Meituan is becoming one of the most aggressive players expanding in the retail market.
Since the start of this year, its Happy Monkey community hard-discount stores have accelerated multi-city expansion, its Xiaoxiang Supermarket front warehouses have continued to push into lower-tier markets, and its thousands-of-square-meter brick-and-mortar flagship stores have been rapidly replicated. Meituan also plans to spend US$717 million to acquire Dingdong Maicai's China business.
From community discount stores to front warehouses and large physical stores, Meituan is speeding up the construction of a retail system in which it directly manages the goods itself. Meituan did not originally have to do this.
As an internet platform with a massive user base, a merchant network, and an instant delivery system, Meituan could have continued connecting consumers with retailers and let the latter bear the costs of procurement, inventory, and store operations. But the 2025 food delivery war intensified competition over delivery and traffic, and also exposed another limitation of the platform model: controlling the transaction entry point does not equal controlling the supply of goods; gathering a large number of orders does not mean being able to directly obtain procurement and inventory efficiency. Meituan's exploration of self-operated retail predates the food delivery war, and now this competition has further accelerated its extension into merchandise operations. From connecting merchants to personally running warehouses and stores, Meituan is trying to convert the scale advantages accumulated as an internet platform into product development, procurement, and supply chain efficiency, in order to win a larger share of the grocery market.
But groceries have always been a low-margin business. By voluntarily taking on heavier operational responsibilities, is Meituan expanding its competitive advantage, or is it making an internet platform that was originally good at improving transaction efficiency carry more of the costs of traditional retail?
Fighting for more shopping carts
In Meituan's increasingly vast retail map, Xiaoxiang Supermarket is one of the most direct manifestations of its heavier operating model. Unlike Meituan Instashopping, which mainly connects third-party merchants, Xiaoxiang is a self-operated instant retail business that provides fresh food, daily necessities, and other goods and delivery services through community front warehouses, directly taking on procurement, inventory management, and warehouse-store operations. Since the start of this year, it has accelerated two expansion paths at the same time: online front warehouses continue to sink into lower-tier cities, while large offline physical supermarkets have shifted from single-store experiments to initial replication. At the end of 2025, Xiaoxiang Supermarket covered 39 cities, rising to 55 by the end of the first quarter this year and further to 68 by the end of the second quarter. In September, Xiaoxiang Supermarket General Manager Gao Yulong publicly said that as of the end of July it had covered more than 100 cities, with more than 1,000 online and offline stores combined. In the Yangtze River Delta, its business expanded from core cities such as Shanghai, Hangzhou, Nanjing, and Suzhou to markets including Ningbo, Wuxi, Changzhou, Jiaxing, Shaoxing, and Nantong, and also entered county-level cities such as Yiwu and Kunshan. In Guangdong, its layout also extended from Guangzhou, Shenzhen, Foshan, and Dongguan to surrounding areas, and in June this year it further entered eastern Guangdong cities such as Shantou, Chaozhou, and Jieyang. In the Beijing-Tianjin-Hebei region, it has already covered cities including Shijiazhuang, Tangshan, Langfang, Baoding, and Cangzhou. Xiaoxiang has also accelerated its entry into regions that were previously less covered. Western central cities such as Chengdu, Chongqing, and Xi'an have successively opened warehouses, and in August it entered markets such as Shenyang and Nanning, beginning to expand into the northeast and Guangxi. Xiaoxiang is trying to replicate its established front warehouse operating model into more markets at a faster pace.
Retail formats are flourishing in every direction. But from the consumer's perspective, there are only two: in-store and home delivery. Over the past few years, relying on Meituan's instant delivery network, Xiaoxiang first built up its home delivery business. Consumers place orders via mobile phone, and front warehouses near their communities complete sorting and delivery. This year, Xiaoxiang began to accelerate its push for in-store consumption. Since the opening of its first Beijing Wanliu store last December, cities including Ningbo, Hangzhou, Shenzhen, Jiangmen, and Nanchang have successively welcomed Xiaoxiang physical supermarkets. So far, Xiaoxiang has opened 8 large offline stores. Judging from the projects already opened, the stores mostly enter shopping malls in mature commercial districts, generally reaching thousands of square meters, and some directly take over operating spaces vacated by traditional supermarkets. The first Beijing Wanliu store took over the former BHG boutique supermarket, the Hangzhou Zijing Tianjie store moved into the former Yonghui Supermarket, and the Shenzhen Link CentralWalk store took over a space previously left by Carrefour. The Nanchang store that opened at the end of September has an operating area of more than 4,000 square meters, offers nearly 10,000 products, and features freshly made bakery items, freshly cut fruit, aquatic product processing, and on-site dining areas. By comparison, Hema Fresh also combines large physical stores with online instant delivery, with typical store areas of about 2,500 to 5,000 square meters and about 6,000 SKUs. Meituan CEO Wang Xing explained at the second-quarter earnings call this year that "a certain number of offline flagship stores will play a very important role in the entire ecosystem." But opening large physical supermarkets also means higher rent, labor, decoration, and equipment investment. In particular, operating areas for freshly prepared cooked food and live aquatic products also require corresponding processing staff, facilities, and shrinkage management. Meituan is willing to bear these costs because the boundaries of retail are breaking down: a single front warehouse can no longer fully capture consumers' diverse grocery spending. To seize a larger market share, Meituan has chosen to make the business heavier and heavier.
