Metrics question

You launched a new feature that determines whether a Facebook message was read by a recipient. What metrics would you collect? How would you know the feature was a success?

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

Metric design for a feature with a clear social or psychological trade off, transparency versus pressure to respond.

State the feature's value, senders get certainty about whether their message was seen, reducing anxiety about being ignored.
Define the primary success metric, adoption or opt in rate if optional, or reduction in duplicate follow up messages sent.
Define a guardrail metric, recipient discomfort or complaint rate, since knowing a message was read but not replied to can create social pressure.
Consider segmenting by relationship type, since read receipts might be welcomed between close friends but unwanted in more distant contexts.
Propose a way to test, A B test with an opt out available, measuring both sender satisfaction and recipient discomfort signals like disabling the feature.

Names the real psychological trade off explicitly, sender certainty versus recipient pressure, rather than treating this as a simple feature with only upside.
Proposes a specific guardrail metric, rate of users disabling the feature or reporting discomfort, to catch the downside directly.
Suggests segmenting by relationship closeness, since the feature likely performs very differently between close friends and looser acquaintances.

Only measuring adoption or usage with no guardrail for the social pressure this kind of feature can create.
Treating all relationship types the same, ignoring that context changes how welcome this feature is.
No mention of giving users control, like the ability to disable it.

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