Market Research and the Consumer Insight Function

Introduction

Product decisions are often made under conditions of uncertainty. A product manager may need to decide which customer problem to solve, how large the opportunity is, which segment to prioritise, what value proposition to communicate, how much customers may pay, or whether an existing product should be changed or discontinued. Market research and consumer insight help make these decisions more disciplined by providing systematic evidence about markets, customers, competitors, contexts and outcomes.

Research does not eliminate uncertainty. Nor does it replace managerial judgement, creativity or strategic vision. Its purpose is to improve the quality of decisions by clarifying what is known, what is inferred, what remains uncertain and what should be tested next. The consumer insight function therefore acts as an organisational capability that connects evidence with action.

This chapter establishes the foundations of market research and consumer insight for product managers. It explains the terminology, evolution and industry structure of the field; distinguishes major forms of research; introduces insight maturity; and examines how evidence should influence product and managerial decisions.

Learning Objectives

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

1. Define market research, consumer research, customer insight and related terms.
2. Explain how the market research industry has evolved from descriptive measurement to strategic, technology-enabled insight.
3. Distinguish exploratory, descriptive, causal, evaluative, primary and secondary research.
4. Describe the roles of clients, research agencies, data providers, technology firms and participants.
5. Assess the difference between data, information, findings, insight and action.
6. Evaluate the maturity of an organisation’s insight capability.
7. Apply principles of evidence-based product and managerial decision-making.
8. Recognise methodological, ethical, privacy and interpretation risks.

1. The Purpose of Market Research

Market research is the systematic design, collection, analysis and interpretation of information relevant to a specified marketing or business decision. It may concern customers, non-customers, competitors, channels, category dynamics, pricing, communications, usage contexts or organisational capabilities.

The word systematic is important. Market research is not simply asking a few colleagues what they think, reading online comments, or selecting statistics that support a preferred conclusion. Systematic research begins with a decision problem, defines the information required, uses an appropriate method, documents assumptions, analyses evidence and communicates implications.

A useful distinction is between a decision problem and a research problem. A decision problem is an action question, such as “Should the company launch a lower-priced subscription?” A research problem translates that question into evidence requirements: “What unmet needs exist among price-sensitive users, how do they evaluate alternatives, what trade-offs will they make, and how might a lower price affect adoption and retention?”

Research commonly supports five decision functions:

Research is most valuable when it is conducted early enough to influence a decision and when its findings are expressed in terms that decision-makers can use.

2. Essential Terminology

The field contains overlapping terms that should not be treated as synonyms.

Market refers to the set of actual and potential exchanges involving a category, need or solution. A market can be defined by products, customer problems, usage occasions, geography or institutional arrangements. The same organisation may define its market narrowly as “online grocery delivery” or more broadly as “convenient access to household food.”

Customer is an individual or organisation that buys, subscribes to, uses, influences or benefits from an offering. The buyer, user, decision-maker and payer may be different people, especially in business-to-business and household settings.

Consumer traditionally refers to an individual who uses or consumes a product. In contemporary practice, the term is also used broadly for people who experience a category, whether or not they purchase from the focal organisation.

User research focuses on how people interact with a product, service or system, particularly their tasks, behaviours, frustrations and context. It is often closely associated with product design and usability.

Customer research investigates existing or prospective customers’ needs, experiences, perceptions, behaviours and relationships with an organisation.

Market research includes customer and consumer research but also covers competitors, market structure, distribution, category trends, pricing and the broader commercial environment.

Consumer insight is a well-supported interpretation of evidence that reveals a meaningful human or market truth and indicates an opportunity for action. A statement such as “60% of respondents prefer option A” is a finding. An insight might be: “People say they want more choice, but in time-pressured situations they choose the option that reduces decision effort.” The latter connects behaviour, context and implication.

Voice of the customer (VoC) is an organised representation of customer needs, expectations, complaints and feedback across touchpoints. VoC systems may combine surveys, interviews, service interactions, reviews and behavioural data.

Analytics is the systematic examination of data to describe, diagnose, predict or prescribe action. Analytics and research overlap, but analytics often uses existing behavioural or operational data, while research frequently generates new evidence to answer a defined question.

Evidence is information that bears on a claim. Its usefulness depends on relevance, validity, reliability, quality, timeliness and suitability for the decision—not merely on volume.

3. From Market Measurement to Consumer-Centred Insight

The market research profession has changed alongside business, technology and social expectations.

