"Estimate Uber rides per day in NYC"

The failure mode is population times a fraction, with no supply-side check and no acknowledgment that New York City has the heaviest subway ridership in North America. Interviewers asking this in 2026 already know the real number from TLC data. What they want is whether you understand the system: driver supply caps, subway substitution, congestion pricing, and regulatory constraints that make volume growth structurally different from every other US city.

What makes NYC harder than “Uber rides in the US”

Four constraints do not exist anywhere else:

• TLC plate cap. NYC froze new for-hire vehicle plates in 2018 and has extended the freeze repeatedly. Supply is legally capped. When demand spikes, prices rise rather than volume, so you cannot grow your way out with driver recruitment.
• Subway substitution. The MTA runs about 3.4 million boardings per day at under $3. A commuter traveling from Astoria to Midtown takes the N train, not an Uber. NYC’s addressable Uber market is much smaller than population alone implies.
• Congestion pricing. Launched January 2025, the $9 toll for passenger vehicles entering Manhattan south of 60th Street shifted Uber demand inside the zone toward higher-value trips and suppressed some discretionary rides.
• Borough asymmetry. Manhattan and Queens (airport runs) generate a disproportionate share of trips. Staten Island is trivially small. A city-wide per-capita approach obscures this.

Structure a strong answer

State your scope, pick two methods, run them independently, and converge.

Scope: NYC five boroughs, all Uber products except Eats, typical weekday.

Demand-side build-up: NYC adult population is roughly 6.5 million. Subtract transit-primary commuters: MTA logs 3.4 million daily boardings, and accounting for repeat riders, roughly 2 million distinct adults are subway-first for most trips and rarely Uber for routine commutes. Addressable adults: 4.5 million. About 35% have the app and use it at least occasionally, giving 1.6 million monthly active users. Daily active riders on a weekday: 15% of MAU, or 240,000 riders, taking about 2 rides each, producing 480,000 rides. Net airport trips (after deduplication with the MAU base) add about 30,000. Adjusted demand total: roughly 510,000.

Supply-side check: NYC has about 90,000 active TLC-licensed FHV drivers. Uber holds roughly 70% of the rideshare market, implying about 63,000 Uber-affiliated drivers. At 55% time-based utilization across a 20-hour service window, each driver completes about 5 to 6 trips per shift. Many work split or extended shifts; effective trip multiplier is about 1.6. 63,000 drivers × 5.5 trips × 1.6 shift factor gives roughly 554,000 trips.

Both methods land within 10% of each other. Central estimate: 550,000 Uber rides per day in NYC on a typical weekday.

Weekend peak (Friday and Saturday nights): 30 to 40% above weekday average, so roughly 700,000 to 770,000.

Real anchor: NYC TLC data shows total FHV daily trips averaged 831,862 in July 2025, up 8.5% year-over-year. Uber holds roughly 70% of rideshare; Lyft is nearing 30%. Uber’s share of total FHV volume implies about 500,000 to 582,000 rides per day. The estimate of 550,000 sits within that band.

What the plate cap means for product strategy

Most estimation questions end at the number. The more interesting signal in this question is what the number reveals about Uber’s strategic situation in NYC. Volume is supply-constrained, not demand-constrained. You cannot grow meaningful UberX trip counts without new plate capacity, which requires city regulatory action. So the viable growth path in 2026 is revenue per trip: higher-tier products, subscription membership, airport premium pricing. A candidate who surfaces this is showing they understand the market’s actual economics, not just the arithmetic.

Typical interview range

Interviewers expect a final estimate in the 400,000 to 700,000 range with a stated central estimate around 500,000 to 600,000 and a clear methodology. Anything below 300,000 or above 900,000 without specific justification is a flag. The TLC Factbook is publicly available; naming it as a calibration anchor, rather than guessing from population alone, separates a strong answer from a competent one.