Regional demand segmentation is the filter a national market size figure cannot provide
Regional demand segmentation in China is the practice of splitting the national market into customs catchments and consumer geographies, used to decide where a foreign brand places its first bonded stock and its first marketing budget. A national total, by contrast, describes a case no operator can execute: every cost and every delivery promise in China is regional.
The national figures are not small. Total retail sales of consumer goods reached RMB 50.12 trillion in 2025, up 3.7 percent, of which online physical-goods sales were RMB 13.09 trillion, or 26.1 percent (National Bureau of Statistics, December 2025 retail release). But the demand pool has stopped expanding: China had 976 million online shoppers in June 2025, up 0.1 percent in six months, at 86.9 percent penetration (CNNIC, 56th Statistical Report). When the number of shoppers is flat, the only growth available to a new entrant is share taken from someone else in a specific place.
Cross-border import destinations show where the infrastructure sits
A cross-border import destination share is the percentage of national cross-border e-commerce import value attributed to the province in which the consignment is declared, used as a proxy for where the bonded operators, importers of record and platform sellers serving that demand are based. China's cross-border e-commerce imports were worth RMB 570.2 billion in 2025, up 2.7 percent, and the goods were entirely consumer products (GACC, 2025 statistics).
The concentration is the story. Five destinations take 44.3 percent of the trade, and three of them form a single contiguous manufacturing-and-consumption belt.
| Destination | 2025 import share | What it rewards | Bonded port that serves it |
|---|---|---|---|
| Guangdong | 14.8% | Hong Kong adjacency, Cantonese content, electronics and fashion | Guangzhou / Shenzhen |
| Jiangsu | 8.6% | Industrial consumers, inland distribution reach | Shanghai / Ningbo catchment |
| Zhejiang | 8.5% | E-commerce operations depth, cross-border seller density | Ningbo / Hangzhou |
| Shanghai | 6.8% | Highest-value repeat purchase, beauty and premium food | Shanghai |
| Shandong | 5.6% | Northern catchment, cold-chain and food categories | Qingdao |
A declaration point is not a consumption point: one Shanghai bonded warehouse can ship to Harbin, and a Guangdong seller can serve a Chengdu customer. Read the shares as a map of where competent operators already are — which is why they also predict where your fulfilment partner, TP and category rivals cluster.
The three core geographies reward different launches
China's three core demand geographies — the Greater Bay Area, the Yangtze Delta and the Beijing-Tianjin-Hebei cluster — are distinct consumer markets rather than three slices of one, used to choose which single region a brand enters first instead of attempting all three at once. Each has a different cost of attention, a different category depth and a different language register.
The Yangtze Delta is where premium categories are decided: the densest cluster of beauty, personal-care and imported-food operators, the largest pool of TP and agency talent, and the highest per-capita incomes outside the two first-tier cities. It is also the most expensive place to buy attention.
The Greater Bay Area, measured nationally as Guangdong, is the volume and speed market: Hong Kong adjacency that simplifies sourcing and payment, a Cantonese-language content ecosystem that runs semi-independently of Mandarin platforms, and deep electronics and fashion supply chains.
Beijing-Tianjin-Hebei is the policy and northern-logistics market. Tianjin is a new tier-1 city in the 2025 Yicai ranking and sits on a bonded port with cold-chain capacity, and the cluster's consumer base is large but more price-stratified than the south. Competition for attention is materially lower than in Shanghai, which makes it a rational first test for food, nutrition and household categories aimed at northern distribution.
City tiers rank infrastructure; they do not rank your customer
China's city-tier system is a commercial classification of 337 prefecture-level and above cities by commercial-resource density, used by brands to sequence a city rollout rather than to size demand for a product. The 2025 edition holds four tier-1 cities, fifteen new tier-1 cities, then 30 tier-2, 70 tier-3, 90 tier-4 and 128 tier-5 (Yicai, 2025 city ranking).
The ranking measures commercial infrastructure: how many brands are already present, how dense the retail and media ecosystem is, how quickly new formats arrive. That is exactly what a launch needs to know about execution feasibility — and it is not the same question as whether shoppers in that city pay a premium in your category. Plenty of tier-2 cities outperform tier-1 cities in a single category, and a brand that treats the tier list as a demand forecast will over-fund the top four cities and under-fund the eight cities that actually buy its product.
Income quintiles, not provinces, set the price ceiling
Household income quintiles in China are the five equal-sized groups into which the National Bureau of Statistics sorts households by per-capita disposable income, used to count how many buyers can afford a given retail price. In 2025 per-capita disposable income was RMB 43,377, up 5.0 percent, with urban income at RMB 56,502 against rural income at RMB 24,456 — a ratio of 2.31 (NBS, 2025 household income release).
The quintile averages ran from RMB 10,150 in the lowest group to RMB 103,778 in the highest, and the top two groups together cover roughly 40 percent of households. That band — not the whole population — is the addressable market for most imported consumer goods, and it is geographically uneven: it clusters in the tier-1 and new tier-1 cities and thins out quickly below them.
Two more filters sit on top. The landed price is set by the cross-border tax regime rather than the shelf price at home: within the RMB 5,000 per-order and RMB 26,000 per-person annual limits, tariff is zero and import VAT and consumption tax are levied at 70 percent of the statutory rate. And checkout price carries platform commission, 2 to 5 percent on Tmall Global plus a cross-border payment fee, so the margin test belongs in the region decision.