What must be added to make retail heavy
The trade-off between "scale and efficiency" runs through Meituan's retail development history. From the retreat of Xiaoxiang Fresh in 2019, to the upgrade of Meituan Maicai into Xiaoxiang Supermarket, and then to the contraction of Meituan Select in 2025 and the shift of resources toward Xiaoxiang and Kuailv, front warehouses became the main line because they can maximize the reuse of online traffic and the rider network. But the 2025 food delivery war further narrowed Meituan's competitive defense line. A Goldman Sachs report in August this year estimated that in the second quarter of 2026, the average daily food delivery order share of Meituan, Alibaba, and JD.com would be about 49%, 41%, and 10%, respectively. The delivery network still has scale advantages, but it is no longer Meituan's exclusive competitive chip. In different markets, Meituan faces competitors with completely different cost structures. In lower-tier markets, Duoduo Maicai relies on centralized delivery and community pickup to win price-sensitive consumers with lower last-mile fulfillment costs. Xiaoxiang can provide faster instant home delivery, but it must bear the costs of dispersed stocking, in-warehouse sorting, and rider delivery. Competition in first- and second-tier cities is more reflected in merchandise operating capability itself. Sam's Club builds consumer stickiness through its membership system, differentiated product selection, and higher average order value; Hema has its own product development and fresh food supply chain; community discount stores such as Aldi rely on selected SKUs and turnover efficiency to control costs. Simply delivering goods faster is no longer enough to ensure that consumers will continue to choose it.
On the one hand, Meituan needs to build a stronger retail brand mindshare. Consumers are used to ordering food delivery through Meituan, but that does not mean they will also prioritize Xiaoxiang when buying fresh food or daily necessities. Even if the platform connects a large number of community supermarkets, consumers' perception of product quality and price often still belongs to the specific retailer. Meituan also needs to control supply sources more directly. As a transaction platform, Meituan can gather a large number of retail orders, but third-party merchants each procure and stock goods on their own, making it difficult for transaction scale to automatically translate into unified procurement and inventory efficiency. To this end, multiple Meituan retail formats are deepening into upstream in different ways. In September this year, Xiaoxiang Supermarket disclosed that it had established hundreds of direct fresh produce origin procurement bases, and that some dairy products jointly developed with Junlebao had annual sales exceeding 100 million bottles, with cooperation further extending to exclusive factory scheduling and logistics optimization. Happy Monkey, meanwhile, takes origin direct procurement, selected SKUs, and private labels as the core of its hard-discount operation. It has so far developed about 300 private-label products, covering categories such as grain and oil, baked goods, and frozen foods, reducing intermediate links through factory cooperation and centralized procurement. Platform businesses are also extending upstream. In September, Meituan Instashopping launched the "Product NEXT" plan to strengthen precise product selection, product development, and supply chain support for third-party merchants. Private-label gross margin is only part of the benefit. More importantly, Meituan hopes to convert huge consumer demand into product development, procurement, and cost advantages, rather than relying only on traffic and delivery to expand transaction scale.
