Operations · Product case study · India
zepto: Build the shop nobody walks into.
Zepto is an Indian quick commerce company that promises groceries in about 10 minutes. This case study explains what Zepto is, how its network of dark stores makes the promise possible, what its 2026 draft prospectus says about speed, scale and losses, and what a product manager can take from it.
See inside a dark store ↓Try it · the promise against the stopwatch
Medians and distances from Zepto’s June 2026 draft prospectus, as summarised by ABI Analytics.
- 1,139
- dark stores, March 2026
- 66
- cities
- 17.5 lakh
- orders a day, FY26
- 4.8 crore
- people ordered in FY26
Independent educational analysis by PMcademy. Figures are from Zepto’s draft prospectus as summarised by ABI Analytics, or from the press sources listed at the end. The models and the practice scenario are fictional teaching material. No affiliation with Zepto.
What Zepto is
Zepto sells groceries, snacks and household goods, along with cafe food and, since 2025, medicines, and brings them to your door in about ten minutes. You order in the app. The order is packed in a small warehouse near you and handed to a rider. There is no shop to visit and no delivery slot to choose.
It was started by Aadit Palicha and Kaivalya Vohra, who left Stanford and were both 19 when Zepto launched. Their first attempt, KiranaKart, delivered from neighbourhood kirana shops. Palicha has said customers of that early service complained of waits of two to three hours, and that it did not keep its users. The lesson they took was that speed was not a gimmick. It was what customers kept asking for.
- KiranaKart: grocery delivery from local kirana shops. It does not catch on.
- Quick commerce begins, from the company’s own dark stores.
- Zepto comes out of stealth, live in Bengaluru, Mumbai and Delhi.
- Zepto Cafe starts in Mumbai, inside the same dark stores.
- Valued at $1.4 billion.
- Headquarters move from Mumbai to Bengaluru.
- Raises $450 million at a $7 billion valuation.
- Files an updated draft prospectus, seeking up to ₹8,010 crore of new capital.
The unique feature: the dark store
- Goods in at the back. Stock arrives from one of 75 warehouses and goes straight onto shelves.
- Fast movers by the desks. The things most orders contain sit a few steps from packing.
- Packing in about a minute. In 2021 the founders said an order was packed and out in under 60 seconds.
- Riders at the door. The average order travelled 1.78 km in FY26.
A dark store is a small warehouse laid out for picking, not for browsing. Zepto’s planned new stores average 3,500 to 4,000 square feet, and its filing counts an average of about 46,600 products available across a store’s area in FY26. Customers never walk in, so every decision about the layout is about seconds saved.
This is the difference from a marketplace that sends a rider to someone else’s shop. Zepto runs the building, the stock and the staff, about 48,000 in-house operating staff such as pickers and packers, while delivery is done by gig riders. That is more to pay for and more to control.
Why it is about density, not rider speed
The ride is the part you cannot rush safely. So the only honest way to shorten it is to start closer. Move the slider to add stores to a fictional city and watch all three numbers change together.
- Average ride
- 1.8 km
- Order to door
- 11.1 min
- Orders per store a day
- 3,750
Too few stores: the ride is too long for the promise, however fast the packing.
Illustrative model: a city of 100 sq km placing 30,000 orders a day, 1 minute to pack, 3 minutes to hand over, riders at 15 km/h. For scale, Zepto reported 1,677 orders per store a day in FY26.
In April 2022 the industrialist Anand Mahindra called ten-minute delivery inhuman for delivery workers. Aadit Palicha replied that the model needed rider speeds of under 15 km/h. Both points are about the same thing: if the store is close enough, nobody has to hurry.
The numbers, three years side by side
Revenue doubled in FY26 to ₹22,624 crore, from ₹11,110 crore in FY25 and ₹4,455 crore in FY24. Orders nearly doubled too, to 64 crore in the year. By the January to March 2026 quarter Zepto was handling 23.3 lakh orders a day, and each store averaged 2,140.
The last chart is the one to study. The loss on each order grew when the network tripled in FY25, then fell as those new stores filled up. It was ₹59 in the final quarter of FY26. Busier stores spread rent and staff over more orders, which is the whole economic argument for density.
Where the money comes from
- ₹17,588 crProduct sales78%
- ₹2,780 crWarehousing, packaging and delivery fees12%
- ₹1,636 crAdvertising7%
- ₹564 crPlatform fees2%
Most of the revenue is the groceries themselves. The interesting line is advertising. Brands pay to appear when you search or browse, and that income grew from ₹49 crore in FY24 to ₹651 crore in FY25 and ₹1,636 crore in FY26. Palicha has credited an advertising system built in-house.
Advertising costs little to deliver once the customers are already there. For a business that loses money on each order, a high-margin line riding on the same traffic matters a great deal. It is the same pattern as CRED’s bill payments funding its lending.
The promise and the stopwatch
Zepto’s own filing reports a median time to delivery above ten minutes in every period it lists: 11.47 in FY24, 10.57 in FY25, 11.40 in FY26 and 12.35 in the latest quarter. A median means half of all orders took longer than that.