Early measurement and opinion research

Early commercial research focused on sales territories, distribution, advertising reach and descriptive market measurement. Large-scale surveys and opinion polling established procedures for sampling, questionnaires and statistical summarisation. The central model was often linear: define a population, ask standardised questions, estimate preferences and report results.

The growth of segmentation and motivation research

As markets became more competitive, firms sought to distinguish groups with different needs and purchasing patterns. Segmentation research combined demographic, geographic, behavioural and psychographic variables. Qualitative approaches—such as depth interviews, observation and group discussions—were used to explore motivations that respondents might not express through structured questions.

The digital and data-rich period

The expansion of websites, mobile devices, loyalty programmes, social platforms and digital advertising produced large volumes of behavioural data. Research became faster and more continuous. Online panels, social listening, digital experiments and product analytics enabled firms to observe actions rather than rely only on stated opinions.

However, more data did not automatically produce better insight. Digital data may be incomplete, biased toward highly active users, difficult to interpret, or disconnected from the underlying customer need. The profession consequently moved toward triangulation: combining different methods and data sources to understand both what people do and why.

Contemporary insight practice

Today, insight teams frequently operate across marketing, product, design, strategy, customer experience and data science. Research may be conducted through agile sprints, remote interviews, in-product prompts, concept tests, behavioural experiments and continuous feedback systems. Artificial intelligence can assist with transcription, coding, summarisation, search and pattern detection, but human researchers remain responsible for research design, context, ethics, interpretation and accountability.

The direction of evolution can be summarised as follows:

4. The Market Research and Insight Industry

Market research is an ecosystem rather than a single profession.

Clients and internal insight teams

The client is the organisation or team that owns the decision. Internal researchers, insight managers, product researchers and analysts translate business problems into research programmes, commission external partners, integrate evidence and facilitate decisions. Their value lies not only in technical research skill but also in stakeholder management and commercial understanding.

Full-service research agencies

Full-service agencies design and execute qualitative, quantitative and mixed-method studies. They may provide sample recruitment, fieldwork, analysis, workshops and strategic recommendations.

Specialist agencies

Specialists focus on particular methods or problems: ethnography, usability, pricing, conjoint studies, brand tracking, cultural research, customer experience, business-to-business research or medical and regulatory research.

Sample and panel providers

Panel providers recruit and maintain groups of people who agree to participate in research. They manage screening, incentives, identity checks and sample delivery. Panel quality depends on recruitment sources, participation behaviour, fraud controls, representativeness and transparent documentation.

Data, technology and analytics providers

These organisations provide survey platforms, social listening, customer data platforms, behavioural analytics, experimentation tools, dashboards, artificial intelligence services and data enrichment. Technology can improve speed and scale, but it does not solve poorly framed questions or biased data.

Participants and communities

Research participants contribute information, time and personal experience. They are not merely passive data points. Ethical practice requires informed participation, appropriate compensation, protection from harm and responsible use of their information.

Professional bodies and standards

Professional associations and standards bodies publish guidance on sampling, disclosure, privacy, quality, accessibility and research conduct. Standards do not guarantee a valid study, but they provide a basis for professional accountability and transparent practice.

5. Major Research Types

No single method is universally superior. Method selection should follow the decision, the type of uncertainty and the nature of the evidence required.

Exploratory research

Exploratory research investigates poorly understood problems and helps generate hypotheses. Common methods include literature review, stakeholder interviews, depth interviews, ethnography, observation, diary studies and open-ended group discussions. It is especially useful at the beginning of product discovery.

Exploratory research is usually flexible and purposive rather than statistically representative. Its goal is depth, meaning and discovery, not estimating the proportion of a population with a known margin of error.

Descriptive research

Descriptive research measures the characteristics, incidence, distribution or prevalence of a market or customer phenomenon. Surveys, structured observation, market sizing studies, brand tracking and customer databases are common examples. It addresses questions such as “How many?”, “Who?”, “When?” and “Where?”

A descriptive result must be interpreted in relation to sampling, measurement and non-response. A large sample with a poorly defined population can still produce misleading conclusions.

Causal research

Causal research examines whether changing one factor produces a change in an outcome. Randomised experiments, A/B tests, field trials and some quasi-experimental designs are used. For example, users may be randomly assigned to two onboarding flows, with activation rates compared after exposure.

