Problem Definition and Research Questions

Introduction

A brand management project becomes academically valuable only when it investigates a clearly defined problem. A broad concern such as “the brand is losing customers” is not yet a research problem. It is an organisational symptom that requires diagnosis. The researcher must determine what is happening, why it may be happening, whose behaviour or perceptions are involved, what is already known, and what evidence is needed to support a defensible decision.

Problem definition is therefore the bridge between managerial concern and systematic research. It converts an uncertain situation into a bounded investigation with a specified context, unit of analysis, concepts, objectives, research questions, and—where appropriate—testable hypotheses. Poor problem definition produces unfocused data collection, inappropriate samples, weak analysis, and recommendations that cannot be implemented. Strong problem definition improves both academic credibility and managerial usefulness.

In brand management, this process is especially important because brands operate simultaneously as market offerings, organisational assets, symbolic meanings, customer experiences, and social identities. A decline in sales might reflect price competition, reduced distribution, declining perceived quality, weak communication, poor service, changing consumer values, or inaccurate internal assumptions. Research questions must distinguish among these possibilities.

Learning Objectives

By the end of this chapter, you should be able to:

• distinguish a managerial problem from a research problem;
• diagnose organisational and market contexts systematically;
• identify a knowledge gap and justify its importance;
• define the unit and level of analysis;
• formulate a clear research aim and measurable objectives;
• develop focused research questions and hypotheses;
• operationalise abstract brand constructs;
• evaluate the quality, feasibility, and ethics of a proposed problem definition; and
• translate research questions into useful brand management decisions.

From Managerial Concern to Researchable Problem

A managerial problem is a situation requiring a decision. Examples include whether to reposition a brand, invest in influencer marketing, redesign packaging, enter a new segment, or respond to declining loyalty. A research problem is a gap in understanding that prevents an informed decision. It states what is not sufficiently known and what must be investigated.

The distinction can be expressed as follows:

A useful problem statement normally identifies five elements:

1. Context: the organisation, market, category, location, or segment.
2. Observed condition: the performance issue, opportunity, or uncertainty.
3. Knowledge gap: what is not adequately known or explained.
4. Consequences: why the gap matters for decisions or theory.
5. Research boundary: the population, period, and constructs to be studied.

For example: “Although Brand A has increased digital advertising expenditure, consideration among urban university students has remained low. It is not known whether this reflects weak perceived authenticity, low message relevance, or limited category involvement. This study will examine the relationships among these constructs and brand consideration during the 2025 campaign period.”

This statement is stronger than “This study examines Brand A’s marketing problems” because it identifies a population, context, competing explanations, and outcome.

Diagnosing the Organisational and Market Context

Research questions should not be developed in isolation from the environment in which the brand operates. Context diagnosis establishes what is already known, what decision is pending, and which explanations are plausible.

Organisational diagnosis

The organisational diagnosis may examine:

• the brand’s mission, promise, values, and positioning;
• portfolio structure and relationships among sub-brands;
• target segments and customer value propositions;
• recent changes in product, price, promotion, distribution, or service;
• performance indicators such as awareness, consideration, conversion, retention, and advocacy;
• internal capabilities, budgets, data systems, and decision authority; and
• competing interpretations among managers, sales staff, customer-service teams, and customers.

Internal documents, brand guidelines, campaign reports, customer complaints, sales records, interviews, and stakeholder workshops can support this diagnosis. Internal data should be treated critically: performance figures may be incomplete, definitions may differ across departments, and organisational narratives may protect established decisions.

Market diagnosis

The market diagnosis considers:

• category growth or decline;
• competitor positioning and communication;
• customer needs, motivations, and barriers;
• cultural and demographic changes;
• regulatory and technological developments;
• channel and platform changes;
• price sensitivity and economic conditions; and
• emerging expectations concerning sustainability, inclusion, privacy, and transparency.

Useful tools include competitor mapping, customer journey analysis, environmental scanning, social listening, category reports, review analysis, and exploratory interviews. These tools do not automatically establish causality. They help generate and refine possible explanations.

The symptom–cause distinction

A symptom is an observable condition; a cause is a mechanism that may produce it. “Lower repeat purchase” is a symptom. Potential causes include dissatisfaction, switching incentives, stock availability, changes in household circumstances, or stronger competitor preference. A research project should avoid assuming that the first explanation offered by management is correct.

