How many restaurants are in Mumbai?

Q: How many restaurants are in Mumbai?

A worked estimation for the PM interview question on Mumbai restaurants, covering organised vs. unorganised sectors, cloud kitchens, and the 2026 product angle.

The single most important move in this question is the clarifying question you ask before doing any arithmetic. The answer range spans an order of magnitude depending on what “restaurant” means: from ~15,000 registered, platform-discoverable establishments to 140,000+ food-serving businesses across all formats.

The clarifying question

Ask: “Are we counting all food establishments including street-food formats and cloud kitchens, or registered dine-in and delivery operations only?” Name the two interpretations and commit to one. This tells the interviewer you understand the structural reality of Mumbai’s food economy before you touch the numbers.

Pick a lane: organised dine-in, QSR chains, cafes, and cloud kitchens. Exclude pure street vendors. That is the range a platform like Zomato or Swiggy would care about.

Demand-side build

Mumbai’s working population: 20 million residents (Greater Mumbai, BMC jurisdiction).

Segment by income and dining-out frequency:

Upper-income (15% of population): eat out roughly 5 times per week
Middle-income (40%): eat out roughly 2 times per week
Lower-income (45%): eat out roughly 0.5 times per week

Weighted daily meals eaten outside the home per person: (0.15 × 5 + 0.40 × 2 + 0.45 × 0.5) / 7 = approximately 0.18 meals per person per day.

Total daily covers across the city: 20 million × 0.18 = 3.6 million meals per day.

Average covers per establishment per day across all formats (dine-in, QSR, cafe, and delivery-only): roughly 28. That gives 3.6M ÷ 28 = approximately 128,000 establishments. Pull back 10-15% for cloud kitchens and delivery-only operations that have no physical covers. Adjusted estimate: 110,000-120,000 food establishments.

Sanity checks

NRAI India Food Services Report 2024: India has 500,000+ restaurants nationally. Mumbai has roughly 8-9% of urban India’s population but disproportionately high income concentration and food culture. 120,000-140,000 is consistent.
Industry listing data (early 2026): 140,000+ food establishments across all formats in Greater Mumbai.
Platform reference class: Zomato lists approximately 60,000 restaurants in the Mumbai metro area. That number represents the organised, discoverable layer, which makes sense as roughly half the broader estimate once you exclude informal operators not on any platform.
Registered floor: approximately 15,000 restaurants appear in GST-registered, platform-listed datasets. That is the bottom of the organised sector, not the total.
Density check: EazyDiner reports 7,100+ restaurants in Western Suburbs alone, consistent with high concentration in upper-income belts.

Final answer: approximately 120,000-140,000 food establishments across Greater Mumbai in all formats. If the question is about what a delivery platform could partner with today, the relevant number is ~60,000. If it is about GST-registered establishments only, the floor is ~15,000.

The 2026 follow-through

A strong candidate does not stop at the number. The organised sector is roughly 20-25% of total food establishments in Mumbai (higher than the national average of 10-15%, due to GST formalisation pressure and platform aggregation). That means approximately 89% of restaurants are single-owner operations with no digital infrastructure.

Volunteer the product observation: “The 100,000+ unorganised establishments are the interesting product design problem, not the 15,000 already on platforms. These operators have thin margins, informal labour, and no appetite for SaaS subscription fees. A viable product for them has to be embedded in what they already use: WhatsApp ordering, UPI receipts, FSSAI compliance nudges. Not a separate app.”

That pivot from estimation to product viability is what interviewers at Swiggy, Zomato, and Zepto are actually screening for in 2026.

strong

"Before I size this, one question: are we counting all food-serving establishments including cloud kitchens and informal eateries, or just registered dine-in and delivery operators? [Interviewer confirms all formats.] OK. Mumbai has 20 million residents. Segmenting by income, I get a weighted daily eating-out rate of about 0.18 meals per person, or 3.6 million meals per day citywide. At roughly 28 covers per establishment per day across formats, that is about 128,000 establishments. Subtract delivery-only operations with no covers and you land at 110,000-120,000. Sanity check: industry data puts the figure at 140,000+, Zomato lists ~60,000 in the metro, and NRAI's national count of 500K is consistent with Mumbai holding 20-25% of the organised sector. My answer is 120,000-140,000 total, ~60,000 platform-accessible. The more interesting number for a product question is the 100,000 unorganised operators who are not on any platform and whose viability problem is unsolved."

weak

"Mumbai has 20 million people. About 5% eat out daily, so 1 million meals. Each restaurant serves 100 people. That gives 10,000 restaurants." This is off by more than 10x. It ignores dining frequency segmentation, uses a Western-calibrated 100-seat assumption for a market dominated by small-format operators, never asks what counts as a restaurant, and cannot respond when the interviewer points out that Zomato alone lists 60,000.

What interviewers are actually testing

Indian consumer tech interviewers (Swiggy, Zomato, Meesho, PhonePe) use this question to see whether you understand the dual-layer market: the organised sector they already serve and the unorganised majority they have not yet cracked. Global tech companies asking this question to candidates with Indian market experience want to see emerging-market nuance, specifically that you do not treat Mumbai like San Francisco with more people.

The cloud kitchen layer matters. Mumbai was India’s fastest-growing cloud kitchen market post-2020. Rebel Foods, Swiggy, and Zomato each operate thousands of virtual brands out of shared kitchen infrastructure in Mumbai. A candidate who does not account for cloud kitchens is using a 2018 mental model of restaurant supply.

The estimation is the setup. The product conversation about who has a viable business is the bar.