The hardest risks to identify have not yet been framed as questions

At the beginning of an investment-related investigation, the target company is usually already known.

For many years, I have supported institutional investors in the private and public markets with China-related investment research. In most projects, the client has already formed an initial investment thesis and enters the investigation with a set of questions to be tested.

Will the market continue to grow?
Does the technology represent a genuine barrier to entry?
Are the company’s relationships with key customers stable?
Can the management team deliver its business plan?

All of these are important questions.

But the risks that matter most to an investment decision are not always included in the initial research scope.

Investors may know which company they are considering investing in without yet knowing what they should be most concerned about.

A defined target does not mean that the risks have been defined

When an investment team decides to investigate a company further, it has usually formed an initial view.

That view may be based on market growth, technological advantages, customer resources, management experience or historical performance. The investigation naturally begins with those factors.

An investment thesis, however, determines where the research begins. It does not guarantee that every material issue has already been captured.

A company’s public materials may emphasize technological leadership, while the real constraint on commercialization is the length of the customer validation cycle.

Management may repeatedly point to growth in orders, while participants elsewhere in the value chain are more concerned about actual delivery, end-market consumption or the quality of cash collection.

A client may initially want to understand the founder’s industry relationships, only for interviews to reveal a different question: whether the company depends too heavily on the founder’s personal capabilities and connections.

At the beginning of a project, risks of this kind may not yet have clear names.

They often appear first as details that do not fit: an unexplained timing gap, conflicting accounts from two types of source, an advantage that is repeatedly asserted but rarely supported by direct experience, or an issue treated as peripheral in formal materials but repeatedly raised by frontline participants.

The difficulty of an investigation is not limited to deciding whether an answer is credible.

Sometimes the more important task is recognizing that the original question has not yet reached the issue that actually needs to be tested.

Financial, legal and independent primary research examine different evidence

Institutional investors may commission independent primary research even when financial and legal due diligence are already under way.

This is not a duplication of the same work. Each discipline enters through a different body of evidence and answers a different set of questions.

The three disciplines do not replace one another. They observe the same investment target through different evidential entry points.

Nor is the sole purpose of independent primary research to look for negative information that management has not disclosed.

It has a more fundamental role: identifying, within fragmented frontline information, issues that have not yet entered the formal verification list.

An anomaly identified in a frontline interview may need to be tested by the financial team against transaction data. A lawyer may need to determine whether a particular arrangement creates a legal or compliance risk. Further technical work may establish that an initial concern is not supported.

The value of independent primary research is not that it can answer every question on its own.

Its value is that it gives questions previously outside the field of investigation an opportunity to be tested.

Standardized tools are effective when the questions are already defined

When the research questions and response categories are clear, structured surveys, standardized interviews and AI-assisted research tools have obvious value.

They can reach more respondents quickly, compare products, prices and services along consistent dimensions, measure the distribution of market preferences, purchasing intentions or industry expectations, and improve the efficiency of organizing and summarizing information.

If the objective is to understand whether customers prioritize price, performance, delivery time or after-sales service when selecting a supplier, a structured survey can produce a clear comparative result. If the task is to compare usage, satisfaction or future purchasing plans across several products, standardized questions also reduce variation in how responses are framed.

These tools solve an important problem: how to answer defined questions more efficiently.

But they depend on one underlying condition.

The researcher must already know what to ask.

A questionnaire can collect information that has already been written into a question. In investment-related research, however, some of the most important risks have not yet entered the initial research framework.

Risks do not always first appear as conclusions

In one investment-related research project, the client regarded a particular operating decision by the target company as a positive step that could support future growth.

Management had presented a clear rationale. The decision was expected to increase short-term volume and accelerate market penetration. At the beginning of the project, it was not treated as a central risk.

As we conducted interviews with distributors, customers and other market participants, however, a different possibility began to emerge.

Several participants described the company’s pricing as unusually aggressive. While the approach might increase sales volume in the short term, it could also reduce the margin available to distribution partners.

That observation produced a new set of questions that had not appeared in the original interview guide.

At this price level, how much margin would remain for the sales channel?
Would distribution partners still have an incentive to promote the product?
Would customers continue to choose the company’s products under the same conditions?
If volume increased, could profitability be sustained?
If margins contracted, would the company still be able to invest in product development and quality control?

Further interviews and desk research revealed a possible chain of consequences:

• Aggressive pricing

• Pressure on channel margins

• Weaker incentives to promote the product

• Actual sales potentially falling short of expectations

• Further contraction in profit margins

• Less capacity for future investment

• An adverse effect on long-term competitiveness

The investigation did not begin by asking, “Will this strategy fail?”

What appeared first was a mismatch between the growth path described internally and the economic incentives described by external market participants.

By following that mismatch, a risk that had not originally been defined as a research question gradually became something that could be investigated and tested.

Independent verification is not intended to oppose management

The purpose of this work is not to disprove management’s decisions or manufacture a conclusion that conflicts with its account.