This is not a scandal. It is what a promise looks like once it is measured. The useful question for a product manager is what the number in the name is for. Ten minutes tells customers what kind of service this is. The median tells the company how close it is. Keep both, and never confuse them.
What it costs
The loss in FY26
Up from ₹4,700 crore in FY25 and ₹1,215 crore in FY24. The filing shows cash going out of operations every year since the company began.
Gig riders
The average number of active delivery partners in the January to March 2026 quarter. The promise depends on people the company does not employ.
Stores that did not work
In FY26 Zepto opened 169 dark stores, closed 59 and moved 46. Picking the wrong street is an expensive mistake.
There are two ways to ten minutes, and it helps to see them side by side. Swiggy Bolt narrowed the offer: nearby restaurants and quick dishes only, using kitchens that already existed. Zepto built the kitchen cupboard itself. The first is cheaper and limited. The second is costly and under your control.
Zepto is not alone on this road. Its filing counts 2,243 stores at Blinkit, part of Eternal, and 1,143 at Swiggy Instamart in March 2026. A Redseer study that Zepto commissioned puts its share of orders among large quick commerce apps at about 35% in the latest quarter, up from about 26% in FY24.
What to take from Zepto
A promise is a system, not a slogan
Ten minutes is the sum of packing, handover and a ride. Break any promise about speed into its parts before you make it.
Measure the promise you make
Zepto’s own filing puts the median above ten minutes. Publish the number your promise is judged by, and watch its counter-metric.
A metric that cannot be gamed →Instrument for the questions you will be asked →
Know the cost of one order
Growth made the loss per order smaller, not zero. Work out the economics of a single unit before you multiply it.
A second business can ride on the first
Advertising grew from ₹49 crore to ₹1,636 crore in two years, on top of traffic the grocery business already had.
Your turn: the promise you cannot keep
Your fictional pharmacy app promises medicines in 30 minutes. This month the median was 38. You can only do one thing this quarter. Pick one, then open it to see the trade-off.
Change the promise to 40 minutes
Honest and cheap, and you start meeting your own number. You give up the headline, so check whether customers chose you for the speed or for the reliability.
Open more stores to shorten the ride
The promise holds, as Zepto’s model shows. Every store adds rent and staff, so each one needs enough orders to carry its costs.
Limit the fast promise to common medicines
Close to how Swiggy Bolt reached ten minutes: narrow what is covered. Fewer orders qualify, so be clear in the app about which ones do.
Turn your pick into a test with the experiment brief template, and decide first whether the test is worth running.
Questions people ask about Zepto
What is Zepto?
Zepto is an Indian quick commerce company founded by Aadit Palicha and Kaivalya Vohra. It delivers groceries and everyday items from its own network of dark stores, with a promise of about 10 minutes. Its draft prospectus reported 1,139 dark stores in 66 cities at the end of March 2026.
How does Zepto deliver in 10 minutes?
By keeping stock close to customers. Orders are packed in small warehouses called dark stores and travel a short distance: 1.78 km on average in FY26, according to its draft prospectus. The same filing put the median time to delivery at 11.40 minutes for that year.
What is a dark store?
A dark store is a small warehouse laid out for picking online orders quickly. Customers never walk in. Zepto’s planned new stores average 3,500 to 4,000 square feet, according to its draft prospectus.
Is Zepto profitable?
No. Its draft prospectus reported a loss of ₹5,905 crore in FY26 on revenue of ₹22,624 crore. The loss per order, measured by adjusted EBITDA, narrowed from ₹136 in FY25 to ₹79 in FY26.
How does Zepto make money?
Mostly by selling products: ₹17,588 crore of its FY26 revenue. It also earns fees for warehousing, packaging and delivery (₹2,780 crore), advertising from brands (₹1,636 crore) and platform fees (₹564 crore), as reported by Entrackr from the filing.
Who founded Zepto?
Aadit Palicha and Kaivalya Vohra, who left Stanford and were both 19 when Zepto launched in 2021. Their earlier attempt, KiranaKart, delivered from local kirana shops.
Sources
- ABI Analytics: summary of Zepto’s draft IPO prospectus ↗Stores, orders, median delivery times, distance, revenue, losses and users from the updated draft prospectus filed in June 2026.
- Entrackr: Zepto doubles revenue in FY26 ↗The FY26 revenue lines and costs. 9 June 2026.
- YourStory: two 19-year-old Stanford dropouts ↗The founders’ 2021 account: dark stores, packing in under 60 seconds, early delivery times.
- Outlook Business: Aadit Palicha on the 10-minute bet ↗Why speed, the early waits of two to three hours, and the advertising business. 19 April 2025.
- TechCrunch: Zepto raises $200 million ↗The founders’ background and the 2022 state of the company. 2 May 2022.
- Wikipedia: Zepto (company) ↗Company timeline, funding rounds and the 2022 exchange about rider speed, used where no company page was available.
Company names and marks belong to their owners and identify the subject of this independent case study. Learn the skills behind it in the free PMcademy levels, or see the whole path on the roadmap.