Skip to content
Interface language: English
Oriens / Signal

B2B customer interviews in Korea, Japan and China: learn from the actual work

Plan neutral B2B interviews, work with interpreters and separate evidence, interpretations and open questions.

Article language: English

An open atlas and notebooks lie on a desk overlooking a city across the river.
In this article

“Very interesting product” is pleasant to hear. It is harder to issue an invoice for it. An interview with a potential B2B customer in Korea, Japan or China should explain the situation: what the company is trying to do, how it manages now and what makes change difficult. Start with events the participant remembers; show the solution later. A polite opinion should not replace evidence of need, budget or ability to buy.

Language differences require preparation, but they do not justify a story about one uniform way of thinking across an entire country. We study the experience of a particular person in a particular role. This guide helps plan a conversation, avoid suggesting answers and separate what was heard from our conclusions. The example questions and situations need adaptation; they are not a script for every industry.

Choose the decision the interview should inform

Before recruitment, write down what you will do with the answers. You might choose an application, examine how a sample is assessed or determine whether local partner support is needed. These are different goals and need not fit into one session. If every question seems equally important, choose the assumption whose being wrong would most change the next plan. One call does not have to solve the whole expansion.

For example: find out how plants replace a worn component and what makes an alternative difficult to approve. This decision leads to questions about the last replacement, the current supplier and assessment of specifications. “Understand the market” leads everywhere. Include hypotheses in the Asia market research brief so the researcher can propose a suitable way to collect information. The guide to briefing market research in Asia (in Polish) describes how to set its scope.

If the interviews are intended to support a choice of market, the guide to comparing Korea, Japan and China (in Polish) helps establish shared criteria and the roles relevant to the questions beforehand. Some interviews serve other decisions, and interviews alone do not establish market size.

Recruit for the experience you want to reconstruct

A person at the right company may not know the process. For equipment use, speak with a user or maintenance person. For approval of a new supplier, find someone who participated in a similar assessment. Briefly describe the subject in the invitation and check whether the participant can discuss that experience. Do not request information they cannot access or disclose. For a service, distinguish the user, buyer and payer, as discussed in the article on roles in a service model (in Polish). Daily use and the reason for continuing a programme may require different participants, including after purchase.

If a distributor provides introductions, document that recruitment route. Speaking only to its satisfied customers may show one channel and one service model. It remains useful material, provided it is not called the opinion of the whole market. Report participants' roles, how they were selected and the interview topics. The result describes the cases collected; estimating a problem's frequency in a wider population would require separately designed research. Connect recruitment with demand verification in Asia. The guide to verifying demand in Asia (in Polish) discusses the evidence behind that decision. Collecting answers alone does not decide what they can support.

Prepare a short guide and try it before the meeting

The GOV.UK Service Manual on in-depth interviews describes them as a way to learn about users, their circumstances and needs. It recommends preparing topics, prompts and a trial of the guide. These are methods for service research, not a study of customer behaviour in Asia. Use them as a starting point for your own B2B interviews.

Include an opening, reconstruction of a specific event, clarification and closing. For each topic record what information you need and what would be too general. If asking about reliability, plan how to reach an example of failure or a check of operation. In a trial with a colleague, see whether the questions make sense without knowledge of your planned offer. Finding a shortcut in your thinking now saves the participant's time.

Reconstruct the last case instead of ordering an opinion about the future

Ask the participant to describe the last time the component was replaced. Establish the reason, sequence, people involved and difficult moment. If they say “it usually takes a long time”, ask what took time in that case. It might have been assessing specifications or access to equipment for a test. Those answers suggest different actions.

When researching a completed decision against your own offer, the guide to reviewing a lost B2B proposal (in Polish) also asks what the buyer compared, when uncertainty arose and which route they chose. Do not replace that history with the seller's guess about price.

Questions about the future can help, but record the answer as a statement about the future. “We would gladly look at an alternative” is not an account of the last purchase. “We cannot change the component before the current project ends” identifies a condition worth clarifying. Listen for workarounds too. Manual records, a service call or postponing replacement can reveal how people actually cope. Do not judge the workaround before understanding why it exists.

Do not suggest the answer by naming your benefit

Compare two questions. “Would automating parts ordering save you time?” contains the desired benefit. “How was the last parts order prepared and where was an extra check needed?” lets you discover whether time is the main problem at all. Ordering might be quick while the difficult part is choosing the correct variant and responsibility for an error.

Do not replace a difficult question with an attractive adjective. “Modern”, “flexible” and “comprehensive” allow agreement without shared meaning. Ask for an example of behaviour or a condition. If the participant uses their own industry term, retain it and check its meaning. Material for a Japanese and Korean technical glossary can be an important interview result: it shows how the audience names the problem. The guide to preparing a technical glossary (in Polish) explains how to organise such terms.

Separate research from the product demonstration

Showing a solution first may turn the rest of the conversation into a discussion of fit with what was just shown. Plan a separate demonstration and clearly signal the change of purpose. First learn how work is done now; then examine the reaction to a specific proposal. Compare notes from both parts rather than treating every compliment as confirmation of the original hypothesis.

Aeneia conceptual photography

For the demonstration choose one use and show limitations too. Ask where the solution would not fit the described process and what would need checking before assessment. A missing feature is a signal to clarify, not an automatic instruction to expand the product. A map of B2B buying participants helps determine whether the participant is assessing their own work, another department's requirements or a hypothetical possibility. The guide to mapping B2B buying participants in Korea and Japan (in Polish) explains how to develop such a map.

Prepare for an interpreter and materials in two languages

Before the session give the interpreter the research goal, product definitions and the difference between a question and a proposition. In the component example, emergency replacement may need to be distinguished from planned replacement. Calling both simply “change” loses an important detail. Agree how to ask for clarification and mark uncertain translations. Do not expect the interpreter alone to settle the technical meaning of an imprecise sentence.

Speak in short segments and leave time for answers. Do not direct the whole conversation to the interpreter: the counterpart remains a participant. Retain original terms in notes where their meaning needs checking. Machine translation may help prepare a working record but should not decide ambiguous requirements. Use agreed terminology in later sales material, using the guide to localisation between Poland and Asia (in Polish) for context.

Agree participation, observation and recording

The participant should know beforehand why the session takes place, who will attend and how the material will be used. The GOV.UK guide to informed consent addresses purpose, data, recipients of findings and informing people about observation and recording. It helps prepare participant information; it does not replace determining the legal and organisational rules for your project.

Before participation, obtain and retain a record of the participant’s informed, voluntary agreement to take part. If observation or recording is planned, explain its scope beforehand and ask for agreement appropriate to the planned participation. Explain that the participant can stop taking part or withdraw consent, how to communicate that decision and what happens next under the agreed arrangements. Separately determine the legal bases for processing, access, storage, retention and the effects of withdrawal with those responsible for legal and data protection matters. Agreement to participate is not automatically the legal basis for every use of personal data.

Do not switch on recording as an automatic background feature. Establish the need and an acceptable way to work. If notes without identifying details suffice, consider them. Let the participant skip a question and do not press for other suppliers' confidential terms. Researching need does not require collecting someone else's secrets. If a participant offers a document, first check whether they can share it and how the team may use it.

Record evidence, interpretation and the question separately

After the session divide the note into three fields: what the participant described or showed; your interpretation; and the open question. Example: the participant reported that the last replacement required an extra test; the team's interpretation is that assessing an alternative may be an important barrier; the open question concerns the conditions of that test. These are different levels of certainty and need different next actions.

Do not count themes without saying whom you interviewed and how participants were found. A few consistent accounts can justify another hypothesis, but do not automatically give the problem's share across a country. Describe differing cases just as carefully. The guide to reading a Korean market report (in Polish) requires checking the scope of the data; your own interviews deserve the same honesty.

Practise distinguishing weaker and stronger signals

Suppose someone says a new system looks interesting and they could return to the topic. That is a reaction to the proposal, not an explanation of the current process. Record it without adding purchase intent. Ask when a return would make sense and which material would help assessment. If the answer remains general, treat it as an open ending. You do not need to force a declaration to regard the interview as well conducted.

A second participant describes a recent part change, an extra test and who assessed its result. This is richer material about the buying mechanism. It still does not prove your offer fits the process or that another replacement will occur soon. Ask about conditions of use and compare them with product capabilities. A stronger account of the problem does not remove the need to test fit. Rushing here can turn good research into a poor sales forecast.

In a third case, the participant explains that the entire process is outsourced. The wrong role may have been recruited or the segment described incorrectly. Record the case and improve recruitment. Do not discard an answer as useless merely because it did not fit the plan. Knowing where a particular audience's need ends can help select an integrator, service provider or another application.

Edit the finding before it reaches a management slide

Take the working sentence “customers expect faster delivery”. Ask who said so, in what situation and compared with what. Perhaps one participant needed a part after a failure and another was planning stock for a future project. Combining them into one recommendation could lead to costly inventory even though one problem needs an emergency procedure and the other better planning. A shorter finding is not always more accurate.

A corrected finding should state its scope: in the described failure, time to obtain a part affected the repair choice; how often this occurs among other recipients remains to be checked. Propose an action separately. If the team cannot distinguish the account from a product idea, return to the notes. An elegant later presentation must not erase uncertainty that still matters to the decision.

Turn the result into a choice of next action

For each important finding record a possible response. If the problem is choosing a variant, prepare a way to compare specifications. If a test is the obstacle, design a product sample assessment plan. The guide to planning a product sample assessment (in Polish) explains how to match a test to its question. If maintenance experience is missing, find someone with it. Do not assign every finding the action “improve the presentation”. Slides cannot resolve every question hidden there.

If, after separate research into needs, the company prepares a proposal for a recipient in Japan, the guide to a B2B offer for a customer in Japan (in Polish) helps organise its scope and next step. An interview is not the participant’s agreement to buy.

When accepting interview work, check whether you can identify the decision, source and limitation for each important finding. If you are preparing Asia market research, Aeneia Oriens can discuss the goal and possible support. Participant recruitment and interviewing need separate agreement.

Aeneia / Aeneia editorial

Editorial article. Source links are included in the text.

About Aeneia

Brands we have worked with

  • L’Oréal
  • Garnier
  • Netflix
  • Carrefour
  • Warner Bros. Discovery
  • Biedronka
  • HBO Max
  • Wolt
  • Żabka
  • COSRX
  • Embassy of China in Poland
  • Mixa
Aeneia / Initium

From knowledge to a decision

Tell us what you want to put in order. We will agree on where to start.

Contact Aeneia

We partner with and use solutions from:

  • Google
  • OpenAI
  • Microsoft
  • Stripe
  • Anthropic
  • Linear