Entrepreneurial Opportunity, Problem and Customer

Review questions

Integrated summary: From Idea to a Learnable Opportunity

This subject begins with a simple but important idea: early entrepreneurship is not a race to build a product; it is structured learning under uncertainty. An idea proposes something that might be offered. A problem describes a difficulty or undesirable outcome in context. An opportunity is a bundle of hypotheses connecting an important problem with a reachable segment and value that may be created in a feasible and potentially sustainable way. FlexSlot therefore does not become an opportunity merely because a scheduling board sounds attractive. We first need to learn who experiences rescheduling problems, when they appear, what the current alternative is, and what makes change important.

Value creation means a change that matters to a person or organization, not simply the addition of features. The value might be less coordination time, fewer errors, an easier experience, or lower risk. A feature such as an automatic reminder has no value in isolation; its value depends on an outcome for the user or customer. To organize uncertainty, we can use practical lenses such as problem importance, feasibility, economic logic, and constraints. These are thinking aids rather than universal laws. Their purpose is to identify the next learning question instead of collapsing every unknown into a vague judgment such as “the idea is excellent.”

We then need a problem statement that supports learning. A useful frame identifies who experiences the situation, the context, the observed behavior or outcome, the current alternative, and what remains unknown. “All tutors need an app” is weak because it generalizes and assumes the solution. A more precise statement might say that some independent tutors with frequent changes update messages and a separate calendar after rescheduling, sometimes creating repeated work or inconsistent details, while the size of that impact remains unknown. The initial segment is therefore a working hypothesis related to the problem, not a final declaration of the market.

A role map prevents us from asking the wrong person about a decision they do not control. A user interacts with the service. A customer or purchase decision-maker makes the contracting decision in the situation being studied. A payer bears the payment if that role is separate. A stakeholder can influence or be affected by the decision. One person may perform several roles. These are practical distinctions for the case, not universal legal definitions. We must also separate observation from interpretation: “someone forgot to update an appointment” is an observation; “too many tools caused it” is an interpretation that needs additional evidence.

Customer discovery examines these frames through direct conversations with potential customers, users, partners, and stakeholders as appropriate to the question. It is usually better to ask about a real past situation: “Tell me about the last time the appointment changed; what did you do step by step?” rather than “Would you use our idea?” Leading or compound questions can shape an answer, while questions about behavior and current alternatives reveal details that can be followed up. NSF I-Corps uses intensive customer discovery, and program materials for accepted teams state a 100-interview requirement. That is a program-specific requirement, not a scientific threshold that proves a market or invalidates learning below it.

When analyzing interviews, record evidence before conclusions. If the problem appears in a case with frequent changes and self-managed coordination but is weaker where a schedule is stable or an assistant handles coordination, we can record an emerging pattern: pain may be associated with those conditions. We cannot turn three or five interviews into a market percentage. Interviews also cannot answer every claim. A usage claim may require a prototype or behavioral trial, while a commitment claim may need behavior closer to a real decision. The appropriate evidence type follows the question we are trying to answer.

An assumption becomes more useful when it is converted into a hypothesis that can weaken. Specify the segment, context, expected behavior, test, a predefined decision threshold, and what would cause reconsideration. A decision threshold is a justified stage-specific rule, not a universal scientific truth, and it is better defined before the result is seen so it is not moved simply to make a test appear successful. Prioritize assumptions by impact if wrong and current uncertainty: an assumption that could break the venture logic and still has weak evidence deserves earlier learning than an easy but low-impact detail.

Finally, screen the opportunity only as far as the evidence allows. If six of eight people from the founder’s network say the problem sounds important, that is a result for those eight cases, not “75% of the market.” Ask what became stronger, what weakened, what remains unknown, and what the smallest useful next test is. Lean Startup and entrepreneurial experimentation can help organize the hypothesis-test-learn-decide cycle, but they are tools with documented limits rather than guarantees of success. A sound decision at this stage may be to continue, modify the segment or problem, or pause. Its quality comes from a clear reason and proportional evidence, not from defending the first idea.

Review questions

Test your understanding with these questions. Marking and saved results are available with a free account.