Random assignment helps balance other influences between groups. A causal conclusion still requires appropriate treatment exposure, reliable outcome measurement, sufficient observation time and attention to implementation effects. Correlation alone does not establish causation.

Evaluative research

Evaluative research assesses an idea, prototype, product, service, campaign or experience against specified criteria. It includes usability testing, concept testing, pilot evaluation, satisfaction measurement and post-launch reviews. Evaluation is strongest when success criteria are defined before results are observed.

Primary and secondary research

Primary research collects new information for the current question, such as interviews, surveys or experiments. Secondary research analyses existing information, such as public statistics, academic studies, company records, industry reports or prior research.

Secondary research is often faster and less expensive, and it should usually be reviewed before commissioning new work. Its limitations may include different definitions, outdated data, unclear methods, restricted access and commercial bias.

Qualitative, quantitative and mixed methods

Qualitative research seeks meanings, interpretations, experiences and processes. Quantitative research measures variables and examines patterns across cases. Mixed-method research deliberately combines both, often using qualitative work to develop measures and quantitative work to estimate prevalence, or using quantitative results to identify groups for deeper investigation.

The choice is not a competition between “soft” and “hard” methods. A statistically precise answer to the wrong question is not useful, and a compelling interview does not establish population prevalence.

6. From Data to Actionable Insight

A disciplined insight process separates five levels:

1. Data: recorded observations, responses, transactions, behaviours or statements.
2. Information: organised data with context and descriptive meaning.
3. Finding: an analysed result relevant to the research question.
4. Insight: an interpretation that explains significance, tension, motivation or opportunity.
5. Action: a decision, experiment, design change, investment or policy response.

Consider a streaming service. Data may show that many users stop browsing after viewing several titles. Information may show that abandonment is highest among new users. A finding may be that new users take longer to select a programme than experienced users. An insight may be that extensive choice creates uncertainty before users have learned the service’s recommendation logic. An action could be a guided first-session experience that reduces choice complexity.

A useful insight should be:

• Grounded in credible evidence;
• Relevant to a meaningful decision;
• Non-obvious, adding interpretation beyond a simple description;
• Humanly meaningful, explaining needs, tensions or context;
• Actionable, indicating what could be changed or tested;
• Bounded, making clear where and for whom it applies.

Insight is not the same as a dramatic quotation, a dashboard, a statistical significance result or a management opinion. It emerges through disciplined interpretation and must remain open to challenge.

7. Evidence in Product and Managerial Decision-Making

Evidence-based management does not mean that every decision requires a large survey or that only numerical evidence is legitimate. It means that decisions should consider the best available evidence, the quality and limitations of that evidence, stakeholder expertise, organisational constraints and the consequences of action.

A practical evidence assessment asks:

Evidence has different roles at different product stages. In discovery, it helps identify problems and contexts. In definition, it supports prioritisation and segmentation. In development, it evaluates prototypes and usability. In launch, it informs positioning, channels and readiness. After launch, it helps diagnose adoption, retention, satisfaction and unintended consequences.

A product manager should avoid two opposite errors. Opinion-led decision-making treats seniority, intuition or isolated anecdotes as sufficient evidence. Dataism assumes that whatever is measurable is necessarily important and that more data automatically produces better decisions. Good practice combines judgement with appropriately designed evidence.

8. Insight Maturity

Insight maturity describes how effectively an organisation turns evidence into better decisions. It is not simply the amount of research purchased or the sophistication of its software.

Level 1: Ad hoc evidence

Research occurs reactively, often after a decision has already been informally made. Definitions are inconsistent, findings are difficult to access and learning is not retained.

Level 2: Project-based research

The organisation commissions competent studies for individual decisions. Quality may be acceptable, but research remains episodic and teams may duplicate work.

Level 3: Integrated insight

Research, analytics, customer feedback and product data are connected. Insight teams participate earlier in strategy and product development, and common frameworks support comparison across studies.

Level 4: Decision-centric capability

The organisation defines learning agendas around strategic uncertainties. It triangulates evidence, links research to outcomes and uses experimentation to learn quickly. Stakeholders share responsibility for acting on insight.

Level 5: Learning organisation

Continuous learning is embedded in governance, product development and resource allocation. Teams test assumptions, document decisions, revisit predictions and treat customers as partners in responsible innovation.

