We tested our own finding in a second city. It broke.
Two days ago we published that Zepto has no uncontested shelf space — every store we read had an advertiser for every search. That was true in Bengaluru. We have now run the identical test in Delhi NCR, and in Delhi it is false. Here is what replicated, what didn't, and what we got wrong.
We wrote that “on Zepto, every store we read had a sponsored placement for every search — there is no uncontested shelf anywhere.” That claim was measured in Bengaluru and published without a city qualifier. In Delhi NCR, 18 of 81 Zepto store-searches returned no advertiser at all. The finding was real; the generalisation was not.
What replicated
Blinkit was almost eerily consistent. Exactly 10 of 81 store-searches returned no advertiser in Bengaluru, and exactly 10 of 81 in Delhi NCR. Different city, different stores, different assortment — same proportion of uncontested shelf. That is a much stronger result than either city alone, and it suggests Blinkit's ad density is a platform-level property rather than a local accident.
Instamart replicated too, in the other direction. Zero uncontested store-searches in Bengaluru, zero in Delhi. Every store, every search, both cities: someone was paying. If you sell on Instamart, there is no quiet corner in either metro.
And one advertiser got even more dominant. Ritebite Max Protein held 240 sponsored placements for “protein bar” on Instamart in Bengaluru — 66% of the category. In Delhi it held 266 placements, 93% of everything paid in that category. That is close to total ownership of a category's paid shelf across two metros.
What didn't
Zepto behaved like a different company between the two cities.
- Uncontested shelf went from 0 to 18 of 81 store-searches.
- Empty shelves appeared. In Bengaluru, no Zepto search returned an empty store, in any category. In Delhi, every one of the nine searches had at least one store return nothing at all.
- Ad concentration loosened slightly — the top advertiser's share fell from a 21–70% range to 21–38%.
The most likely explanation is network maturity: Zepto's Bengaluru operation appears denser and more uniformly stocked than its Delhi NCR one. We can measure the difference; we cannot see inside the company, so we are not going to speculate further than that.
Why we are publishing this rather than quietly fixing it
The entire argument of this site is that quick commerce has no single national shelf — that averages hide what happens store by store. It would be incoherent to make that argument and then treat a one-city read as a national fact about a platform. We did exactly that, and a second city caught it within 48 hours.
So the standard changes from here: a platform-level claim requires at least two cities, and every claim carries the city it was measured in. The affected pages have been updated with that qualifier.
There is a wider point for anyone reading quick-commerce coverage. Almost every published figure about this market — including ours until today — comes from a single sample presented as a general truth. When you next see a claim about how a platform behaves, the useful question is: measured where, on what date, across how many stores?
Nine searches across coffee, protein bars and baby care, read in 9 dark stores per platform per city in Bengaluru (22 July 2026) and Delhi NCR (24 July 2026) — 486 store reads in total. Public shopper-facing shelf only; no seller accounts, no brand-supplied data. A store-search counts as uncontested when the shelf returned no sponsored placement at all. Ads are counted as slots owned, never rupees spent, because bids are not public. Two cities on two days is still a small sample: it is enough to disprove a general claim, not enough to prove one.
Your brand's numbers are also city-specific
If a platform behaves this differently between two metros, so does your shelf. Read your brand store by store, in the cities you actually sell in. Free, no signup.
Keep reading
- Blinkit vs Zepto: two completely different shelves — the original Bengaluru finding, now city-qualified.
- Who's actually buying the Blinkit shelf
- Delhi NCR coffee report · protein bars · baby care
- Our method, and its limits