A consumer messaging product is preparing to launch an in-app video-calling feature and wants to understand demand, set evaluation metrics, and manage trade-offs.

Using only the provided SQL dataset, how would you determine whether users actually need or want the new video-calling feature? If you were granted access to all internal data sources, what additional analyses would you perform to gauge demand for the feature? After release, how would you evaluate or measure the quality and success of the video-calling feature? If the primary success metric improves while a non-primary (guardrail) metric declines, how would you proceed and communicate the decision?

Consider penetration/activation rates, funnel or cohort analyses, A/B test design, metric hierarchy, guardrails, and follow-up monitoring.

Meta Data Scientist interview question: Determine User Demand for New Video-Calling Feature. {"blocks": [{"key": "f7426046", "text": "Scenario", "type": "header-two", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {...