How to test demand in Korea, Japan or China before starting to sell
Form a hypothesis, plan conversations and test behaviour to distinguish polite feedback from a purchase signal.
Article language: English

In this article
“Interesting product” sounds encouraging, but it is not an order. When testing demand in Asia, ask about a specific problem, the way it is currently handled and the conditions under which someone would consider a change. Look for behaviour and verifiable next steps. A friendly conversation has value in itself, but a market-entry calculation should not turn every compliment into future revenue.
At this stage you are not trying to prove that an entire country wants your offer. You are testing a hypothesis about a segment. The choice of Korea, Japan or China may follow an earlier comparison of markets. A demand test should lead to a more precise offer, another test or a justified decision to stop, with the limits of the evidence visible. The guide to choosing between Korea, Japan and China (in Polish) discusses criteria for the earlier comparison.
How do you write a testable hypothesis?
Name the audience, the situation and the cost of the problem in its own process. Add how it is solved now and what you propose to change. “Companies in Japan need automation” covers too many conditions. A more useful hypothesis says that a particular kind of team, performing a specific task, is looking for a way to reduce a stated difficulty. It remains an assumption until tested. If commissioning research, the guide to briefing market research in Asia (in Polish) can help define the questions and expected output; outside commissioning is not required for every test.
Define a signal sufficient for the next step. Depending on the product and stage, that might be agreement from the right person to assess an application, requirements supplied for a demonstration, or the start of a formal supplier-qualification process. Do not prescribe one universal number of interviews as proof of a market. More important is whether the sample reached the right people and produced material that answers the question.
What can existing data tell you?
Official statistics can help you understand the size and structure of a segment if their definition matches yours. Japan's e-Stat provides statistics by theme; China's National Data is a gateway to statistics from its statistical authority. Read each table's definition, period, unit and population. Do not bend a category to fit your offer.
Trade data may show imports of a group of goods, but cannot establish a need to buy from a new supplier. The guide to using HS trade data (in Polish) explains these limits. An industry report can justify another conversation, but not replace it. In your record, distinguish reasons to test a hypothesis from evidence obtained from an appropriate participant.
Consumer material calls for another kind of reading. As the article on reading Xiaohongshu reviews (in Polish) explains, these posts can reveal a use situation or a question, but a count of posts does not become a segment size or a confirmed purchase.
How do you select interviewees?
Set inclusion criteria: role, application, organisation type and relationship to the purchase decision. A user may understand the problem but not the purchase conditions. A distributor's salesperson may understand the channel without representing a user's daily work. Separate these roles as needed and record whose experience supports each answer.
For a service, distinguish user, buyer and payer, as discussed in the article on roles in a service model (in Polish). The needs of the person doing the work and the conditions for approving a purchase may require conversations with different participants.
Do not base the whole sample on one referral. Contacts from one circle may share experiences. Record recruitment methods and refusals. If you find companies through a directory or industry meeting, check their actual role. The guide to checking an Asian partner (in Polish) helps establish identity, but suitability for the question needs separate assessment.
How do you ask without leading?
Begin with the last specific occasion: what happened, who took part, what was difficult and what solution was chosen. Ask about actions and conditions the interviewee actually knows. Instead of putting a benefit into the question, ask for the current procedure. An answer based on an event is usually easier to interpret than a general opinion about the future.
Only then present the scope of your solution and ask about fit. Do not treat every reservation as an invitation to defend the product. Record the objection, its role and whether it can be checked. Persuading someone to agree politely is not the purpose. You need information that may also challenge the team's idea.
How do you prepare language and materials?
Agree the language of the conversation, a glossary of important terms and the scope of translation. The interviewer or interpreter should understand the aim of the test, not just the product. Questions must preserve the distinction between current practice, preference and readiness to act. The team can agree terminology and review of language versions in its localisation brief; the guide to preparing a localisation brief for Korea and Japan (in Polish) explains how to prepare one.
Show enough material to assess the hypothesis: an application description, a diagram or a demonstration. Mark what is ready and what is a concept. If the product does not yet exist, do not imply that delivery is available. Prepare an answer about the next step. Research should be fair to the interviewee and should not create commitments the company does not intend to fulfil. If an existing catalogue is used as material, the article on product catalogues for Korea and Japan (in Polish) discusses identifying a product variant; a catalogue does not establish delivery readiness.
What separates interest from a purchase signal?
Distinguish an opinion, an action and a formal commitment. Praise for a presentation is an opinion. Naming the person responsible for an assessment or supplying specific requirements is an action whose meaning depends on context. An order has a separate status. Do not combine these events into one “interested customers” category: it obscures the important differences.

