Tmall Global, cross-border, or a local entity: how to actually pick your way into China
Short version: if you're testing China and don't have a registered company there yet, start on Tmall Global (cross-border, no China entity required, roughly a ¥50,000 deposit to open a store). If you want to own the relationship and the data instead of renting a marketplace storefront, run your own independent cross-border setup through a bonded warehouse. And once China is a real, proven line of your business — not an experiment — that's when you set up a local entity (a WFOE) and sell domestically. The order matters more than the logos. Most brands pick the structure first and discover the strategy never fit it.
This is the question I get asked first, almost always in the wrong form. People ask "should we set up a WFOE?" or "do we need a Tmall flagship?" as if the legal structure were the strategy. It isn't. The structure is plumbing. It should follow from how sure you are about the market, how much control you need, and what you're willing to commit — not the other way round. So before the table: the three routes, what they actually are, and the trade you're making with each.
The three doors into China
Strip away the jargon and there are really three ways in. They differ on one axis that matters more than any other — how much of yourself you have to put on the ground.
1. Tmall Global (cross-border marketplace). You open a storefront on Alibaba's cross-border platform without registering a company in China. Goods ship in from overseas, usually through a bonded warehouse in a free-trade zone, and clear customs as personal imports when a consumer orders. You need your home-country incorporation documents and a trademark you own or are licensed to use, plus a store deposit in the region of ¥50,000. The platform owns the traffic, takes its commission, and largely owns the customer data. You're renting reach.
2. Independent cross-border. Same legal/customs status — goods come in cross-border under the personal-import regime — but instead of living on Tmall, you sell through your own channels: a WeChat mini-program store, a Douyin storefront, your own bonded-warehouse fulfilment. More work, more setup, but you own the relationship, the first-party data, and the margin you'd otherwise hand to a marketplace. This is the route brands underrate most.
3. A local entity (WFOE). You register a wholly foreign-owned enterprise in China, get a local business licence, a China corporate bank account, a domestically registered trademark, and you import under general trade. Now you can sell on domestic Tmall (Tmall Classic), run domestic logistics, invoice in RMB, and operate like a Chinese company — because for most purposes you are one. Full control, full commitment, full compliance load.
What you're actually trading
Here's the same three routes laid against the things that decide the call — not the brochure features, the real ones.
| Tmall Global | Independent cross-border | Local entity (WFOE) | |
|---|---|---|---|
| China company needed | No | No | Yes |
| Speed to live | Weeks | Weeks to a few months | Several months |
| Upfront commitment | Low (≈¥50k deposit) | Low–medium | High |
| Who owns the customer | Mostly the platform | You | You |
| Product registration | Lighter (personal-import rules) | Lighter (personal-import rules) | Full domestic registration |
| Sells domestically (offline, B2B, JD/Tmall Classic) | No | Limited | Yes |
| Best when | You're testing demand | You want data + margin | China is a proven, scaling line |
The tax point everyone gets backwards
People assume cross-border is the cheap door because the tax is lower. The first part is true: under the cross-border retail-import regime, customs duty is set to zero and VAT and consumption tax are levied on 70% of the statutory amount — so a typical good carrying 13% VAT lands around an effective 9.1%, with no tariff. That's genuinely favourable. But it comes with a leash: a single order is capped at ¥5,000 and each shopper has a ¥26,000 annual cross-border quota. Cross the single-order cap and you lose the duty exemption; cross the annual quota and the order simply can't come through cross-border at all.
What that means in practice: cross-border is brilliant for repeat consumer purchases under those limits, and useless for high-ticket items or anything you want to sell to businesses, sell offline, or scale past the quota. The tax break is real, but it's a consumer-import break, not a wholesale-import break. Plenty of brands pick cross-border for the tax and then hit a ceiling they could have seen coming.
The registration door swings both ways
The quieter advantage of cross-border is regulatory, not fiscal. Because goods enter as personal imports, many categories skip the heavy domestic pre-market registration — a cosmetic or a health-style product that would need a full filing to sell through general trade can often go cross-border far faster. For regulated categories, that head start is frequently the real reason to start cross-border, more than the tax.
The flip side is the same coin. A local entity carries the full compliance load — registration, an ICP filing to run a Chinese site, ongoing accounting and tax in-country — and in exchange you get the things cross-border can't give you: domestic shelf space, offline retail, B2B sales, the ability to take your brand anywhere a Chinese company can go. You're not buying convenience. You're buying permanence.
So which one?
Match the door to the question you're actually answering:
- "Will Chinese consumers even want this?" — Tmall Global. Lowest commitment, fastest read, and you're paying for the platform's traffic precisely because you don't yet know if the demand is there.
- "They want it — now I want the data and the margin." — Independent cross-border. Once you've proven pull, stop renting the relationship. Own the customer list, the retargeting, the repeat purchase. This is usually the most underused move on the board.
- "China is now a real, growing line of the business." — Local entity. When you're hitting cross-border quotas, want offline or B2B, or need to be a fully domestic player, set up the WFOE. Not before — the compliance overhead of a WFOE you're not using yet is just a tax on indecision.
And you don't have to choose once and forever. The most sensible path is sequential: prove demand on cross-border, build owned channels alongside it, and only stand up the local entity when the numbers make the overhead obviously worth it. Brands that start with the WFOE because it feels "serious" usually spend a year paying for infrastructure ahead of any demand to justify it.
Bottom line
Tmall Global, independent cross-border, and a local entity aren't a ranking from beginner to advanced — they're three different bets on how certain you are. Cross-border buys you a cheap, fast, low-commitment read with a real tax break but real quotas. A WFOE buys you permanence and full domestic reach at the cost of time and compliance. Pick the structure that fits how much you actually know, not the one that looks most committed in a board deck. The commitment should follow the evidence.
If you're weighing which door to walk through — or you started on one and aren't sure when to graduate to the next — that's the work I do. Reach out and we can pressure-test the plan. For the strategy that sits on top of whichever structure you choose, read the cross-border marketing playbook for entering Greater China.