Dark stores explained: why the network decides your reach

Most brand teams treat a quick-commerce listing as a switch. You go live, you turn on spend, you expect orders.

It does not work that way. A listing is national. Availability is not. Availability is decided store by store, and the store is a dark store.

This page sits under our wider guide to e-commerce marketing and covers one thing only: the fulfilment unit, its economics, and what those economics do to your media plan.

What a dark store actually is

A dark store looks like a shop inside and behaves like a warehouse. Shelves, aisles, chillers, labels — no customers.

Writing in California Management Review Insights on 16 January 2026, Alok Raj of XLRI Jamshedpur and Akash Tiwary describe it as a retail outlet that "looks like a supermarket or convenience store but is not open to walk-in customers", optimising "operational efficiency over customer-facing experience". They put a typical store at 3,500–6,000 sq ft carrying 17,000+ SKUs.

The clearest way to see it is against the two things people confuse it with.

That last row is the commercial one. A warehouse is rewarded for stocking everything. A dark store is rewarded for stocking the few thousand things its own neighbourhood buys this week. "Micro fulfilment centre" is the logistics term for the same idea.

One listing, three realities.

How India's platforms run dark stores right now

Treat the numbers below as this quarter's state, not a permanent fact. The category moves every quarter. Both figures come from listed-company disclosures, so they are datable.

• Swiggy Instamart reported 1,171 active dark stores across 131 cities, 4.92 million sq ft of store area, 1,089 orders per dark store per day and an average order value of ₹691 for the quarter to June 2026 — Swiggy Limited's Q1 FY2027 shareholders' letter, results approved 30 July 2026. Quick-commerce contribution margin that quarter was −0.2%.
• Blinkit reached 2,443 dark stores after adding 200 in the same quarter, with capex of roughly ₹2.5 crore per store — Business Standard, 23 July 2026.

What is structural, and will still be true next year: the store is small, the catchment is small, the range is rationed, and the whole thing only works at high order density.

The economics, worked through on hypothetical numbers

Working through the maths: the inputs below are invented for illustration. They are not any platform's real cost base. Run the same structure on numbers you can defend.

Daily contribution = orders/day x (basket x gross margin % - variable cost/order)
Store breaks even when daily contribution = daily fixed cost

Assume a hypothetical store with ₹12,00,000 of monthly fixed cost — rent, power, cold chain, store staff — which is ₹40,000 a day. Assume an 18% gross margin on the basket and ₹32 of variable cost per order for the rider, packaging and payment fees.

Read the pairs. Case A and case B have the same basket; only density changes, and density alone closes a ₹16,000 daily hole. Case A and case D have the same density; basket alone does the same job.

Two levers, same result. That is why every quick-commerce platform chases both — more orders per store, and bigger baskets per order.

It also explains the one number you should care about. A store sitting below its break-even line has no reason to give shelf space to a slow-moving SKU. A store above it can afford to experiment.

The assortment constraint: listed nationally is not available nationally

Here is the part most media plans get wrong.

Your catalogue might run to thousands of SKUs. The store holds a finite number of facings. Something has to be cut, and the cut is made locally, on rate of sale.

So the assortment constraint bites like this:

• A national listing proves you are sellable. It does not prove you are in stock anywhere specific.
• Your hero SKU may be in most stores while your variants sit in a few dozen.
• Pack size decides a lot. A ₹99 pack helps the store's order count; a ₹499 pack helps its basket. Both are useful to the store, for different reasons.
• A slow fortnight in one cluster can get you dropped there while you are still live elsewhere.

This is the same discipline as any D2C distribution plan, except the shelf resets far faster than a supermarket's does.

The pin-code reality, and what to do about it

Because each store carries its own stock list, availability varies by pin code within a single city. Two customers two kilometres apart can see different things.

National campaign planning assumes a national shelf. Quick commerce does not have one. What to do instead:

What the platform shows you, and what to ask for

Reporting depth varies by platform and by how deep the commercial relationship is, and none of it is publicly documented. So do not assume — ask for the data dictionary before you sign.

The second column is why a quick-commerce listing cannot be your only data source. Buying media on the platform is a separate craft — that sits with retail media and, for Instamart specifically, Swiggy ads.

What we check before funding a quick-commerce listing

In our own work at Digital Hangover, five things get answered before any budget moves behind a quick-commerce listing. No answer means no spend.

Over a longer horizon the same questions feed customer lifetime value work, because repeat buying on these platforms follows availability more than affinity.

Regulation now reaches the store, not just the app

Two live developments, both datable.

• Food safety enforcement has moved inside the store. Telangana's food safety authority suspended the FSSAI licences of four quick-commerce dark stores in Hyderabad after inspecting 16 fulfilment hubs, citing expired stock, pest infestation and faulty cold-room controls — Adgully, 2 October 2026. Separately, FSSAI is considering e-commerce-specific rules covering manufacturing date, expiry date and batch number on listings, and stricter checks on dark-store storage — Inc42, 13 October 2026. These are proposals; no effective date has been published.
• Gig and platform work is now recognised in law. India's four labour codes took effect on 21 November 2025. Under the Code on Social Security, aggregators contribute 1–2% of annual turnover to a social security fund, capped at 5% of payments made to gig and platform workers — Press Information Bureau, 9 December 2025.

Neither is a marketing problem on its face. Both land in the same place: cost per order goes up, the break-even line in that table above moves right, and marginal SKUs get cut sooner.

Plan for the network and your spend works harder

A dark store is not a detail of someone else's supply chain. It is the gate between your advertising and a completed order.

Brands that plan against the store list instead of the city list stop paying for unservable impressions, stop blaming creative for a stock problem, and get a clean read on what their media actually did. That is a pure efficiency gain — no extra budget required.

If you are building the paid side of this, our performance marketing guide covers the measurement discipline it needs.