Estimation question

How long will it take to store Strava's data on the phone while offline?

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 structured estimation combining data-size reasoning with a systems constraint (offline caching), rather than a pure market-sizing exercise.

State the assumption: this is asking how much of a user's workout history and route data Strava needs to cache locally to support offline access, not literally all server-side data.
Estimate per-activity data size: GPS points every few seconds for a typical hour-long activity, plus metadata (pace, heart rate), roughly a few hundred KB per activity.
Estimate the offline window needed: likely a rolling recent window (last 30 to 90 days of activities) rather than a user's entire multi-year history, since offline caching is about immediate access, not full backup.
Multiply activities per period by size per activity to get a total local storage estimate, then state it as a range given the uncertain assumptions.
Address the actual question, how long to store, by reasoning about cache eviction policy and sync timing rather than treating it purely as a data-size problem.
State the estimate clearly labeled as built on stated assumptions, not measured fact.

Reframes the ambiguous question into a concrete, answerable one, offline cache retention policy, and states that framing explicitly.
Breaks the estimate into per-activity size times activity frequency times retention window, showing clear decomposition.
Uses illustrative numbers as assumptions: assume 300 KB per activity, 20 activities a month, a 60-day rolling cache, roughly 12 MB of cached data.
Distinguishes local device storage constraints from server-side unlimited history, correctly scoping what offline actually requires.

Treating this as a pure data-volume estimate without addressing the offline caching and retention angle the question implies.
Giving a number with no shown decomposition or stated assumptions.
Ignoring that offline storage on a phone is a meaningfully constrained resource compared to server storage.

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