70% growth and 500,000 merchants are both big numbers, but neither of them tells you whether a store can actually fulfil an order. As of November 11, 2026, Taobao Shangou reported more than 500,000 retail brand merchants taking part in Singles Day, 935 brands doubling transaction value over a single weekend, and fresh food, baby care, home appliances and household goods growing more than 70% year on year. For teams running on-shelf availability across a store network, the festival is less a marketing event than a stress test of distribution coverage, and the failure mode is almost always the same: the stock exists, but not where the order lands.
Why the festival breaks availability, not demand
Platform reporting describes reach, not readiness. A store being connected to an instant retail channel means it can receive orders; it does not mean it can fill them. The gap between those two states is where most of the operational damage happens during a peak, and it appears in three places at once: inventory distribution, pick routing, and the penalty structure that follows a stockout.
It is worth being precise about the scale involved. Singles Day sales across major platforms reached about 1.02 trillion yuan in the three weeks to the end of October, up nearly 2.4% on the same stretch a year earlier, according to the research firm Syntun. Growth of that size does not arrive evenly: it concentrates in a few hours, in a few categories, and in the stores closest to where demand spikes.
That concentration is why availability, not assortment, becomes the binding constraint. Most brands have already widened their catalogue for the festival; what they have not necessarily done is make sure every store in the network can sell the widening. The retailers that spend heavily on instant retail, including Alibaba, JD.com and Meituan, have collectively pledged tens of billions of yuan to subsidise one-hour delivery, which raises the cost of a failed order for everyone in the chain.
Three breaks that appear at the same hour
Break one: the stock is in the wrong store
A neighbourhood store that ships a few dozen orders on a normal day may receive several hundred during a festival. Total inventory can look healthy while a single fast-moving item is sold out in one store and sitting three kilometres away in another. Platforms route the order to the store closest to the customer, not to the store holding the stock, so the shortage is created by geometry rather than by supply.
Break two: the aisle was designed for browsing
A store is not a warehouse. Aisles are laid out for shoppers, promotional displays block the main path, and pickers walk around them. When order volume doubles, average pick time typically climbs from around three minutes to seven or eight, and late orders move from occasional to systemic. The customer does not see a busy store; the customer sees a late delivery, and the review lands on fulfilment rather than on the product.
Break three: the record outlives the order
A single stockout costs one order. A pattern of stockouts costs visibility, because platforms down-rank stores that repeatedly fail to fulfil. Availability therefore stops being an internal operations metric and becomes a distribution mechanism: improving it is a way to buy traffic more cheaply than bidding for it.
What availability looks like by store type
The failure modes differ by format, which is why a single threshold rarely works. A grocery store loses the order when fresh food runs out at 7pm; an appliance specialist loses it when the installation slot is three days away. Treating both as one availability number hides the operationally different problems underneath.
| Store type | Typical SKU range | Peak-day failure | First action |
|---|---|---|---|
| Neighbourhood grocery | 800-1,500 | Fresh and prepared food sells out | Time-slot replenishment plus substitutes |
| Baby care | 600-1,200 | Formula stage gaps, slow transfers | Stage mapping and transfer thresholds |
| Appliance specialist | 300-800 | Installation slots block same-day orders | Appointment booking and staggered dispatch |
| Convenience | 1,500-3,000 | High-frequency items bought by the case | Purchase limits and front-of-store picking |
A weekly routine that holds under load
The work has to start before the peak, and it has to use one definition across every region. Otherwise the review meeting turns into an argument about whose numbers are right, which is the most expensive way to lose a week.
None of this is sophisticated. What makes it work is consistency: with tens of thousands of stores, a single definition lets a team see which failure type is most expensive, while a dozen local definitions only produce anecdotes. For brands negotiating for traffic allocation, being able to explain your availability rate is the entry ticket.
It is also worth noting what the platform numbers do not cover. Singles Day totals are reported at platform level, and the operating detail a store team needs, such as pick time by hour or stockout by SKU, is not public. That gap is why the routine matters more than the dashboard: the dashboard tells you the peak arrived, the routine decides whether you can still fill the order at nine in the evening.
FAQ
Is availability the same as in-stock rate?
No. In-stock rate measures inventory, while availability measures whether a customer can order right now. An item can be in stock and still unavailable because it is not listed, the store is closed, or the delivery radius excludes the address, which is why store teams should track both.
Why do stockouts cost more than the lost order?
Because platforms down-rank stores that fail repeatedly. A pattern of stockouts reduces future traffic allocation, so the cost compounds beyond the individual basket and shows up in the next campaign's order volume.
How should a team detect late orders early?
Track average pick time and the share of late orders hourly rather than watching order volume. Volume rising with stable pick time is healthy; pick time rising with flat volume usually means routing or staffing, not demand.
Can substitute mapping replace replenishment?
It covers fresh and fast-moving categories where replenishment cannot arrive in time. For standard goods, transfer from a nearby store is usually better, provided the transfer cost is priced into the margin.