A diagnostic tree can help separate outcomes from explanations:

Figure: A diagnostic tree separates an observed brand symptom from competing explanations.

Literature Review and Knowledge Gaps

A knowledge gap is not simply a topic that has not been studied by the researcher. It is a meaningful limitation in existing knowledge. Common forms include:

• Descriptive gap: insufficient evidence about the characteristics or prevalence of a phenomenon.
• Explanatory gap: inadequate understanding of why an outcome occurs.
• Contextual gap: evidence exists, but not in the relevant country, category, segment, or channel.
• Conceptual gap: important constructs are defined inconsistently or relationships are poorly specified.
• Methodological gap: previous research uses methods that cannot address the present question adequately.
• Practical gap: theory exists, but managers lack evidence for a current decision.

A literature review should compare sources rather than list them. The researcher should identify definitions, theories, populations, methods, findings, contradictions, and limitations. Academic journal articles are central, but industry reports, regulatory documents, company data, and credible market studies may be relevant to context. Search terms should combine the brand issue with constructs and outcomes, such as “brand authenticity,” “purchase intention,” “service experience,” or “customer-based brand equity.”

A useful evidence matrix records the author, year, context, sample, method, constructs, principal findings, limitations, and relevance to the proposed project. This matrix makes it easier to distinguish an actual gap from a claim based on limited reading.

Unit and Level of Analysis

The unit of analysis is the entity about which conclusions are drawn. The level of analysis refers to the conceptual layer at which the phenomenon is examined. These terms are related but not identical.

Possible units in brand management include:

• individual consumers;
• households;
• customer accounts;
• retail outlets;
• social-media posts;
• campaigns;
• brands;
• firms; and
• market categories.

For example, a study may collect survey responses from individuals but analyse brands as cases. Alternatively, it may analyse individual perceptions nested within retail stores. Confusing these levels creates an ecological fallacy—drawing conclusions about individuals from aggregate data—or an atomistic fallacy—drawing conclusions about groups from individual observations.

The unit should match the research question. “How do consumers perceive Brand X?” suggests individual consumers. “How do competing brands differ in positioning?” suggests brands or brand cases. “Which campaign characteristics predict engagement?” may use campaigns or posts as units, depending on the design.

The sampling frame must also correspond to the unit. If the target population is current customers, a sample of social-media followers may exclude customers who do not use that platform. If the study concerns lapsed customers, surveying only active loyalty-program members is inappropriate.

Research Aim and Objectives

The research aim is a concise statement of the overall purpose. It should normally contain an action verb, the central phenomenon, the population or context, and the intended contribution.

Example: “To examine how perceived authenticity and message relevance influence brand consideration among young adult consumers of sustainable clothing in London.”

Objectives divide the aim into achievable tasks. Strong objectives are specific, observable, and feasible. They may involve reviewing theory, measuring constructs, comparing groups, testing relationships, or developing recommendations.

Example objectives:

1. To review literature on perceived authenticity, message relevance, and brand consideration.
2. To measure these constructs among young adult sustainable-clothing consumers.
3. To examine whether perceived authenticity is associated with brand consideration.
4. To compare perceptions between frequent and occasional category buyers.
5. To develop evidence-based recommendations for brand communication.

Objectives should not promise more than the design can support. A cross-sectional survey may identify associations, but it generally cannot prove that a communication exposure caused a later purchase. Verbs such as “explore,” “describe,” “compare,” and “examine associations” are often more defensible than “prove” or “demonstrate causation.”

Developing Focused Research Questions

Research questions are the precise questions the project will answer. A coherent sequence often includes one central question and several subordinate questions.

A central question might be: “How do perceived authenticity and message relevance relate to brand consideration among young adult sustainable-clothing consumers?”

Subquestions could ask:

• How do consumers describe authenticity in this category?
• What levels of authenticity, relevance, and consideration are reported?
• Is perceived authenticity associated with brand consideration?
• Does message relevance differ between frequent and occasional buyers?
• What communication improvements do consumers perceive as credible?

Good research questions are clear, focused, answerable, empirically investigable, and aligned with the available time and resources. They should specify the population and context where these affect interpretation.

Avoid questions that are too broad (“What is the future of branding?”), contain an assumed answer (“Why is the ineffective campaign failing?”), or combine unrelated projects (“How do price, culture, technology, psychology, distribution, and leadership affect the brand?”). A question should also distinguish normative and empirical forms. “Should the brand reposition?” is a decision question; “How do target customers evaluate the current positioning?” is an empirical research question that can inform the decision.

