Metrics question

You are a PM at a food delivery firm. Your data analyst comes up to you and tells you that there is a spike at breakfast, dinner and lunch time. However, at lunch, the conversion is 2% when compared to dinner and breakfast which is 10%. How would you go about solving this problem?

Practice this question out loud. An AI interviewer asks it, follows up like a real interviewer would, and scores your answer. Type or speak.

What this question tests

Diagnosing a segment specific conversion gap using funnel thinking and mealtime context.

Confirm the numbers: verify the 2 percent versus 10 percent gap is measured consistently across the same funnel steps.
Consider mealtime-specific behavior: lunch orders are often time constrained, from work, so slower decision making could hurt conversion.
Segment by restaurant supply: check if fewer restaurants are open or available for delivery during the lunch window.
Segment by delivery speed: lunch orders often need to arrive within a tight window, so slower ETAs could cause more drop off.
Check pricing and promotions: verify if lunch has fewer active discounts compared to dinner and breakfast.
Propose the top hypothesis, likely restaurant availability or ETA during lunch, and outline a data check before recommending a fix.

Frames the gap through mealtime context, noting lunch orders are often time pressured from work, unlike more relaxed dinner ordering.
Proposes checking restaurant supply and ETA specifically during the lunch window as a concrete, checkable hypothesis.
Considers promotion parity across mealtimes rather than assuming the same discounts apply evenly all day.
Sequences investigation before proposing a fix, avoiding a guess based only on the two numbers given.

Proposing a fix immediately without investigating why lunch specifically underperforms.
Ignoring restaurant supply and delivery time constraints unique to the lunch rush.

Chapters of the AI PM course that teach what this question tests.