中国商品进入海外的经营路径

Start with the factory's two options. Selling 8,000 units to a German chain at US$4.20 is traditional or B2B trade: one buyer, one negotiation, one shipment, and payment after delivery. Retailing at US$9.90 on AliExpress is B2C: thousands of small orders, platform commission, advertising, shipping and returns. The price is higher in the second route, but so are the costs and the work.

The textbook's §2.1 and Alibaba's platform layout provide a map. Traditional foreign trade is a relay: factory to trading company to foreign importer to wholesaler to retailer. Cross-border B2B shortens the relay: an overseas firm buys in bulk directly from a Chinese supplier through a platform such as Alibaba.com. Cross-border B2C or DTC removes most intermediaries: a Chinese seller reaches overseas consumers through AliExpress, SheIn, TikTok Shop or its own site. These routes coexist; they are not stages that replace one another.

Alibaba's layout illustrates the mapping. Alibaba.com is export B2B; AliExpress is export B2C; Tmall Global and Kaola are import B2C; Lazada serves Southeast Asia, Trendyol serves Turkey, and Daraz covers Pakistan, Bangladesh, Sri Lanka, Nepal and Myanmar. The same group runs different routes because target buyers and markets differ. A supplier reading these names can tell who the buyer is and which direction the goods flow.

The customs series gives the scale, with a scope warning. China's cross-border ecommerce exports rose from about CNY 611.6 billion in 2018 to about CNY 2,267.2 billion in 2025, and the export share of total cross-border ecommerce rose from about 58% to about 80% (SRC-038). These are customs-counted cross-border transactions. They are not domestic online retail, and they do not measure platform GMV. Comparing them with a domestic retail figure would produce a meaningless growth claim.

Route choice is a control trade-off. In B2B, the buyer is concentrated, orders are large and payment terms matter; the seller negotiates with a professional buyer but does not pay for consumer traffic. In B2C, the seller reaches many consumers but must pay for acquisition, fulfilment and returns. §1.7 flags rising traffic cost as one of the hardest challenges, which is exactly why some firms stay in B2B while others build DTC.

DTC is often misunderstood. It is not simply "a store on a platform". DTC means the brand owns the customer relationship and much of the data, so it can manage repeat purchase and pricing directly. A marketplace store may carry the brand name, but the platform still controls discovery, rules and much of the customer data. DTC therefore requires capabilities that a factory may not have: content, acquisition, service and retention.

Overseas warehouses and FBA change the delivery equation. Goods are shipped in bulk before orders arrive, stored locally and dispatched quickly. This shortens last-mile time and can lift conversion, but it requires a forecast. If demand is lower than expected, storage fees and tied-up cash accumulate. If demand is higher, stock-outs lose sales. The warehouse does not remove duty or import compliance: goods still enter the destination country.

Offline channels remain part of the mix. The Canton Fair, held twice a year since 1957, continues to bring buyers and suppliers together; the 138th session ran from 15 October to 4 November 2025 with 1.55 million square metres of exhibition space and 74,600 booths (SRC-045). Many B2B relationships are "met online, closed offline": the platform starts the conversation, the fair or a factory visit closes it.

A practical decision therefore uses four questions. Who is the buyer: a firm or a consumer? How large is the order and how concentrated is demand? Who bears inventory, traffic and returns? What evidence do you have for demand? The answers point to a route or a mix. The factory in the opening can run both: B2B for volume and cash, B2C for testing and learning, DTC only if it can fund acquisition.