Metrics question

70% of shoppers of an eCommerce site are reviewing the site's return policy page prior to shopping. What do you do?

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

Tests interpreting an ambiguous signal correctly, distinguishing healthy shopping confidence from return-policy anxiety, before proposing any action.

Don't assume the number is bad: 70% checking the return policy could be healthy diligence, or anxiety suppressing conversion.
Segment by outcome: compare conversion and return rate for shoppers who visited the policy page versus those who didn't.
Check timing: did this spike after a negative PR moment, a policy change, or new less-trusting traffic, suggesting a trust problem.
If it correlates with lower conversion, treat it as friction: surface key policy facts directly on the product page.
If it correlates with similar or better conversion, treat it as a positive signal and make the policy even more prominent.
Either way, A/B test inline policy info versus click-through, measuring both conversion and post-purchase return rate.

Refuses to label the 70% figure good or bad and instead defines the exact analysis needed to know which it is.
Considers timing and a specific trust-eroding event as a possible root cause rather than assuming steady behavior.
Proposes a concrete, testable fix (inline policy info) that addresses the friction interpretation directly.
Holds both interpretations open until data resolves it, showing calibrated, non-reactive judgment.

Assuming the stat is automatically bad and rushing to hide the return policy page.
Assuming it's automatically good with no data cited to support that read.
No proposed segmentation or test, just a general reaction to the number.

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