Synergy and separation
Meituan does not intend to let these retail formats fight alone. At the first-quarter earnings call this year, Meituan executives explicitly stated that Xiaoxiang Supermarket hopes to achieve a sustainable "low single-digit profit margin" over the long term and always pays attention to return on investment. In the second quarter, Wang Xing defined Xiaoxiang and Happy Monkey as differentiated and mutually complementary retail models, stressing that Meituan's long-term retail competitiveness comes from organizational and supply chain capabilities. Some practical synergy has already appeared within Meituan. On the platform side, although Meituan Instashopping mainly connects third-party merchants, it has been able to standardize massive product information and, by analyzing regional sales and supply data, help brands adjust distribution. This provides a technical foundation for Meituan to identify consumer demand in different regions and optimize product supply. Among self-operated businesses, the sharing of products and supply chains is more direct. Many Happy Monkey products directly reuse Xiaoxiang Supermarket's supply chain, especially in ready-to-eat and fresh food areas, and shelves can even directly sell Xiaoxiang Supermarket's private-label products. But compared with the sharing of products and supplier resources, the "hard integration" of warehousing and fulfillment networks is much more difficult, a challenge reflected in Meituan's acquisition of Dingdong Maicai. In February this year, Meituan announced plans to acquire Dingdong Maicai's China business for an initial cash consideration of US$717 million. Although both use the front warehouse model, their product structures and warehousing methods are clearly different. Dingdong operates around fresh food and family meals, with about 3,000 daily SKUs; Xiaoxiang has expanded into groceries and daily necessities, with more than 8,000 SKUs in large warehouses. In overlapping markets such as East China, some of their warehouse locations duplicate coverage, but directly merging warehouse points would require readjusting delivery radii, in-warehouse product allocation, sorting processes, and capacity arrangements. A senior retail industry figure close to Meituan told All-Weather Technology that, in addition to preventing competitors from obtaining Dingdong, Meituan is also "spending money to buy time" through the acquisition. The person also revealed that, besides Dingdong, Meituan has also approached some traditional supermarket acquisition opportunities involving retail scenarios it had not previously covered. For Meituan, buying ready-made operating capabilities can shorten expansion time, but truly forming scale advantages still requires effectively organizing the products, procurement, and fulfillment capabilities of different systems.
The bigger retail battlefield
Front warehouses are a precise numbers game, but retail ultimately remains a hard business that returns to the common sense of "more, faster, better, cheaper." No single model can solve all retail needs, because the long-term coexistence of distant and near fields is determined not only by the cost structure shaped by centralized versus dispersed inventory, but also by the real match between timeliness and experience in different consumption scenarios. If a consumer wants to buy an obscure imported seasoning, distant-field e-commerce only needs to keep a small amount of inventory in a regional large warehouse to cover the whole country; but if a front warehouse were to stock a bottle in every community dark warehouse, it would face a long inventory turnover cycle. If someone urgently needs a carton of eggs before dinner, the fulfillment advantage of minute-level delivery from a local front warehouse is irreplaceable. The same is true of online and offline. Instant delivery can certainly meet efficient replenishment needs, but for fresh products, many people are still used to judging quality by seeing and selecting them in person, so physical supermarkets retain independent operating value. Operating models are becoming increasingly segmented, while the business boundaries of large platforms are becoming increasingly broad. This change is taking place in a stock era where traffic dividends have peaked.
Platforms are finding it increasingly difficult to achieve growth simply through new users and overall consumption expansion, while shopping channels for high-frequency consumption such as groceries are still changing, and winning transactions that originally occurred in other channels has become a new source of growth. The 2025 food delivery war further intensified this competition. Alibaba and JD.com increased investment in instant retail and delivery networks, and Meituan's original fulfillment advantages faced challenges. At the same time, the industry has also reached a moment of "portfolio reconstruction." After nearly a decade of exploration, the models of front warehouses, hard discounts, large physical stores, and regional supply chains have all matured. Platforms no longer need to explore from scratch, but can directly reorganize these retail capabilities through self-construction and acquisitions. The pressure of stock growth, intensifying competition, and new opportunities for supply integration have jointly driven this round of retail integration and expansion. In places few people notice, the relationship between retail transaction entry points and product supply is also changing. On October 21, Walmart announced a shopping access mechanism for third-party AI agents, planning to allow consumers to place orders directly through external AI assistants. This signals a possibility: future traffic entry points may be diluted by technology, and platforms may no longer be able to keep consumers firmly enclosed in their own apps. When delivery speed is no longer an exclusive moat, the value of traffic entry points is also being reshaped, and product quality, procurement costs, and supply chain depth naturally rise to become the key to competition.
The platform's potential advantage lies in allowing different retail formats to share users, technical systems, procurement resources, and fulfillment networks, reducing duplicated construction. But the cost and management logic of front warehouses, physical supermarkets, and community discount stores are vastly different, and forced integration may also bring enormous management friction. Retail, in the end, is a hard business of bending down to pick up coins. What Meituan ultimately needs to prove is not how many kinds of retail businesses it can operate, but whether, after putting these formats into the same system, it can truly reduce costs and increase efficiency through back-end supply chain synergy and create value greater than the sum of each business operating independently. Otherwise, an ever-expanding retail map is merely pouring huge amounts of capital into a swamp that was already as thin as a cicada's wing.
Comments