← Meta Interview Insights

Meta·Product Manager·Hiring Manager Screen·Senior

Senior
Apr 2026

Summary

Meta PM interview, one question about diagnosing a metric drop on Instagram Stories. Short and focused, felt more like a screen than a full loop.

Questions Asked (1)

Q1

You're the PM for Instagram Stories and creation is down 5%. How do you diagnose what's going wrong?

Root Cause AnalysisProduct Analytics & MetricsProduct Sense & Ideation
Author's notes

I went straight to segmentation and probably moved too fast.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying what 'creation' means and how it's measured, then segment the metric to isolate where the drop is concentrated. Form hypotheses across internal and external factors, and prioritize the most likely causes to investigate with data and qualitative research.

Pro tip: Always validate that the drop is real and not a data instrumentation issue before diving into product changes—many 'metric drops' are actually logging bugs or definition changes.

1. Clarify the metric and scope

Define 'creation' precisely (e.g., number of stories created, users creating, creation attempts) and confirm the 5% drop is statistically significant and not due to a tracking change.

2. Segment the data

Break down the metric by dimensions like platform, region, user cohort, entry point, and time to identify where the drop is concentrated and when it started.

3. Generate hypotheses

List potential internal causes (e.g., recent releases, bugs, UI changes) and external causes (e.g., seasonality, competitor launches, platform policy changes).

4. Prioritize and investigate

Use data to validate or eliminate hypotheses, starting with the most likely and impactful ones. Leverage logs, A/B tests, and user research to confirm root cause.

5. Recommend next steps

Propose immediate fixes if a bug is found, or further research if the cause is unclear. Outline how to monitor and prevent future drops.

Key Points to Mention

  • Define 'creation' clearly and check for data quality issues first.
  • Segment by platform, region, user type, and entry point to localize the drop.
  • Consider both internal factors (recent code changes, bugs) and external factors (seasonality, competition).
  • Use a mix of quantitative analysis (funnels, cohorts) and qualitative methods (user interviews, session replays).
  • Prioritize hypotheses based on likelihood and potential impact.
  • Propose a clear action plan: fix if known, or further investigation if not.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.