Question 1Which statement is primarily a solution idea rather than a problem or evidence? Changing an appointment sometimes requires repeated updates to messages and calendars. A unified board that sends automatic reminders when an appointment changes. Three tutors described a recent case in which they resent details. We do not yet know how frequent the problem is among tutors.
Question 2Five people say the FlexSlot idea sounds nice. What is the most accurate conclusion now? The opportunity is confirmed because every participant liked it. The problem is proven because five people approved the solution. We have limited positive reaction and still need evidence about the problem, behavior, and willingness to change. The market size can be estimated from the approval rate in this group.
Question 3The team is unsure whether rescheduling is an important problem at all. What is the best step before building a full technical integration? Run a low-commitment learning step about rescheduling situations and current alternatives. Build all features so people can judge the final solution. Finalize the brand name because identity must come before testing. Set a precise annual revenue forecast before speaking with customers.
Question 4Which question examines value rather than focusing on a feature? Can an automatic reminder be programmed? Should the interface use one color? Is the name shorter than competitors’ names? What meaningful change would occur for the user if the solution works, and how would we notice it?
Question 5Which problem statement is strongest for initial learning? Everyone wants easier scheduling. Some independent tutors with frequent changes update messages and calendars separately, which may create repetition or inconsistencies; we do not yet know the size of the impact. We need a better app than the ones already on the market. The problem is that users do not understand the value of our idea.
Question 6The team defines its segment as “all adults who have appointments.” What is the most useful improvement for problem discovery? Add problem-related conditions such as frequent rescheduling and personally managing coordination. Segment people by their preferred interface color. Keep the segment broad so no potential customer is lost. Choose one city at random and use location alone as the segment.
Question 7An organization buys a tool, employees use it daily, and an authorized role signs the purchase order. Who is best placed to answer, “What criteria would block or allow the purchase?” Any end user because they see the interface every day. An outsider who is not affected by the decision. The purchase decision-maker or the role managing the actual contracting criteria. The founder because they know the product better than anyone.
Question 8A tutor says, “I forgot to update an appointment last week.” Which note separates observation from interpretation? Observation: they forgot to update an appointment. Interpretation: too many tools may be a cause and needs verification. Observation: too many tools are definitely the cause. Interpretation: they forgot the appointment. Observation: they need the new app. Interpretation: they will certainly buy it. Observation: the market has the problem. Interpretation: one case confirms it.
Question 9Lian wants to open a discovery interview. Which question gives the best window into actual behavior? Do you like the idea of an app that sends reminders? Tell me about the last time an appointment changed after confirmation. What did you do step by step? Isn’t using more than one tool frustrating? If we lowered the price, would you use the product every day?
Question 10What is the main problem with the question, “Isn’t it frustrating to use more than one tool?” It is too short to allow a long answer. It is leading and suggests that frustration is the correct feeling. It does not mention the proposed product name. It does not ask the respondent’s age or background.
Question 11The team reads that NSF I-Corps requires a large number of discovery interviews for accepted teams and states 100 interviews. What is the correct use of that information? Treat 100 as the scientific threshold that proves any market is valid. Conduct 100 random interviews without defining a decision or segment. Reject any learning from fewer than 100 interviews. Understand it as a program-specific requirement and an example of intensive discovery, not a universal magic number.
Question 12In three interviews, one case has frequent changes and clear pain, one uses an assistant and has low pain, and one has a stable schedule. What is the best conclusion? Only one-third of the market has the problem. The problem does not exist because two cases showed little pain. There is a signal that pain may relate to frequent changes and self-managed coordination, and it should be examined in additional cases. The product should be built immediately because one case is strong.
Question 13Which statement is closest to a testable hypothesis? People hate rescheduling. Among tutors with frequent changes, we expect a defined behavior; we will test it with a specified method and predefine what result would strengthen or weaken the hypothesis. Our idea is great and we will find customers. The market must be large because everyone uses calendars.
Question 14The team has four assumptions. Which should usually be tested first using the lenses of “impact if wrong” and “uncertainty”? An assumption with high impact if wrong and weak supporting evidence. A low-impact assumption with strong evidence. A cosmetic detail that is easy to test but does not change the venture decision. An assumption already settled by contract or proven capability and not central to opportunity logic.
Question 15A founder interviews eight people from their network; six say the service sounds useful, and no behavior is tested. What is the most accurate description? The sample proves that 75% of the market will pay. The idea has failed because two people did not agree. We have limited positive opinion in a convenience sample and need a test closer to behavior or commitment before generalizing. The percentage is enough to build a precise revenue forecast.
Question 16How should Lean Startup and entrepreneurial experimentation be used in this diploma? As a law that guarantees success if the steps are followed in order. As a replacement for judgment because experiments always give a clear answer. As a tool that requires every venture to build an app prototype first. As tools for organizing hypotheses, tests, and learning while recognizing their limits and matching the test to context.