Why Management Education Must Teach Students to Ask Better Questions

From the Faculty's Desk

A student presents a polished recommendation: reduce prices to improve sales. The analysis is organised, the slides are confident and the calculation is correct. Then a simple question changes the discussion: why have sales declined? If the problem concerns unreliable delivery, a price reduction may weaken margins without restoring demand. An excellent answer to the wrong question remains a poor managerial decision.

Management education must therefore develop more than the ability to apply a framework. It must cultivate the ability to decide which problem deserves analysis, what evidence could illuminate it and where an apparently sensible interpretation might fail. Asking better questions is the discipline that connects knowledge with judgment.

Questions define the problem before analysis begins

A business problem rarely arrives in a neutral form. A complaint that employees resist change already implies an explanation. A request to improve advertising assumes that communication is the constraint. A proposal to automate a process may bypass the question of whether the process should exist in its current form.

Problem framing determines which variables receive attention, which stakeholders are heard and which solutions become imaginable. The person who defines the question therefore influences the answer before any formal analysis takes place. Students need to recognise this influence, including when the framing comes from a senior manager or a persuasive case narrator.

A useful classroom habit is to request several plausible formulations of the same situation. Declining profitability could concern pricing power, cost structure, product mix, capacity utilisation or the quality of reported revenue. These are connected possibilities, but they imply different evidence and interventions. Keeping alternatives open prevents premature commitment to the most familiar explanation.

Move from description to explanation

What happened, why it happened and what should be done are distinct questions. Description establishes a pattern. Explanation proposes a mechanism. Prescription adds a judgment about objectives, alternatives and consequences. Moving quickly between these levels can create arguments that sound complete while leaving their most important links unsupported.

Suppose a hypothetical company observes higher sales after introducing a training programme. The sequence is compatible with a beneficial training effect. It is also compatible with seasonal demand, a price change or managers selecting stronger employees for training. Recognising these alternatives does not require cynicism. It requires separating an observation from a causal claim.

Students should learn to ask what would have happened without the intervention and which comparison makes that counterfactual credible. Not every business decision permits a controlled experiment. Even then, the counterfactual question improves interpretation by exposing the assumptions on which a recommendation depends.

The purpose is proportional confidence: conclusions should be as strong as the evidence permits, and no stronger.

Better questions require subject knowledge

An emphasis on inquiry must not become an argument against expertise. A student cannot interrogate a cash-flow statement effectively without understanding accruals, working capital and financing. Nor can a student evaluate an algorithm without some understanding of data, prediction and error.

Knowledge supplies the distinctions that make questions precise. It enables a finance student to ask whether cash improved because collections strengthened or supplier payments were deferred. It enables an HR student to distinguish employee dissatisfaction from a poorly designed incentive. General curiosity becomes professionally useful when connected to disciplinary understanding.

The educational task is therefore reciprocal. Concepts help students ask better questions; questions reveal when concepts are useful or insufficient. A framework should function as an instrument for investigation, rather than a template applied because it appeared in the previous lecture.

Transfer across contexts cannot simply be assumed. Halpern’s model combines willingness to engage in effortful thinking, explicit skills, attention to problem structure and monitoring of one’s reasoning (Halpern, 1998). This suggests asking students to recognise the same reasoning problem in different business settings, rather than expecting exposure to one case to generalise automatically.

This also explains the importance of faculty expertise. A skilled teacher can recognise the consequential issue within an imprecise student question and help the student reformulate it without prematurely supplying the answer.

Make discussion an intellectual method

Harvard Business School’s guidance on case discussion treats questioning, listening and responding as connected teaching skills. It encourages instructors to plan questions, use follow-up probes and attend to what participants leave unsaid (HBS Christensen Center, n.d.). The value lies in the structure of inquiry, rather than the mere presence of classroom conversation.

A discussion becomes analytical when claims are examined against evidence and alternative explanations. Asking a student to justify a recommendation is a beginning. Asking what evidence would reverse it is more demanding. Asking another student to identify the strongest part of the argument can prevent disagreement from becoming a contest of confidence.

Participation must also be distinguished from frequency of speech. A concise question that exposes an unsupported assumption may contribute more than several fluent restatements. Faculty can make this distinction visible through feedback that recognises reasoning, listening and constructive revision.

The goal is not to make every class inconclusive. A strong discussion should clarify what can be concluded, what remains uncertain and which further information would matter.

Give AI a demanding educational role

Generative AI can produce an answer before a student has understood why the question is difficult. This creates a challenge for assessment, but also an opportunity to make evaluation more explicit. A fluent response can become an object of inquiry: what assumptions does it make, which sources support it and what alternatives does it omit?

UNESCO’s guidance on generative AI in education and research advocates a human-centred approach and the validation of pedagogical and ethical suitability (Miao & Holmes, 2023). For management education, that supports a clear principle: the educational value of a tool depends on what the learner is required to do with it.

Students might compare an AI-generated recommendation against original sources, identify an unsupported causal claim and revise the analysis. They could also defend their choices orally or show how a conclusion changed after receiving new evidence. These activities make the reasoning visible instead of relying solely on the polish of the final submission.

AI literacy should include the ability to reject an attractive answer for a well-explained reason.

Assess the quality of inquiry

Assessment signals what an institution values. If marks reward only a neat final answer, students may conceal uncertainty and avoid difficult questions. An alternative is to assess the formulation of the problem, the relevance of evidence, the treatment of alternatives and the justification for the eventual decision.

For a project on declining customer retention, students could submit an initial question, an evidence plan and a revised question after preliminary investigation. The difference between the first and final formulation would show learning. A change of mind, supported by evidence, should count as intellectual progress rather than inconsistency.

There must still be standards. Not all interpretations are equally credible, and not every question is productive. A question should be relevant to the decision, answerable to some degree and capable of changing what a manager believes or does. Vague complexity is not a substitute for disciplined thought.

Faculty can help students recognise when additional analysis has diminishing value and when a decision must be made under remaining uncertainty.

Build a culture in which inquiry is shared

Students are more likely to ask substantive questions when they see faculty model curiosity and revision. A teacher who distinguishes established knowledge from a contested interpretation demonstrates intellectual authority through clarity about limits. That example can be more powerful than a general invitation to speak.

A faculty-led article, case note or discussion can begin with a puzzle from practice and invite students to develop competing explanations. Colleagues from different specialisations can then show how the same situation changes when viewed through finance, operations, behaviour or strategy. Such participation makes disciplinary depth visible without pretending that one field contains every answer.

The lasting outcome is a professional habit. Before accepting a dashboard, a forecast or an established routine, a graduate should ask what it represents, what it excludes and what evidence would justify acting on it. Management education succeeds when students can use knowledge to improve the question itself. Better decisions often become possible at precisely that moment.

References

1\. Harvard Business School, Christensen Center for Teaching & Learning (n.d.). Questioning & Discussions. Official case-method teaching guidance.

https\://www\.hbs.edu/teaching/case-method/leading-case-discussion/questioning-discussions

2\. Miao, F. & Holmes, W. (2023). Guidance for generative AI in education and research. UNESCO. Institutional guidance; not an experimental evaluation of the classroom proposals in this article.

https\://www\.unesco.org/en/articles/guidance-generative-ai-education-and-research

3\. Halpern, D. F. (1998). Teaching critical thinking for transfer across domains: Dispositions, skills, structure training, and metacognitive monitoring. American Psychologist, 53(4), 449–455. doi:10.1037/0003-066X.53.4.449. Author abstract indexed by PubMed.

https\://pubmed.ncbi.nlm.nih.gov/9572008/

Sources verified online: 20 October 2026\.