Metrics question

Measure the success of the save feature in LinkedIn.

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

Metrics thinking for a low visibility, utility feature where usage patterns reveal intent, not just raw counts.

Clarify the save feature's purpose: users bookmark posts, articles, or jobs to revisit later, signaling delayed intent rather than immediate action.
Define the north star: percentage of saved items that are later revisited or acted on, like a saved job actually applied to, not just total saves.
Add a frequency metric: percentage of active users who use the save feature at least weekly, showing habitual adoption.
Add a content type breakdown: which categories of saved content, like jobs versus articles, drive the most revisits, informing where to invest further.
Add a guardrail: saved items that are never revisited within 30 days, since a high abandonment rate signals the feature isn't delivering real value.
Set an illustrative target, for example assuming a goal of 40 percent of saved job postings being revisited within a week.

Correctly identifies that revisit or action rate, not raw save count, is the true measure of value for a save feature.
Breaks down usage by content type, showing awareness that jobs and articles serve very different user intents.
Gives an illustrative number, a 40 percent revisit target, grounding the metric in a checkable goal.
Names abandoned saves as a guardrail, a subtle but meaningful signal of low feature value.

Measuring only total saves, which overstates value since many saved items are never revisited.
Treating all saved content types the same, missing that jobs and articles likely have very different revisit patterns.

How would you re-engage users with items they saved but never revisited?
How would you know if the save feature is cannibalizing more valuable actions like immediate applying or sharing?