Management speaks from information and experience inside the company. Customers, suppliers, distributors, former employees and competitors observe the same business from different positions.

A difference in their accounts does not necessarily mean that anyone is distorting the facts.

The same business can look different because the observers occupy different roles, see different parts of the operation, focus on different periods and respond to different economic incentives.

Independent verification is therefore not a binary exercise in deciding whether management is right or wrong.

It examines whether there is a distance between the internal account and the external market response that could matter to the investment decision.

A new lead is not a new conclusion

Information that falls outside the interview guide may be important, but it does not automatically become a risk merely because it appeared unexpectedly.

An issue raised by one interviewee may apply only to a particular region, project or period. It may come from direct experience or from internal hearsay. It may reveal a genuine anomaly, or it may simply reflect a normal difference between what people in different roles are able to observe.

After identifying a new lead, the investigation must therefore ask:

• Is the statement based on the interviewee’s direct experience, or is it an opinion or second-hand account?

• Did the interviewee’s role and scope of responsibility place them in a position to observe the relevant facts?

• Have other customers, employees, business partners or industry participants observed a similar pattern?

• Is the issue isolated and temporary, or does it represent a continuing operating pattern?

• Is it consistent with public information, the organization’s structure and historical developments?

• If the lead is valid, which part of the investment thesis would it affect?

• Can it be verified independently through other evidence?

The value of an in-depth interview is not that every remark outside the guide should be included in the final report.

Its value is that it gives the researcher an opportunity to discover a new question and then determine, through further interviews, desk research and other professional due diligence, whether that question belongs in the final assessment.

A complete research process is not simply “ask questions and collect answers.” It is closer to:

Set an initial question
→ Obtain a lead
→ Redefine the question
→ Identify new sources
→ Cross-check the evidence
→ Assess its significance to the investment decision

This is a concrete application, within a single project, of the iterative process of independent verification.

Evidence discipline separates confidence from evidential strength

At FUWA LAB, “evidence discipline” means refusing to stop at the fact that someone has made a statement. It requires asking whether the information is observable, verifiable and supported by independent evidence.

A source speaking with confidence does not make the statement a fact. Several interviewees expressing the same view does not necessarily mean that their accounts are based on independent underlying sources. And when new information conflicts with an existing investment thesis, the conflict alone does not prove that the opposite conclusion is correct.

The strength of the language in a report must correspond to the strength of the evidence.

FUWA LAB distinguishes among the following formulations:

• “We tend to believe that…”

• “The information currently available supports…”

• “We are currently unable to verify this independently.”

These are not merely stylistic variations.

They distinguish between what has been verified, what is supported by the available evidence and what still requires qualification.

Caution is not the same as ambiguity.

Avoiding conclusions that go beyond the evidence is itself part of the professional discipline required in investment research.

Unknown risks cannot be found simply by increasing the sample size

When a question has not been correctly defined, repeating the same set of questions with more people does not necessarily bring the investigation closer to the facts.

If every interviewee is asked only whether customers are satisfied, the researcher may collect a large volume of satisfaction data without ever examining whether the customer relationship depends on the founder personally.

If every expert is asked to forecast market size, the investigation may produce a neat set of growth projections without anyone asking what observable traces those projections should leave in raw-material demand, capacity, orders or end-market consumption.

This does not mean that sample size is unimportant.

The number of interviews becomes stronger evidence only when the questions are continually refined and the structure of the source base is expanded.

What this work requires is not simply “a large number of interviews,” but multiple rounds of in-depth interviews across different roles, with each round informed by what emerged from the last.

The first round establishes an initial understanding.
Contradictions determine which sources should be approached next.
New sources are used to test earlier leads.
The results determine whether the direction of the research should change.

Each round is not another attempt to collect the same answer.

It is a step toward narrowing down the problem that actually needs to be resolved.

The researcher’s value also lies in discovering the question

Institutional investors do not need an open-ended exploration of an industry.

Within a defined project period, they need to understand which facts support the current investment thesis, which assumptions still lack independent evidence, and whether a potentially decision-relevant issue remains outside the existing research framework.

This requires the researcher to avoid being fully constrained by the initial interview guide, while also avoiding an unlimited expansion of scope whenever a new lead appears.

Is the anomaly real?
Is it representative?
Is it relevant to the investment decision?
Is it worth using limited project time to pursue?
What type of evidence should be used to test it?

These judgments must be made throughout the investigation.

Independent primary research is therefore not only a service for finding answers to the questions a client has already posed.

At times, its more important role is helping the client recognize that the question that truly needs to be answered is a different one.

A questionnaire can collect information about risks that have already been framed as questions.

One of the values of in-depth interviews is discovering risks that have not yet been framed as questions but may still affect the investment decision.

For FUWA LAB, the interview guide is the starting point of an investigation, not the boundary of where the investigation can go.

The hardest risks to identify are not questions for which no answer exists. They are the questions we realize need to be asked only after the conversation has truly begun.