State the objective in one line, split the problem into two or three mutually exclusive buckets, and pick the riskiest bucket first. Sketch a quick issue tree as you talk. If you freeze, paraphrase the question to buy seconds and ensure alignment before proceeding decisively.
Lead with a sharp, falsifiable claim, such as demand, not price, drives losses. Then list two or three tests you would run immediately, naming data you need and expected signals. Update publicly as evidence appears, demonstrating flexibility without abandoning ownership of your initial direction.
Tell the interviewer your plan and time budget: two minutes to structure, three to analyze, one to synthesize. Use a visible countdown on paper marks. When time slips, cut scope deliberately, narrate what you are dropping, and commit to a crisp recommendation.






All Rights Reserved.