Most Dark Stores Should Never Have Been Built

By Dipankar Biswas · Head of Quick Commerce & Online Supermarket Operations, Carrefour UAE · July 2026

Unpopular opinion from someone who runs quick commerce for a living: most dark stores should never have been built.

I say this as someone who has personally launched twenty of them.

The $14 billion lesson

Look, I get it. VCs put roughly $14 billion into dark-store startups in 2021. Getir was worth $12B at its peak. You know how that ended: Getir retreated from the US, UK and Europe. Gorillas sold. Jokr, gone.

Here's the thing nobody wanted to admit at the time — the product was never the problem. Customers loved 15-minute groceries. They still do. Rent is just a terrible business partner.

A dark store opens with three liabilities on day one: a lease, a fit-out cost, and zero customers. Every order it will ever fulfill has to climb out of that hole. When the cost of capital was zero, you could pretend the hole didn't exist. When rates went up, the hole ate the industry.

Didn't India crack it?

India is the counter-argument everyone raises, and it deserves a serious answer. There are now 6,000+ dark stores across Indian quick commerce, and the scale is genuinely staggering.

But look at the unit economics: Instamart loses ~₹85 an order. Zepto ~₹79. Blinkit — the best operator of the lot — needed 1,300 stores to get its loss down to ₹3 per order.

Read that again: a three-rupee loss per order is the industry's success story. India hasn't disproven the dark-store math; it has demonstrated, at enormous scale, exactly how narrow the path is. It works only with extreme density, ruthless assortment discipline, and a war chest that can outlast your competitors' war chests.

The boring alternative that actually makes money

At Carrefour UAE, we did the unfashionable thing. No new buildings. We taught 40 of our existing supermarkets to deliver in 20 minutes.

It's not elegant. At 6pm my pickers dodge trolleys, and the online order and the lady in aisle 7 sometimes want the same last pack of chicken. Store-based picking forces you to solve co-existence problems that a dark store never has.

But consider what those stores bring to the fight. The rent is already paid — by walk-in customers. Our stores covered their fixed costs years ago. Online volume rides on infrastructure the offline business already justifies. Inventory never sleeps. Walk-in shoppers keep stock turning daily, so dead inventory — the silent killer of dark-store P&Ls — never gets a chance. (Well. Almost never.) Density on day one. A supermarket that has traded for a decade sits where the customers already are. We didn't have to guess catchments; the catchments chose the stores years ago.

The results, twelve months in: roughly 6,200 daily orders across rapid and express, ~AED 230M in annualized sales, fulfillment cost per order down 47% (AED 15 to 8), on-time-in-full up from 70% to 99% — while we cut the delivery promise from 30 minutes to 20.

Store picking at 6pm is chaos. Profitable chaos.

So are dark stores dead?

No — and this is where the nuance lives. We still run dark stores where there's no store nearby: catchments with real demand and no retail asset to piggyback on. Used that way — as infill, not as the default — the dark store is a precision tool.

What I'd want to see a business case for is the other pattern: signing leases for sheds next to your own shops. If you already own the box, the shelf, and the customer, building a second box across the street is not a strategy. It's a subsidy to your landlord.

The operator's test

1. Do you already own retail assets in the catchment? Convert them first. Capex-light beats capex-heavy when the service level is identical.
2. Is there demand density but no asset? That's a dark-store catchment — if the assortment is disciplined and the radius is tight.
3. Is the dark store there because a competitor opened one? That's not a business case. That's FOMO with a lease attached.

Speed is a product feature. Rent is a business model. The winners of the next phase of quick commerce will be the ones who never confuse the two.