer-term prevention through monitoring, alerting, experiment guardrails, complaint taxonomy, and segment-level health dashboards.

The goal is to demonstrate a clear RCA approach: define the anomaly, validate the data, isolate affected cohorts, generate testable hypotheses, use evidence to prioritize investigation, and outline how Spotify should mitigate user impact while preventing similar Discover Weekly issues from recurring.