Estimation question

How much would it cost to set up WiFi in New York?

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 cost estimation for a large infrastructure project, breaking a vague ask into components with reasonable, stated assumptions.

Clarify scope: 'WiFi in New York' likely means citywide public outdoor WiFi coverage (like LinkGNYC), not indoor building WiFi; state this assumption upfront.
Estimate area to cover: NYC is roughly 300 square miles across 5 boroughs; assume focusing on the ~30 square miles of dense public/commercial areas (Manhattan plus dense corridors) for meaningful ROI.
Estimate access point density: assume outdoor WiFi access points cover roughly 100-150 meters radius each in dense urban settings, requiring several thousand access points for full coverage.
Estimate cost per access point: assume $10,000-15,000 per installed public AP including hardware, installation, and backhaul connectivity (a reasonable public infrastructure estimate).
Multiply: e.g., 5,000 access points x $12,000 = $60M capital cost, then add annual operating cost (maintenance, bandwidth, support) at roughly 15-20% of capex per year.
Sanity-check against a real comparable: NYC's LinkGNYC kiosk program is a known real initiative, so compare the order of magnitude and note the estimate is directional, not precise.

States the scope assumption explicitly (public outdoor coverage vs indoor, which neighborhoods) before estimating any numbers.
Breaks the cost into capex (access points, installation) and ongoing opex (maintenance, bandwidth), not just a single lump sum.
Uses defensible per-unit assumptions (AP coverage radius, cost per AP) rather than guessing a total figure directly.
Sanity-checks against a known real-world comparable (an actual NYC public WiFi initiative) to validate the order of magnitude.

Jumping to a single guessed total dollar figure with no decomposition.
Not distinguishing one-time capital cost from ongoing operating cost.
Ignoring scope ambiguity (indoor vs outdoor, which boroughs) and estimating for the wrong thing.

How would the estimate change if you only covered subway stations instead of streets?
What's the biggest cost driver you'd try to reduce first?
How would you phase this rollout to manage the capital cost?