Strava offline recording: estimate the local storage runway for workout data
Company: Strava
Role: Product Manager
Category: Product / Decision Making
Difficulty: medium
Interview Round: Other
Strava's app records workouts such as runs and rides on the phone, for example a run's GPS track, and uploads each activity to the service once the phone has a connection. Suppose the phone stays offline, so every activity the app records has to remain on the device. Estimate how long the app could keep storing this data locally, and walk through how you reached the number.
```hint Start from one sample
Work out what a single recorded point holds and how often the app writes one, then scale up to an hour and a day of use. Keep each input separate so you can vary it for different users.
```
### Clarifying Questions
- Does "how long" mean until the phone's free storage is full, or until a storage budget the app reserves for itself is used up?
- Which data is recorded: the GPS track only, or also heart rate, cadence and power from paired sensors, plus photos attached to activities?
- How much does the user record each day: one hour-long workout, several hours, or continuous all-day tracking?
- How much free storage should we assume the phone has?
### What a Strong Answer Covers
- A clear equation linking free storage, bytes per recorded sample, sampling rate and hours recorded per day.
- Explicit, labeled assumptions for each input, with reasoning for the per-sample size and the recording frequency.
- Correct arithmetic carried through to a time figure, with a range for light, heavy and extreme users.
- A sanity check against something familiar, such as the size of an exported workout file.
- What the result implies for the product: which limit (storage, battery, or the risk of losing data that has not synced) actually matters first.
### Follow-up Questions
- How does the estimate change if users attach photos to activities or download offline maps for their routes?
- When the phone reconnects after a long stretch offline, how long would uploading the backlog take, and how should the app handle it?
- What would you build so an offline workout is never lost if the app crashes or the phone dies mid-recording?
Overview: An estimation question about Strava's offline mode: how long a phone could keep storing recorded workout data, such as GPS tracks and sensor readings, before running out of space. It tests building an equation from sample size, sampling rate and daily usage, labeling assumptions, checking the arithmetic and drawing product conclusions.