Seven & i's licensee has closed all 31 of its 7-Eleven stores in India, a market where a kirana on the corner sells the same biscuits cheaper and an app delivers them in ten minutes. As of 5 November 2026, the exit reads less like one difficult market than a test of which formats survive when a low-cost neighbourhood network and a dense dark-store fleet already occupy both ends of the convenience job. The evidence sits in three numbers: 31 stores closed, about 6,563 dark stores built and 15 consecutive months of falling traffic in the format's home market.
Thirty-one stores close in one market
The closure is small in absolute terms and large in signal. Thirty-one stores is a rounding error against a global network of more than 80,000, and the licensee absorbed the cost without the parent company losing its appetite for international growth. But India was the market where the format met both of its structural competitors at once: an informal retail network with almost no fixed cost, and a delivery fleet that turns a phone into the nearest shelf. Neither competitor existed in the form the konbini model was designed to beat.
The middle position loses its customer
A convenience store sits between two promises. It is faster than a supermarket and better stocked than a street stall, and it charges for that position through higher prices and prime street-front rent. When a kirana undercuts it on price and a dark store beats it on speed, the middle position stops being a compromise and starts being a cost. The customer does not walk to the store; the store comes to the customer, or the customer never leaves home at all.
| Dimension | Kirana | Konbini | Dark store |
|---|---|---|---|
| Fixed cost | Very low, often family-owned premises | High, prime frontage plus fit-out | Moderate, back-street warehouse space |
| Speed to customer | Walk-in, minutes | Walk-in, minutes | Delivery, roughly ten minutes |
| Assortment depth | Narrow, tailored to the street | Broad, standardised | Broad, up to a lakh of SKUs at scale |
| Labour model | Owner-operated | Shift staff plus food production | Pickers plus rider fleet |
The table explains why the format struggled in a market where both alternatives were improving every year. A dark store does not need expensive street-front space, attractive fixtures or a branded in-store experience; it needs cheap back-street square footage, fast picking and a dense rider fleet. Those inputs get cheaper as the network grows, while the inputs a konbini depends on, prime frontage and shift labour, do not.
The konbini machine needs conditions Japan spent decades building
Reading the India exit as a story about India misses the more useful question. The model that Seven & i refined over five decades works because of a supply chain most markets have not built, and because the store does several jobs that a delivery app cannot. Japan runs about 21,600 stores, roughly a quarter of the global network, and those stores are not mainly places to buy snacks.
Factories and three deliveries a day
Dedicated factories produce rice balls, sandwiches and bento boxes to a schedule, and delivery trucks restock stores several times a day so that fresh food arrives within hours of being made. That system needs dense cities, food production at scale and a customer willing to pay a premium for consistent quality. Where those conditions exist, the format has travelled well: Thailand runs more than 14,000 stores under a licence, and Taiwan and South Korea treat the format as daily infrastructure.
The store as a service counter
The second leg is the counter itself. Japanese convenience stores take utility payments, collect parcels, dispense cash and print documents, which brings people in for reasons unrelated to food. A pure retail box cannot justify prime rent on snacks alone, but a box that also settles bills and holds deliveries can. Delivery apps replace the snack trip; they do not replace the service trip, which is why formats that copied the shelves but not the counter struggled to defend their rent.
Even Japan is losing store traffic
The format's home market is not immune. The Japan Franchise Association reported that same-store customer traffic at Japanese convenience stores has fallen year on year for 15 consecutive months, based on September 2026 data. Inflation is pushing up the price of everyday food, delivery apps compete for the same quick purchases, and the category growth that once came from adding fresh food has flattened, so chains are testing new categories including apparel to bring younger shoppers back. Consumer demand has not disappeared; it has moved to whichever channel answers a specific trip fastest, and that shift is the market trend store networks now have to plan against. The forces that beat the format in India are not exotic; they are the same forces, at different intensity.
For store networks elsewhere, the lesson is not that the format is finished. It is that the format's economics depend on a bundle, and the bundle can be unbundled. If the food supply chain is weak, the fresh-food premium disappears. If the service counter is absent, the rent has to be justified by merchandise alone. If a dark store can deliver in ten minutes, the walk-in trip has to offer something the app cannot, and price is rarely it.
What store networks should test before copying either model
Two tests matter more than the rest. The first is whether fresh food can be replenished more than once a day at a cost the basket can carry, because that single capability separates a konbini from a small supermarket. The second is whether the catchment generates enough service visits to justify prime rent. A network that cannot answer both should not assume that a convenience format will earn a convenience margin, whatever the store sign says.
FAQ
Why did 7-Eleven leave India entirely instead of shrinking?
The stores were operated under a licence by Reliance Retail, and the decision closed the whole 31-store estate rather than a subset. The format was squeezed between very low-cost neighbourhood shops and ten-minute delivery, which left little room for a prime-rent convenience box in the catchments it had entered.
Does the India exit mean the convenience format is failing globally?
No. The format remains dominant in Japan, Thailand, Taiwan and South Korea, where the supporting supply chain and the service counter exist. What the exit shows is that the format is a bundle of capabilities rather than a store design, and the bundle does not travel without its parts.
What does a dark store do better than a convenience store?
It removes prime rent and in-store labour from the cost base and replaces them with back-street space, pickers and riders. That makes the marginal cost of serving a top-up trip much lower in dense areas, which is exactly the trip a convenience store relies on for footfall.
Should retailers in other markets expect the same squeeze?
Where a low-cost informal network and a dense delivery fleet both exist, the squeeze is likely. Where one of the two is missing, the convenience format retains room, because it still offers speed that a supermarket cannot and assortment that a street stall does not.
