As of 31 October 2026, cumulative Singles Day sales had reached just over 1 trillion yuan according to Syntun, and the figure says less about demand than about where the order is fulfilled. Which stores should carry the on-demand set now that delivery is measured in minutes rather than days? For store teams the operative change is not the total but the clock, because one-hour delivery has quietly redefined which outlets are worth stocking.
What the campaign data actually shows
The macro backdrop explains why delivery speed became the differentiator. Retail sales of consumer goods rose just 1.1% in the first nine months of 2026, while the economy grew 4.6%, and platforms responded by cutting the friction out of promotions rather than deepening discounts. Alibaba replaced cross-store coupons with straightforward 15% discounts, and shoppers who no longer need to plan a basket are also less willing to wait two days for it, as Asia Daily reported on the campaign.
Once delivery is measured in minutes, the store stops being a display surface and becomes a picking location. That single change reorders the store list, because the attributes that make a store good at display are not the attributes that make it good at picking.
| Store attribute | Weight in a display-led model | Weight in an on-demand model |
|---|---|---|
| Foot traffic | High | Medium |
| Households inside the delivery ring | Low | High |
| Aisle width for picking carts | Low | High |
| Chilled and frozen capacity | Medium | High |
| Back-room space for staged orders | Low | High |
The catchment shrinks, and the ranking inverts
A store that serves a three-kilometre ring is not competing with the store across the city, it is competing with the two stores inside the same ring. In dense districts that means the marginal store stops being marginal and the flagship stops being automatically first. Trade area analysis has to be redone at the ring level rather than the district level, and the output is a different list of high-potential stores than the one built on footfall.
Picking labour is the real constraint
Every added SKU raises the number of aisles a picker walks and the number of decisions made per order. A store carrying four thousand items for walk-in shoppers cannot simply expose all four thousand to on-demand orders without slowing pick times. The practical answer is a curated set of several hundred items chosen for local search demand and picking efficiency, not the full catalogue.
Services extend the same logic
Delivery plus installation is the clearest example of the shift. JD expanded the service to 12 Chinese cities during the campaign, with orders nearly quadrupling, and reported more than 300 million service orders across the festival period, according to Yicai Global. Services cannot be shipped from a warehouse; they are executed from a store or a local partner, which makes store capability part of the assortment decision.
What to fix before adding SKUs
Most networks discover the same thing in the same order. Availability for the twenty items that carry the category is worse than assumed, because the shelf is refilled for walk-in traffic and not for pickers. Fixing that number lifts on-demand conversion more than widening the set, and it costs nothing beyond a change in refill priority.
The festival total of 1.78 trillion yuan, up 4.8% from 1.695 trillion a year earlier, was built on a longer campaign rather than faster daily spending, as The Edge Malaysia reported. Growth of that shape rewards networks that can serve small baskets quickly, and penalises those still optimising for the single large stock-up trip.
FAQ
Why does one-hour delivery change which stores matter?
Because the catchment shrinks to the delivery radius. Households inside the ring replace foot traffic as the ranking metric.
Should every store carry the full on-demand assortment?
No. A curated set of several hundred items keeps pick times workable and avoids dead stock.
What should be fixed first?
On-shelf availability for the twenty items that drive the category. Adding SKUs before that spreads picking labour thinner.
How is a stockout different in an on-demand store?
It is visible instantly in the app, so the loss shows up as a lost basket rather than a slower week.
