Dark store tiering: build q-commerce distribution store by store

Being live on Blinkit is not distribution

A brand goes live on Blinkit. The team tells the founder the brand is now national. The monthly number lands at Rs 4 lakh and stays there. The brand is stocked in 38 dark stores out of roughly 2,000. That is under 2 percent weighted distribution. Nobody blames the store count. Everyone blames the platform.

City selection sits above this decision and most brands handle it reasonably well. Store selection sits below it and almost nobody does it at all. Quick commerce is the only channel in India where distribution is readable store by store, weekly, at no cost to you. A general trade equivalent needs a retail audit and three months. Here it needs a spreadsheet and a Monday.

Build the store list before the first purchase order

Ask your category manager for the store master for the cities you are entering: store code, locality, pincode, cluster. You will usually get 60 to 400 rows per city. Do not accept a launch list picked for you by someone with 40 other brands to place. Rank the stores yourself against three inputs you already own or can get in a week.

• Your own D2C order density by pincode over the last 12 months. This is the cheapest demand signal available to an Indian brand, and the platform does not have it.
• Presence of two or three comparable brands in the same subcategory in that store. If the habit exists there, shoppers are already typing your keyword.
• Cluster value tier. Category teams will describe clusters as premium, mid, or value even when they will not hand over numbers.

For a first launch, 60 to 90 stores across two cities beats 200 stores across six. Velocity per store is what earns a second facing, a second SKU, and eventually a cluster expansion. A thin national spread produces the same monthly revenue and reads as a slow mover in every store level report the category team opens.

Tier the stores and give each tier a job

Split the launch list into three tiers before inventory moves, and write down what each tier is for.

• Tier A, about 20 percent of stores. Full range, ad support, weekly availability checks, any sampling budget. These stores build the case for expansion.
• Tier B, about 50 percent. Hero SKU plus one variant, no ad spend for the first six weeks. This is your read on organic pull.
• Tier C, the remainder. Hero SKU only, minimum order quantity, and an exit date already in the plan.

Then set a cut line, expressed as units per SKU per store per day. For an impulse snack at Rs 99, that number is usually 3 to 5. For a Rs 449 personal care SKU, 1 a day can hold a facing. For a Rs 1,200 accessory, 2 a week. Set the number before launch, not after the first bad month. Brands that arrive with their own threshold get taken seriously by category managers who spend all day being asked for favours.

The 45 day store level review

Six weeks in you have enough data to act. Pull three columns per store: units sold, days in stock, and availability percentage. Then compute the only distribution metric that matters at this stage, which is live SKU-store combinations holding above 85 percent availability. Four SKUs in 60 stores is 240 combinations on paper. If 90 of those were out of stock for half the month, your real distribution is closer to 150 combinations, and your velocity per combination is being reported against the wrong denominator.

Sort every store into one of four buckets.

• Selling and in stock. Give these more range and the ad budget.
• Selling and out of stock. The most expensive bucket in quick commerce, and usually a drop quantity or forecast problem rather than a demand problem.
• In stock and not selling. Needs a pack size, price point, or content fix before it needs an exit.
• Neither. Exit without argument.

Exit the bottom bucket deliberately. Ask for those stores to be delisted and the inventory reallocated into Tier A. Brands resist this because delisting feels like retreat. It is housekeeping. A dead store drags your sales per store per day, and that average is what the category team quotes internally when your expansion request comes up.

What to ask for once the data is on your side

Take the store level view into the monthly review and ask for three specific things rather than general support. First, expansion into the next 60 stores inside the same clusters where you are already above cut line, not a scatter across new cities. Second, a second SKU in Tier B stores where the hero has cleared the cut line for four straight weeks. Third, a corrected drop quantity for the stores that keep selling out mid-week.

Each of those asks is easy for a category manager to approve because you have done the work that justifies it. Vague asks get vague answers and a promise to revisit next quarter.

Sequencing beats ambition here, and the arithmetic is not subtle. A brand in 300 well chosen stores at 88 percent availability will outsell a brand in 900 stores at 60 percent availability, on less working capital, with less near-expiry risk and a cleaner claims file. The platform reads the first brand as a partner worth backing. It reads the second as a risk to be managed.