出海 Market Entry
Chinese companies going abroad often win on product and lose on everything around it: a market chosen by headline size, payments that do not fit local habits, a translation instead of localisation, data rules discovered after launch. This skill plans the whole entry: which market, and the workstreams that decide whether the product can actually sell there.
Write in Chinese, English or both, as the person prefers. For data transfer law, see pipl-gdpr-crosswalk; for marketplace listings, see cross-border-listing.
What This Skill Produces
- A market scorecard: candidate markets scored on the factors that matter for this product
- A twelve-month entry plan by workstream, with milestones
- A compliance checklist for the chosen market: entity, tax, product rules, data
- The top risks, each with a mitigation and an owner
Required Inputs
Ask for these if not provided:
- The product: what it is, price point, B2B or B2C, physical goods, software or service
- Candidate markets, or "help us choose"
- Current traction in China, and what made it work
- Budget and team available for the first year
- Constraints: data must stay in China, payment needs, sensitive categories
Working from a brief
Given the company and the markets in question, make the call and write the first-year plan anyway: choose between the markets named on stated assumptions, then give the twelve-month plan with entity, payments, localisation, compliance and channels. Mark figures to verify instead of asking first.
Framework
1. Choose the market. Score each candidate from 1 to 5 on:
- Demand evidence for this category (search data, competitor revenue, imports)
- Willingness and ability to pay at this price
- Competition, including local and other Chinese players
- Channel access (marketplaces, app stores, distributors, direct)
- Regulatory and data difficulty for this product
- Fit of the China playbook (what transfers, what does not)
- Cost of operating (people, logistics, marketing)
Weight the factors for the business model and show the arithmetic.
2. Plan by workstream:
- Entity and tax: local entity, branch, or operate cross-border first; VAT or GST registration
- Payments: local methods (cards, e-wallets, bank transfer, cash on delivery); settlement and currency
- Localisation: language, pricing, imagery, customer service hours and channels, not only translation
- Compliance: product standards and certification, consumer law, advertising rules, data protection
- Channels: marketplaces, social commerce, search, partners, sales team
- Team: local hires versus China-based; the first three roles to hire
- Brand: name check in the local language, trademark registration before launch
3. Phase it: months 1 to 3 (validate), 4 to 6 (launch), 7 to 12 (scale or stop), with a go or stop metric at each gate.
Output Format
出海计划:[产品] → [市场]
1. Market scorecard | Factor | Weight | [Market A] | [Market B] | [Market C] |, with totals and the recommendation
2. Twelve-month plan | Workstream | Months 1 to 3 | 4 to 6 | 7 to 12 | Owner |
3. Compliance checklist | Area | Requirement | Status | Source to verify |
4. Risks | Risk | Likelihood | Impact | Mitigation | Owner |
5. Gates | Gate | Metric | Go if | Stop if |
End with: "Regulatory points are a starting checklist, not legal advice. Confirm each with a qualified adviser in the target market before committing spend."
Quality Checks
Anti-Patterns
- Choosing by population or GDP alone. Demand for this product at this price is what matters.
- Translation as localisation. Pricing, payment, imagery and service hours matter as much as language.
- Copying the China playbook whole. Channels and consumer habits differ; test what transfers.
- Trademark after launch. Register first; squatting is common.
- Treating regulatory notes as final. Rules change; verify locally.
Example Trigger Phrases
- "我们是做智能家居的,想出海,东南亚还是中东?"
- "帮我做一个进入美国市场的出海计划。"
- "Plan our expansion from China into Europe for the first year."
- "出海需要注意哪些合规问题?"