Record what the recipient had to do. If a colleague of the presenter arranged a meeting, obtaining it may say little about the strength of the offer. If the right person starts an application assessment independently, that is a different signal. Describe each case. The goal is to explain the behaviour, not choose the most favourable interpretation for a report.
How do you explore willingness to pay?
Discuss the purchase process, existing solution, budget for the task and required approvals, if the interviewee is able to speak about them. Prepare a consistent offer with an explicit scope. A reaction to price alone, without delivery or support details, will be hard to interpret. Do not assume that stated acceptance of a price equals readiness to order.
If you plan a real offer or paid trial, approve its terms and ability to deliver. The guide to a B2B offer for Japan (in Polish) illustrates how scope, evidence and a next step can be connected. A refusal may reflect price, the problem's low importance, the wrong option or a procedure. Ask why and record how well the explanation is confirmed.
How should you interpret no response?
Silence is an observation about the contact process. By itself it does not show that a product is unnecessary. Check the addressee, delivery, clarity of the request and ability to take the proposed action. The article on an unanswered message to a Korean customer (in Polish) examines these questions without stereotypes. The test must allow you to see where reaching someone ends and assessing the offer begins.
Respect contact boundaries and plan any follow-up. Changing the message during the test may be necessary, but date and describe the change. If replies appear afterwards, identify the message version when comparing earlier and later contacts. Otherwise it is difficult to tell whether the product assessment changed or the recipient only then understood the proposal. Version records directly affect the conclusion.
What does a properly limited test look like?
A methodological example: a company wants to offer technical-document organisation to a defined group of manufacturers. First it asks people responsible for documents about their last update, difficulties and approval process. Only then does it present the service scope. It records whether the problem occurs and whether the interviewee influences the order. It invents neither a research result nor representativeness for the whole country.
If a need emerges, the next step is for the right person to assess an example scope. The company checks confidentiality requirements and its ability to prepare an offer. In Korea a subsequent process trial may take the form of a sales pilot, as described in the guide to a Korean sales pilot (in Polish). Research into the need thus ends with a specific condition for moving forward, instead of the general claim that “the market has potential”.
How do you document material and disagreements?
For each conversation record role, context, date, important observations and limits. Use quotations under agreed rules and do not attribute words the person did not say. Separate notes from the team's interpretation. If two people understand the same statement differently, return to the material or leave the uncertainty visible in the summary.
Compare cases using the same questions. Pay particular attention to answers contradicting the hypothesis: they may point to another segment or a wrongly named problem. The guide to analysing competitors in Korea (in Polish) helps test whether the recipient already solved the difficulty in a way the team missed. Do not remove a case because it makes a positive recommendation harder.
When should you narrow the segment instead of changing the product?
Compare situations where the problem matters clearly with indifferent cases. The difference may be the process, scale of use or responsible person's role. Do not change the product after every comment before examining the pattern. If the need occurs only in a particular application, narrowing the segment may be more justified than adding features for every interviewee at once.
Record the new group definition and test whether it follows observations or merely a wish to preserve a positive hypothesis. Contrary cases are useful: an organisation meets the criteria but sees no problem. Try to explain the difference without discarding the inconvenient answer. It may mark a boundary of application or a missing question to add next time.
Describe the new test separately, including the material version and recruitment method. Do not present a narrowed result as confirmation of the original broad claim. Summarise what the team learned about the audience and the conditions needed for the offer to make sense. This supports more precise contact and reduces the risk of investing in communications aimed at a group for which the problem does not arise in practice.
Organise interviews using a plan for B2B problem interviews. Ask about the last similar case and what people actually did. A specific story gives a different basis for a decision than a general statement that the product sounds interesting. The guide to B2B problem interviews in Asia (in Polish) develops the interview plan; a few interviews do not measure population frequency.
How do you end research with a decision?
The result should include the hypothesis, evidence, contrary cases and conditions for further work. Separate a decision to refine the proposal from a decision to start selling. With a limited sample, explain which group the findings concern. If the basis is insufficient, identify the gap and the cost of checking it. The company can then choose knowingly whether to continue or stop.
- The segment and interviewee roles match the hypothesis.
- Questions concern real situations and current solutions.
- The material distinguishes opinion from action.
- Versions of the offer and messages are recorded.
- Silence has not automatically been treated as lack of demand.
- Contradictions and limits remain visible.
- The next step follows from specified evidence.
Aeneia Oriens can help turn market research into questions for specific audiences. Present the product, segment and assumption to be tested. The scope can be based on the evidence needed for a decision, with language and materials suited to the market conversation.











