916.6 million orders across 4,525 stores in FY26 set the base for India's next expansion wave, and the number that will decide whether it holds is not order volume but item-level availability as the network moves into smaller towns. Through November 15, 2026, the public record shows Blinkit at 916.60 million completed orders, Zepto at 640.18 million and Swiggy Instamart at 412.20 million, with 2,243, 1,139 and 1,143 stores respectively, while festive quick commerce GMV rose about 60 percent year on year to 22,528 crore rupees, roughly 16 percent of online festive sales.
Density is the variable, not city count
A Redseer-Google report released on October 7, 2026 put India's quick commerce market at roughly 13 billion dollars in FY2025-26 and projected 70 to 90 billion dollars by FY2030-31, with metropolitan markets contributing about 60 percent of incremental GMV and non-metro markets about 40 percent. That split is the part most brand teams skip, because it looks like a growth statistic when it is really a warning about where availability gets harder to hold.
The arithmetic is straightforward. A store handling a few hundred orders a day cannot economically carry the depth of a store handling several thousand, so its assortment gets cut, and the cut items are exactly where availability breaks first. Expansion therefore converts a coverage problem into an availability problem, and the two are measured with different instruments.
What tier town economics do to assortment
Depth falls before breadth does
When a store's demand base thins out, the first thing to go is depth on slow movers, not the count of listings. A brand can therefore still be listed in a town while being unpickable for most of the week, and nothing in the listing report will show it. The only early signal is fill rate measured per store and per item, refreshed often enough to catch a two-week slide.
Frequency decides whether the store matures
Store economics rest on order frequency, not on order count. High frequency spreads fixed costs across more baskets and justifies deeper inventory; low frequency does the opposite, and each new store added to a thin market dilutes the frequency that made the earlier stores work. That is why store count is a lagging indicator of a network's health and fill rate is a leading one.
| Metric | What it shows | Why it matters |
|---|---|---|
| Store count | Coverage breadth | Lagging: shows where the network was built |
| Item fill rate per store | Availability inside the node | Leading: shows whether stock is pickable |
| Order frequency per store | Demand density | Decides whether store economics hold |
Read together, the three rows explain a familiar failure. A brand reports distribution in 60 cities, the network reports a 60 percent GMV increase, and both look healthy while individual stores run out of the two or three items that carry the basket. Nobody is wrong, but nobody is measuring the thing that determines repeat purchase.
The availability benchmark a brand should run
Sample the top sellers, not the catalogue
Auditing every listing is unnecessary. Sampling the twenty to thirty items that generate most of the basket, and checking them store by store for four consecutive weeks, will surface more than a full catalogue sweep run once a quarter. The sample has to move with the footprint, because a store added last month has no history to compare against.
Separate listing from pickability
Two numbers, kept apart, tell the whole story: the share of sampled stores where the item is listed, and the share where it was actually picked in the last seven days. The gap between them is the availability loss, and it is usually larger in newer stores than in mature ones, which is precisely what a tier town expansion produces.
Data quality sits underneath both numbers. Fragmented records across marketplaces, warehouse systems and store-level tools produce overselling and cancellations that look like demand volatility, and industry estimates cited by DigitBridge put the cost of stockouts, overselling and operational errors linked to fragmented data at 5 to 15 percent of potential annual revenue for multichannel retailers. The estimate's methodology is not disclosed, so treat it as a range to test rather than a figure to quote.
FAQ
Why is store count a lagging indicator?
It shows where the network was built, not whether those stores are stocked. Item-level fill rate inside each store is the leading indicator of real availability.
What decides whether the current expansion holds?
Order frequency and basket value. High frequency spreads fixed costs and justifies deeper inventory; thin demand in smaller towns does the opposite and dilutes the stores that already work.
How many items should a brand audit?
Twenty to thirty that carry most of the basket, checked weekly for four weeks. A smaller, refreshed sample beats a one-off catalogue sweep that goes stale immediately.
Where does data quality fit in?
Record errors create overselling and cancellations that read as demand volatility. Reconciling store stock with the listing feed separates a broken record from a genuine stockout.