Qualitative, quantitative, and mixed-method questions

Qualitative questions commonly use “how,” “what,” and “in what ways” to explore meanings, experiences, and processes. Quantitative questions may ask about prevalence, differences, associations, prediction, or effects. Mixed-method projects combine these purposes deliberately rather than merely collecting both kinds of data.

The question determines the evidence. Interviews are suitable for exploring how customers interpret authenticity. Surveys can estimate perceptions and examine relationships. Experiments may test whether alternative brand messages produce different responses under controlled conditions. Digital analytics can describe behavioural patterns, but clicks do not necessarily represent preference, trust, or purchase.

Constructs, Variables, and Operationalisation

A construct is an abstract concept, such as brand trust, perceived quality, authenticity, loyalty, or involvement. A variable is an observable or recorded representation of a construct that can vary across cases. Operationalisation specifies how the construct will be measured or identified in the study.

An operational definition should identify the target population, measurement approach, timeframe, and interpretation. A single item may be practical, but multi-item measures often capture multidimensional constructs more reliably. Researchers should use established scales where appropriate, check their fit with the context, and report any adaptation. Translation, cultural interpretation, and response-style differences can affect validity.

Measurement quality includes:

• Reliability: consistency of a measure.
• Content validity: adequate coverage of the construct’s relevant dimensions.
• Construct validity: evidence that the measure represents the intended concept rather than a different one.
• Criterion-related validity: relationship with a relevant external outcome.

High reliability does not guarantee validity. Respondents may answer consistently to a poorly designed question.

Hypotheses and Conceptual Models

A hypothesis is a testable proposition about an expected relationship or difference. It should identify constructs, the expected direction where justified, and the population or context.

Example: “Among young adult sustainable-clothing consumers, higher perceived authenticity is positively associated with brand consideration.”

A null hypothesis may state that there is no association in the population, while a research hypothesis states the expected relationship. Hypotheses should arise from theory or prior evidence, not from arbitrary prediction after seeing the data.

A simple conceptual model may propose that perceived authenticity and message relevance are predictors of brand consideration, while category involvement influences the strength of the relationship. This model does not prove causality; it represents a theoretically reasoned structure to be examined.

The researcher should distinguish:

• Independent or predictor variable: proposed explanatory factor.
• Dependent or outcome variable: phenomenon to be explained.
• Mediator: mechanism through which a predictor may influence an outcome.
• Moderator: condition that changes the strength or direction of a relationship.
• Control variable: factor included to reduce alternative explanations.

Claims should match design. A longitudinal study can establish temporal order more effectively than a cross-sectional study. A randomised experiment provides stronger evidence for causal effects than an observational survey, provided manipulation and control are successful.

Worked Application: Illustrative Case of Northstar Coffee

Background

Northstar Coffee is a fictional regional coffee chain with a strong reputation for ethical sourcing. It has introduced mobile ordering and increased social-media activity, yet repeat visits among customers aged 18–30 have declined. Managers believe that competitors appear more contemporary and socially responsible.

Problem

The observed problem is declining repeat visits. Management initially frames the issue as a communication failure, but this assumption has not been tested. Possible explanations include perceived inauthentic sustainability claims, inconvenient mobile ordering, weak service experience, higher prices, or stronger competitor attachment.

Analysis

Exploratory interviews with customers, review analysis, and internal transaction data suggest three issues: customers value ethical sourcing but doubt whether claims are transparent; mobile-order collection is sometimes slow; and competitors offer more personalised rewards. The literature indicates that authenticity and service convenience may influence trust and repeat intention, but evidence for regional coffee chains in this age group is limited.

The unit of analysis is the individual customer. The level of analysis is customer perception and intention, with transaction records used as supplementary behavioural evidence. The project excludes supply-chain auditing and does not attempt to measure the entire coffee category.

Approach and research questions

Aim: “To examine how perceived sustainability authenticity, service convenience, and reward-value perceptions relate to repeat-purchase intention among Northstar Coffee customers aged 18–30.”

Research questions:

1. How do customers interpret Northstar’s sustainability claims?
2. What levels of perceived authenticity, convenience, reward value, and repeat-purchase intention are reported?
3. How are these perceptions related to repeat-purchase intention?
4. Do relationships differ between frequent and infrequent customers?

