Metrics question

How would you measure success of Twitter Lists?

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

Whether the candidate can build a metric tree for a niche curation feature and separate light trial from a habit that actually sticks.

Confirm with the interviewer whether success here means discovery, retention, or content quality, since Lists serve several jobs.
State the product goal: Lists should give users a narrower, higher-signal feed than the main algorithmic timeline.
Pick a north star such as weekly active list viewers as a percent of weekly active Twitter users.
Build input metrics: lists created per creator, accounts added per list, and list feed session length.
Add a retention metric: share of list creators who revisit their own list within seven days.
Set guardrails so time on the home timeline and ad impressions do not drop as a side effect.

Separates creation from consumption, assuming for example 5 percent of active users create a list but a far smaller share return weekly, and treats the return rate as the real signal.
Splits public curator lists, which act like content marketing, from private organizational lists, since they succeed on different metrics.
Tracks list feed depth, tweets actually scrolled per session, not just opens, since an unread list has not delivered value.

Reporting lists created as the headline metric without checking whether anyone reads them later.
Ignoring cannibalization of the main timeline and ad load.

How would you separate power users like journalists from casual list creators?
What experiment would show Lists cause retention rather than just correlate with it?