JD.com's 34 JD MALL locations, more than 200 Appliance City flagship stores and nearly 18,000 home appliance stores recorded over 13.4 million visits during this year's Singles' Day campaign, the company said in a disclosure reviewed as of Nov 12, 2026. The same release put JD NOW, its instant retail service, up 42 percent year on year. Those two numbers describe a store network being asked to do two jobs at once: demonstrate products to people who walk in, and fulfil orders for people who never will.
Two Jobs, One Store Network
JD's own campaign disclosure frames the stores as showrooms and professional service hubs rather than points of sale. That framing matters for anyone who manages distribution: a showroom is judged on how many products a visitor can actually touch, while a fulfilment node is judged on whether the item on the shelf can be picked, packed and handed to a rider within the promised window.
The instant retail number is the harder one to copy. A 42 percent increase in transaction value does not come from adding riders; it comes from knowing, at any given minute, which store within three kilometres holds which SKU and can commit to it. Retailers that cannot see store-level stock in real time end up promising availability they cannot deliver, and the resulting cancellations cost more than the missing sale.
What Counts as Availability
On-shelf availability used to be a monthly audit question: did the store receive the SKU, and is it on the shelf. Under a thirty-minute promise the question splits into three. Is the unit in stock, can it be picked from the sales floor rather than a locked back room, and is the store open for orders in this time slot. Any one of the three failing shows up to the customer as a stockout, even when the warehouse report says otherwise.
Trade-Area Math Changes When Delivery Is Thirty Minutes
Trade area analysis traditionally weighs population, competitor density and drive time. Instant retail adds a fourth variable that most models still ignore: the density of pickable inventory inside the delivery ring. Two stores can sit in identical trade areas and perform completely differently because one holds the high-turn categories and the other holds display models only.
| Store type | Demo capability | Availability metric |
|---|---|---|
| Flagship in core city | Full range, powered on | Demo-to-order conversion |
| County-level partner store | Core categories, video for the rest | Same-day order volume |
| Community experience store | Scenario display, few units | Repeat visit rate |
This table is a selection tool, not a description. Once demo capability and availability are scored separately, store selection stops being a single ranking and becomes a portfolio question: how many stores in a trade area need full demo capability, and how many only need to hold stock for delivery. Brands that answer it with one number tend to over-invest in showrooms and under-invest in pickable depth.
The Showroom Problem
Showrooms have a cost structure that instant retail does not fix. Floor space in a mall is expensive, and every square metre devoted to demonstration is a square metre not holding sellable stock. The stores that handle both jobs well tend to separate them physically: a demonstration zone near the entrance, a picking zone behind it, and a rule that demonstration units are never counted as available inventory.
What Brands Should Change This Quarter
The first change is reporting. Most brand dashboards still merge online and offline sales into one line, which hides whether growth came from a store visit or from a delivery order. Splitting the two reveals a different picture per city and per category, and it is the only way to tell whether a store is a showroom, a warehouse or both.
The second change is the definition of a high-potential store. Population and competitor counts remain useful, but they no longer separate winners from losers. What separates them is whether the store can commit to a delivery window and whether a visitor can test the product. Both are operational facts that can be scored before signing a lease.
The third change is pricing discipline across channels. When the same SKU is sold in a mall and delivered in thirty minutes, a price gap between the two becomes visible to consumers within a single search. China Daily's review of the festival noted that platforms pushed simpler discount rules this year, which makes cross-channel price gaps easier to spot rather than harder.
Action List
The sequence below is deliberately operational. Each step produces a field that the next step needs, so skipping one leaves the later steps without an input. Each step below produces a field that the next one consumes, and none of them can be skipped without leaving a gap in the reporting that follows it.
- Split the sales report into store visit sales and delivery order sales, by city and by category.
- Add three availability columns to the store master file: in stock, pickable, open for orders in this slot.
- Score every store on demo readiness and picking depth, then rank the portfolio on both axes.
- Exclude demonstration units from available inventory in every system that feeds a delivery promise.
- Review the price gap between in-store and delivered orders weekly, by SKU, before consumers report it.
FAQ
Does 13.4 million visits mean store traffic is recovering?
Not on its own. The figure covers 34 malls and roughly 18,000 stores over a promotion period, and the company did not break out unique visitors.
Why treat instant retail and store visits as one system?
Because both draw on the same inventory. A unit promised to a delivery order is not available for a walk-in customer, and vice versa.
Is on-shelf availability still a useful metric?
Yes, but it needs splitting into in stock, pickable and open for orders. The combined figure hides which of the three failed.
How should a brand pick high-potential stores now?
Score demo readiness and picking depth separately, then decide how many stores in a trade area need each capability.