Possible hypotheses include:

• Higher perceived sustainability authenticity is positively associated with repeat-purchase intention.
• Higher perceived service convenience is positively associated with repeat-purchase intention.
• The relationship between authenticity and repeat intention is stronger among customers who consider ethical sourcing important.

Outcome and learning points

The project would combine interviews for meaning and language with a survey for structured comparison, while transaction data could provide a behavioural check. Recommendations might concern clearer evidence for sustainability claims, improved collection processes, and more relevant rewards. The case demonstrates that a visible performance problem should not be prematurely reduced to a single marketing explanation.

Feasibility, Ethics, and Governance

A well-defined question must be feasible. Consider access to participants, data quality, time, budget, analytical competence, and organisational permission. A question that requires confidential customer-level transactions may be unsuitable if the researcher cannot obtain lawful access.

Ethical research requires informed consent, voluntary participation, confidentiality, secure data handling, and honest reporting. Brand-funded research creates potential conflicts of interest. Sponsors should not control findings, suppress inconvenient results, or require participants to provide identifiable opinions without justification. Researchers must avoid deceptive claims, manipulative recruitment, discriminatory segmentation, and the misuse of vulnerable consumers’ data.

Privacy legislation and institutional policies may regulate personal data, profiling, cookies, location information, and social-media content. Public availability does not automatically mean that content is ethically risk-free to collect or publish. AI tools may assist with brainstorming, transcription, or coding, but researchers remain responsible for accuracy, confidentiality, bias, copyright, and disclosure of use. Sensitive organisational or participant information should not be entered into unsecured generative systems.

Common Errors and Best Practices

Frequent errors include beginning with a method rather than a problem, asking too many questions, using undefined terms such as “engagement,” treating social-media metrics as direct evidence of loyalty, confusing correlation with causation, and selecting a sample for convenience without acknowledging its limitations.

Best practice involves:

1. starting with the decision context but maintaining research independence;
2. separating symptoms, assumptions, and explanations;
3. reviewing diverse and credible evidence;
4. defining the unit and level of analysis explicitly;
5. limiting the project to a manageable population, period, and set of constructs;
6. aligning aim, objectives, questions, hypotheses, measures, and analysis;
7. piloting instruments and checking whether respondents interpret items as intended;
8. documenting exclusions and limitations; and
9. ensuring that recommendations follow from evidence rather than from predetermined preferences.

Chapter Summary

Problem definition transforms a managerial concern into a focused and researchable investigation. It requires diagnosis of organisational and market contexts, separation of symptoms from possible causes, and a defensible account of the knowledge gap. The researcher must define the unit and level of analysis so that the population, sample, data, and conclusions are aligned.

A clear research aim expresses the overall purpose, while objectives divide that purpose into achievable tasks. Research questions should be focused, answerable, and appropriate to qualitative, quantitative, or mixed-method evidence. Hypotheses are testable propositions derived from theory or prior evidence, but causal language must reflect the strength of the research design. Abstract brand constructs require careful operationalisation, and reliable measurement is not necessarily valid measurement. Finally, feasibility, ethics, privacy, sponsorship, and responsible technology use are integral to problem definition rather than afterthoughts.

Key Terms

Problem definition: The process of specifying a researchable issue, context, gap, and boundaries.

Managerial problem: A situation requiring an organisational decision.

Research problem: A gap in knowledge that prevents informed understanding or action.

Knowledge gap: A meaningful limitation, absence, inconsistency, or weakness in existing evidence.

Unit of analysis: The entity about which conclusions are drawn.

Level of analysis: The conceptual layer at which a phenomenon is examined.

Research aim: A concise statement of a study’s overall purpose.

Research objective: A specific, achievable task supporting the research aim.

Research question: A focused question that a study is designed to answer.

Hypothesis: A testable proposition about an expected relationship or difference.

Construct: An abstract concept such as trust, authenticity, or loyalty.

Variable: An observable representation that can vary across cases.

Operationalisation: Specification of how a construct will be identified or measured.

Reliability: The consistency of a measurement process.

Validity: The extent to which evidence supports the intended interpretation of a measure.

Mediator: A mechanism through which one variable may influence another.

Moderator: A condition that changes the strength or direction of a relationship.

Ecological fallacy: Incorrectly inferring individual characteristics from aggregate data.

