I started by breaking down the user types: someone loading a map tile vs.
Start by clarifying the scope of Google Maps (e.g., which features and user segments) and then break down bandwidth into major components like map tile loading, search queries, and navigation. Use a top-down estimation: estimate daily active users, average sessions per user, and data transferred per session, then sum across features to get total daily bandwidth.
Pro tip: Focus on the dominant bandwidth drivers (e.g., map tiles and navigation) and state your assumptions clearly; interviewers care more about your structured thinking than precise numbers.
Ask clarifying questions to define which Google Maps features (e.g., mobile app, web, navigation, Street View) and user segments to include. State assumptions about daily active users, session frequency, and average session duration.
Break down bandwidth into key activities: map tile loading, search/geocoding requests, navigation updates, and optionally Street View or traffic data. Prioritize the components that likely dominate usage.
For each component, estimate the number of requests per user per session and the average data size per request. For example, map tiles: number of tiles per session × average tile size.
Multiply usage metrics by the number of daily active users to get total daily data volume. Sum across components to get overall daily bandwidth, then convert to average throughput (e.g., Gbps).
Validate the estimate by comparing to known scales (e.g., total internet traffic) or breaking down further. Summarize key drivers and acknowledge uncertainties.
AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.