{"blocks": [{"key": "e50f38cd", "text": "Scenario", "type": "header-two", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "39ac5a82", "text": "DoorDash operates a three-sided marketplace (consumers, dashers, merchants). Leadership is debating whether to shift dasher compensation from per-order payments to an hourly (time-based) model and also wants a framework for diagnosing sudden drops in key metrics.", "type": "unstyled", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "67f47b2c", "text": "Question", "type": "header-two", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "fcaea693", "text": "How would you determine whether DoorDash should pilot paying dashers by time instead of by order? Describe the experiment design, success metrics, and how you would control for marketplace effects across consumers, merchants, and dashers. Suppose a critical marketplace metric (e.g., order completion rate) suddenly declines. Walk through a structured process to identify the root cause and quantify its impact.", "type": "unstyled", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "3bf715c8", "text": "Hints", "type": "header-two", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "11078fb8", "text": "Cover A/B test setup, sampling, guardrail metrics, segment analysis, and hypotheses tree for root-cause analysis; address external factors and data instrumentation issues.", "type": "unstyled", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}], "entityMap": {}}