Feasibility: The practicality of completing a proposed study with available resources.

Research ethics: Principles governing responsible treatment of participants, data, and findings.

Review Questions

5 Short Answer Questions

1. What is the difference between a managerial problem and a research problem?
2. Identify three elements that should appear in a problem statement.
3. What is meant by the unit of analysis?
4. Why should researchers distinguish a symptom from a cause?
5. Define operationalisation and give one brand-management example.

5 Long Answer Questions

1. Explain how organisational and market diagnosis contributes to effective problem definition.
2. Discuss the different types of knowledge gap that may justify a brand management project.
3. Explain how research aims, objectives, research questions, and hypotheses should be aligned.
4. Compare the suitability of qualitative, quantitative, and mixed-method questions for brand research.
5. Discuss why reliability, validity, feasibility, and ethics must be considered before data collection.

3 Analytical / Application Questions

1. A retailer reports declining loyalty and assumes that its advertising is ineffective. Develop a problem statement, three research questions, and two possible hypotheses without accepting the assumption uncritically.
2. Design a conceptual model for examining how brand trust may influence purchase intention. Identify a possible mediator, moderator, unit of analysis, and suitable evidence.
3. Using the Northstar Coffee case, evaluate whether a cross-sectional survey alone would be sufficient to support a causal claim about declining repeat visits. Justify your answer and propose a stronger design if necessary.

Multiple Choice Questions with Answers

Question 1

Which statement best distinguishes a research problem from a managerial problem?

A. A research problem is always numerical, whereas a managerial problem is always qualitative.

B. A research problem identifies a gap in knowledge that affects an organisational decision.

C. A research problem is a solution selected by senior management.

D. A research problem is limited to customer complaints.

Correct Answer: B. A research problem identifies a gap in knowledge that affects an organisational decision.

Question 2

What is the unit of analysis in a study asking how individual customers perceive a brand?

A. The market category

B. The advertising campaign

C. The individual customer

D. The organisation’s annual report

Correct Answer: C. The individual customer

Question 3

Which is the strongest example of a focused research question?

A. What is the future of branding?

B. Why is modern marketing difficult?

C. How do perceived authenticity and convenience relate to repeat intention among the brand’s young adult customers?

D. Should every company use social media?

Correct Answer: C. How do perceived authenticity and convenience relate to repeat intention among the brand’s young adult customers?

Question 4

Which term refers to an abstract concept such as brand trust or authenticity?

A. Construct

B. Sampling frame

C. Unit of analysis

D. Transaction record

Correct Answer: A. Construct

Question 5

Which statement about reliability is correct?

A. Reliability guarantees validity.

B. Reliability concerns the consistency of a measurement process.

C. Reliability proves that one variable causes another.

D. Reliability eliminates sampling bias.

Correct Answer: B. Reliability concerns the consistency of a measurement process.

Question 6

A variable that changes the strength or direction of a relationship is called a:

A. Mediator

B. Moderator

C. Dependent sample

D. Research objective

Correct Answer: B. Moderator

Question 7

Which conclusion is generally most appropriate for a cross-sectional survey?

A. It proves that advertising caused later purchases.

B. It establishes temporal order with certainty.

C. It can examine associations measured at one point in time.

D. It eliminates all alternative explanations.

Correct Answer: C. It can examine associations measured at one point in time.

Question 8

Which is an example of a contextual knowledge gap?

A. A construct has no possible definition.

B. A theory has been tested in large firms but not in the relevant regional category.

C. The researcher has not yet selected a font for the questionnaire.

D. A manager prefers one competitor’s logo.

Correct Answer: B. A theory has been tested in large firms but not in the relevant regional category.

Question 9

Which practice best protects research integrity in sponsor-funded brand research?

A. Allowing the sponsor to remove unfavourable findings.

B. Reporting methods and findings honestly, including inconvenient results.

C. Using only data that support the preferred strategy.

D. Excluding limitations from the final report.

Correct Answer: B. Reporting methods and findings honestly, including inconvenient results.

Question 10

Which research question is most appropriate for exploratory interviews about sustainability claims?

A. What percentage of customers will repurchase next year?

B. How do customers interpret the brand’s sustainability claims?

C. What is the exact population mean of brand trust?

D. Did the randomised treatment produce a statistically significant effect?

Correct Answer: B. How do customers interpret the brand’s sustainability claims?