In our previous article, we discussed a common but easily overlooked problem: the investment target is usually clear, but the risks that truly need to be investigated may not yet have been written into the initial list of questions.

How, then, do risks that have not yet been framed as questions begin to emerge?

Often, they do not arrive as a “major revelation” powerful enough to change an investment judgment on its own. They appear in what seems like an ordinary aside during an interview. It may be a single word, a moment of hesitation, or a comment that falls outside the interview guide.

That comment may not prove anything by itself. But it can alert the researcher that the issue requiring verification is not the original question in the guide, but another question that no one had previously recognized.

An interview guide is built on what is already known

Every customized investment research engagement needs an interview guide.

Before interviews begin, researchers typically define the themes to be verified based on information provided by management, public sources, issues identified through financial and legal due diligence, and the client’s existing investment thesis.

This preparation is essential. It helps define the scope, identify and organize relevant sources, and make findings comparable across interviews.

But an interview guide also has an inherent boundary: it can reflect only the questions that were already recognized before the research began.

If subsequent interviews merely work through the prepared questions in sequence, the research will often do little more than add detail to the existing framework. A genuinely important new risk may be filtered out of the conversation simply because it “does not belong to the current question.”

The purpose of an interview guide, therefore, is to give the research a starting point—not to determine in advance where the investigation must end.

One positive comment changed the original question

In one engagement, details have been generalized to protect client confidentiality. The client asked us to assess whether a target company’s customer relationships were stable.

Under the initial research framework, the questions focused on several areas: how customers viewed the company’s products and services, whether business relationships were continuing, whether purchasing intentions had changed, and how the target compared with similar suppliers.

One customer we interviewed gave an overall positive assessment. The customer valued the product and considered the target company reasonably responsive. Under the original interview guide, this interview appeared to provide another piece of support for the view that the company’s customer relationships were relatively stable.

But while discussing day-to-day communication, the customer added in passing that important matters were still handled mainly through direct communication with the target company’s founder.

This was not a negative comment, nor was it presented by the interviewee as a risk. On the contrary, in the customer’s account, the founder’s personal involvement signaled attention and efficiency.

Yet the comment changed the research question.

We had originally asked:

Are customers satisfied?

The question that then required further investigation became:

Are customer relationships built on the company’s products, processes, and team, or do they depend primarily on the founder’s personal credibility and continued involvement?

These two questions may look similar, but their implications for an investment decision are fundamentally different.

A company’s core customers may be highly satisfied. But if key relationships are maintained mainly by the founder, customer satisfaction does not necessarily mean those relationships can be replicated by the organization, transferred to the team, or sustained as the business expands.

The value of a lead is that it allows the question to be redefined

The most important role of unexpected information in an interview is not to provide an immediate answer. It is to change the answer the research needs to seek next.

The comment that the customer “mainly communicates with the founder” cannot, by itself, prove that the target company has key-person risk. Still less can it justify a conclusion that the company lacks organizational capability. At this stage, it is only a lead.

But that lead breaks down the broad question of “customer relationship stability” into a set of more specific questions that are more directly relevant to the investment decision:

·   Who maintains the day-to-day customer relationship?

·   Who is responsible for commercial negotiations, delivery coordination, and after-sales issues?

·   When an important problem arises, does the customer have a reliable point of contact other than the founder?

·   Has the company created internal records of customer information, communications, and decision-making processes?

·   As the number of customers grows, can the current relationship-management model be replicated?

·   If the founder reduces direct involvement, can those customer relationships continue?

The original question focused on whether the current situation was favorable. The new questions begin to test what sustains that favorable situation and whether it can endure.

This is the process of moving from an observable operating condition to an examination of organizational capability and key-person risk.

Moving beyond the interview guide is not the same as following intuition without discipline

Moving beyond the interview guide does not mean pursuing every unusual statement without limit. Nor does it mean allowing the researcher’s personal interests to replace the objectives of the engagement.

A valuable departure from the guide should be a structured adjustment. We generally take four steps:

1. Identify what changed

First, determine whether the new information changes one of the assumptions embedded in the original question.

In the case above, what changed was not the customer’s level of satisfaction. What changed was the possibility that a relationship structure not previously included in the analysis might sit behind that satisfaction.

2. Rewrite the question

Turn a vague sense that “something does not fit” into a question that can be investigated.

The question should not remain at the level of “Is the company too dependent on its founder?” It should be broken down further: Which relationships are maintained by the founder? Can other team members take them over? Have the relevant processes been institutionalized? Could this dependence constrain expansion?

3. Expand the source base

Once the question has been redefined, the original list of interviewees often needs to change as well.

A customer can explain whom they communicate with, but may not understand the target company’s internal division of responsibilities. A former employee may understand customer handover practices, but may not know how customers perceive those arrangements. Customer contacts at different levels may also see entirely different relationship structures.

The new question must therefore be matched with new sources. It cannot be verified simply by asking the original interviewees the same question repeatedly.

4. Return to the investment thesis

Finally, even if the lead receives further support, the researcher must determine which investment judgment it would affect.

If the investment case depends on a team-based sales model, expansion across regions, or the founder gradually stepping back from frontline operations, the ability of customer relationships to function independently of one individual may become a material issue. Conversely, if the company’s scale, governance arrangements, and the investment thesis do not depend on such replicability, the same fact may carry a different level of significance.

Research is not about adding questions indefinitely. New questions must ultimately be brought back to the judgments that matter to the transaction.

A new question requires a new verification path

In this engagement, the next stage of work was no longer limited to asking additional customers whether they were satisfied. We reorganized the verification process around the newly defined question:

The research path therefore expanded from a static assessment into a sequence of connected questions:

Customer satisfaction → Who maintains the customer relationship → Whether the organization can take over the relationship → Whether the business model can be replicated → Whether the growth thesis holds

The point is not that the research must ultimately conclude that a risk exists. The point is to prevent a seemingly positive fact from obscuring the structural conditions behind it.

As discussed in our previous article, the credibility of a lead and the level of certainty in the conclusion must remain proportionate. The additional point here is that the primary value of a new lead is often not that it answers the original question, but that it forces us to ask a more accurate one.

In-depth interviews produce more than a set of answers

Standardized information collection is well suited to questions with clear boundaries and comparable answers. It can help researchers confirm quantities, frequency, preferences, and trends, while improving the efficiency of processing information at scale.

In customized investment research, however, the research target, transaction context, and investment thesis differ from one engagement to another. Researchers are not always working with a set of questions that has already been fully defined. Sometimes the most important output of the research is not another answer, but the realization that the original question was too broad, directed at the wrong person, or missing an assumption that determines the nature of the risk.

The difference between in-depth interviews and standardized information collection is therefore not simply that the answers are longer or contain more information. More importantly, interviews allow the research path to change when new information appears: the original questions can be broken down, interviewees can be reselected, and the order of verification can be adjusted.

This is not a departure from the research framework. It is a timely correction of that framework when the facts begin to change the question.

The real investigation begins when the question is redefined

A well-developed interview guide can ensure that the prepared questions are covered. It cannot guarantee that every important question has already appeared.

A comment outside the guide may ultimately prove not to indicate a risk. After cross-checking, it may even have a reasonable explanation. But if the researcher fails to recognize that the comment changed the original question, no subsequent verification path will be built.

For FUWA LAB, Independent Primary Research does not mean amplifying every piece of information that falls outside the interview guide. It means identifying which information changes an existing assumption and turning it into a new question that can be verified.

Because in investment research, meaningful progress does not always come from finally obtaining an answer.

Sometimes it comes from finally recognizing that a different question should have been asked.