The county market is a second-wave question, not a first-entry one
County-level and rural markets are China's consumption outside city districts, measured separately because their growth rate and channel mix now diverge from urban China. In 2025 they accounted for 38.7 percent of all retail sales, and rural retail grew 4.1 percent against 3.6 percent in cities, with rural online retail passing RMB 3 trillion for the first time (MOFCOM, 2025 wholesale and retail review).
The number is real and it is regularly misused. County consumption is driven by availability and value-for-money formats, not by the premium, provenance-led positioning an imported brand carries, so county coverage is usually a year-two consequence of platform reach rather than a reason to open a channel. Urbanisation reached 67.89 percent at the end of 2025 (NBS statistical communiqué, 2025) — a third of the population outside the urban retail footprint, and very few buyers at a premium price point.
The bonded port decides what you can promise
A bonded warehouse location is the customs-supervised zone where 1210 retail-import stock is held before an order is placed, used to convert a national market into a delivery promise for a defined geography. Bonded retail import accounts for more than 90 percent of China's cross-border retail imports, on the customs administration's own account of the 2025 destruction rules for unsold stock (GACC Order No. 79).
Placement is therefore a demand decision, not a logistics afterthought. Three constraints interact: the catchment the port can reach inside the delivery window you will advertise; the category supervision the port holds, since fresh and frozen goods are restricted to bonded retail import and must clear inspection at a designated site; and the return route, because a bonded return that cannot be resold is a write-off rather than a refund.
Note what the port map and the demand map do and do not share. Four of the five largest destinations are covered by a bonded port in the same province or municipality; the fifth is Beijing-Tianjin-Hebei, which carries far less cross-border import value than its population implies. That gap is an opportunity for a northern-first launch and a warning for anyone who assumes northern demand equals northern import volume.
A two-week method for choosing the first region
Region selection is the sequence of filters that narrows China from 337 cities to one bonded port and one launch city, used before any inventory commitment is made. It can be run on public data in about two weeks.
- Pull the customs destination shares and confirm which ports serve your two candidate regions.
- Check your category's weight in the import mix — beauty, fragrance and personal care took 27.2 percent of cross-border imports in 2025, food and fresh produce 23.9 percent — and ask whether your category is a leader or a rounding error in each region.
- Apply the income quintile test to your landed price, not to your home-market price.
- Verify the port's supervision status for your category, including cold-chain and designated-site requirements.
- Run a bonded test with a narrow SKU set in one port, then compare conversion, return reason codes and repeat purchase against the second region before scaling.
How GOODSINFINITE sequences regional entry
GOODSINFINITE acts as importer of record and bonded fulfilment partner, which makes the region question concrete: the first decision we take with a brand is which of the five bonded locations holds the launch stock, because that single choice fixes the delivery promise, the return route and the replenishment cycle. From there the sequence runs category check against the positive list, income-band price test, single-port 1210 launch, then expansion on evidence. For the wider sequence, start with our China market entry guide and the cross-border operations glossary.
FAQ
Which Chinese provinces take the largest share of cross-border e-commerce imports? In 2025 the top five were Guangdong at 14.8 percent, Jiangsu at 8.6 percent, Zhejiang at 8.5 percent, Shanghai at 6.8 percent and Shandong at 5.6 percent — 44.3 percent of the national total. Read them as infrastructure gravity rather than a consumer census: they record where the consignment is declared and where the bonded operator, importer of record and platform seller sit.
Should a foreign brand enter the Greater Bay Area or the Yangtze Delta first? It depends on category depth. The Yangtze Delta — Shanghai, Jiangsu and Zhejiang — was 23.9 percent of cross-border imports in 2025 and holds the deepest concentration of beauty, personal-care and premium-food operations. Guangdong alone took 14.8 percent with Hong Kong adjacency, Cantonese content and a strong electronics and fashion base. Choose the region where your category is less crowded and your price band is thicker.
Are Chinese city tiers a reliable guide to where imported products sell? No. Yicai's 2025 ranking sorts 337 cities into four tier-1, fifteen new tier-1, thirty tier-2, seventy tier-3, ninety tier-4 and 128 tier-5 cities on commercial-resource density, connectivity and consumer activity. That ranks infrastructure and execution feasibility, not demand for your category at your price.
How many Chinese households can afford an imported product at a given price? Use the quintiles. NBS put per-capita disposable income at RMB 43,377 in 2025, with quintile averages from RMB 10,150 to RMB 103,778 and urban income at 2.31 times rural income. The top two quintiles — roughly 40 percent of households — behave like a market for imported goods, and they cluster in the tier-1 and new tier-1 cities.
Does the bonded warehouse location change what a brand can sell? Yes. Stock can only promise the delivery window its port catchment supports, and only carry categories that port supervises — fresh and frozen goods are restricted to bonded retail import under 1210 and must clear a designated site. With bonded import above 90 percent of cross-border retail imports, port choice largely sets the service promise and the replenishment cycle.
Sources
- GACC — 2025 cross-border e-commerce statistics: imports of RMB 570.2 billion, destination shares by province
- NBS — Total Retail Sales of Consumer Goods in December 2025: RMB 50.12 trillion, online physical goods 26.1 percent
- NBS — Households' Income and Consumption Expenditure in 2025: per-capita disposable income RMB 43,377, quintiles RMB 10,150 to RMB 103,778
- MOFCOM, Department of Circulation Development — 2025 wholesale and retail review: county and rural markets 38.7 percent of total retail
- Yicai — 2025 city-tier ranking of 337 Chinese cities (four tier-1, fifteen new tier-1)
- CNNIC — 56th Statistical Report on China's Internet Development: 976 million online shoppers, 86.9 percent penetration