Quick Commerce Replenishment: The DC Cadence Brands Miss

The layer brands never look at

Ask a brand about quick commerce availability and they will pull up a dark store report. Percentage in stock, pincode by pincode. Useful, but downstream. On Blinkit, Zepto and Instamart the stock that reaches a dark store was pushed there from a regional distribution centre. You do not supply the store. You supply the DC. If the DC runs thin, no amount of store-level chasing fixes it.

This is the operator blind spot. Teams optimise the shelf they can see and starve the warehouse they cannot. Getting the DC replenishment cadence right is the single highest-leverage availability lever most brands have never touched.

How the flow actually works

All three major platforms run an inventory-led model for packaged brands. The platform raises a purchase order, buys your stock, and holds it at a city or zone DC. From there its own systems allocate units to the dark stores in that cluster based on local sell-through. A single Blinkit DC in a metro may feed 40 to 70 stores.

So availability is a two-stage problem. Stage one is DC in stock. Stage two is store allocation. You control stage one through how fast and how fully you answer POs. You influence stage two by keeping the DC deep enough that the platform’s allocation engine never has to ration.

Most fill misses are not manufacturing problems. They are appointment problems. Every DC takes inbound only in booked slots. Miss the slot, or arrive with a mismatched ASN, and the PO ages. By the time the next appointment opens, the dark stores it feeds have already emptied. Your factory did its job. Your logistics desk lost the window.

Set cadence off DC sell-through, not calendar

The common failure is reordering on a fixed monthly rhythm against last month’s numbers. Quick commerce demand does not respect the calendar. A heat wave, a long weekend, a creator post, and a metro DC drains in days.

Instead, reconcile DC-level sell-through weekly and reorder against the live rate. Ask your category contact for the DC dispatch or sell-out feed. Convert it to a run rate, then to days of cover. When cover on any DC drops below your trigger, you raise supply regardless of what the monthly plan said.

• Steady movers: hold 21 to 30 days of cover at each DC.
• Launch and festive SKUs: hold 35 to 40 to survive the spike.
• Never let a live SKU sit below 14 days at a metro DC.

These bands assume a two to four day appointment-to-inbound lead time, which is typical across Blinkit and Zepto DCs. If your lead time is longer, widen the band.

PO discipline is a discipline, not an afterthought

Treat PO handling as an owned operating rhythm with a named person. The mechanics that quietly cost you fill:

• Accept or short-close POs within the platform’s stated window, usually 24 to 48 hours. Silence is read as inability to supply and hurts your vendor score.
• Match the ASN exactly to the physical dispatch. Case count, batch, MRP, expiry. A mismatch triggers a gate rejection and you lose the slot.
• Book the appointment the moment the PO lands, not the day you dispatch. Slots at a busy metro DC fill fast.
• Send fresh stock with adequate residual shelf life. DCs reject short-dated inbound, and a rejection reads as a fill miss on your record.

None of this is glamorous. All of it decides whether the shelf holds.

Why OTIF alone will fool you

Brands celebrate a 98 percent OTIF and then wonder why availability sits at 80. The two are not the same. OTIF asks whether you fulfilled the PO you were given. It says nothing about whether the PO was big enough. If your DC cover was thin, the platform simply orders less, you fulfil that smaller order perfectly, and the dark stores still run dry.

That is why cover is the leading indicator and OTIF is the lagging one. A healthy operation watches both. Cover tells you the tank is deep. OTIF tells you the pipe is open. You need both true at once.

Build the weekly rhythm

Turn this into a fixed cadence your team runs every week without being asked. Pull DC-level sell-through and closing stock. Compute days of cover per DC per SKU. Flag anything under trigger and raise supply. Confirm every open PO is accepted and slotted. Review last week’s rejections and fix the root cause, whether it was an ASN error or a shelf-life gate.

Do this for a quarter and the pattern becomes clear. Availability stops swinging. Festive spikes stop causing citywide zeros. Your category manager starts trusting your supply, which is the currency you spend when you negotiate better placement and terms. Availability is not a shelf metric. It is a warehouse habit, set two stages upstream of the shelf everyone else is staring at.