Maturity should be assessed across several dimensions: leadership sponsorship, methodological capability, data accessibility, research governance, stakeholder participation, speed, actionability and learning from outcomes. A technically advanced organisation may still be immature if its evidence rarely changes decisions.

9. Designing a Strong Research Programme

A professional research programme usually follows these steps:

Frame the decision

State the decision owner, decision date, alternatives under consideration and consequences of delay or error. Avoid beginning with a method, such as “We need a survey.” Begin with uncertainty.

Define learning objectives

Specify what must be learned, what is already known and what evidence would change the decision. Prioritise questions rather than attempting to answer everything.

Choose the research design

Select methods based on the question. Interviews may be suitable for discovering language and motivations; a survey may estimate incidence; an experiment may test causal impact; behavioural data may reveal actual usage.

Plan sampling and recruitment

Define the relevant population, inclusion criteria, sampling approach and recruitment controls. In qualitative work, seek meaningful variation. In quantitative work, consider coverage, sample size, non-response and weighting where appropriate.

Protect measurement quality

Use clear wording, suitable response options, neutral sequencing and pilot testing. Avoid leading questions, double-barrelled questions and assumptions that participants cannot answer reliably.

Analyse and triangulate

Examine patterns, exceptions and alternative explanations. Compare evidence across methods and identify contradictions rather than hiding them. Contradictions may reveal segment differences, context effects or measurement problems.

Communicate implications

A strong deliverable identifies the decision, evidence, interpretation, uncertainty and recommended action. It distinguishes observed results from hypotheses and makes limitations visible.

Follow through

Record the decision, assumptions and expected outcomes. After implementation, compare results with expectations. This converts isolated research into organisational learning.

10. Ethics, Governance and Regulation

Research involves power, personal information and potential influence. Ethical practice is therefore central, not an administrative afterthought.

Participants should understand the general purpose of the study, what participation involves, how information will be used and whether participation is voluntary. Consent should be meaningful and appropriate to the sensitivity of the research. Researchers must protect confidentiality, minimise unnecessary collection and retain data only as long as justified.

Privacy obligations vary by jurisdiction, but responsible practice commonly includes purpose limitation, data minimisation, access controls, secure storage, deletion procedures and transparent notices. Sensitive topics, children, vulnerable groups, employee research and health-related information require additional safeguards.

Researchers should disclose sponsorship where it could affect interpretation. They should not fabricate participants, manipulate findings, suppress inconvenient results or present exploratory patterns as proven facts. Incentives must encourage participation without creating undue pressure.

Artificial intelligence introduces further governance questions. Automated transcription, coding and summarisation can expose confidential data, reproduce bias or create plausible but unsupported interpretations. Before using AI, teams should assess data permissions, vendor controls, human review, traceability and the consequences of error. AI-generated summaries should never be treated as a substitute for examining the underlying evidence.

11. Challenges and Limitations

Representation and sampling bias

A sample may differ systematically from the population because some people are unreachable, unwilling or excluded by the recruitment process. Online access, language, disability, geography and platform use can affect coverage.

Stated versus revealed behaviour

People may misremember, misunderstand, rationalise or answer in socially desirable ways. Stated intention can be useful, but actual behaviour may differ because of price, convenience, habit or context.

Measurement error

Poor wording, confusing scales, survey fatigue, interviewer effects and inappropriate constructs can distort results. Reliability—consistency of measurement—is not identical to validity, which concerns whether the measure captures what it claims to capture.

Context and transferability

A finding from one segment, market, device or occasion may not transfer elsewhere. Qualitative depth and experimental control can come at the cost of breadth or naturalness.

Organisational bias

Researchers may unconsciously favour a sponsor’s hypothesis, while stakeholders may selectively cite evidence. Pre-registration of key hypotheses, transparent analysis plans and independent challenge can reduce these risks.

Speed and quality trade-offs

Agile research can support rapid learning, but compressed timelines may reduce recruitment diversity, piloting and analytical depth. The appropriate standard depends on decision risk: a reversible interface change may require less evidence than a safety-critical or high-cost market entry decision.

12. Illustrative Case: Improving Adoption of a Digital Banking Feature

The following case is fictional and intended for educational use.

Background

A digital bank introduced an automated savings feature that transferred small amounts of money into a separate account. Product analytics showed strong initial interest: many customers opened the feature page, but relatively few completed setup.

Problem

The product team initially assumed that customers did not trust automated transfers. It proposed adding reassurance messages and a prominent security badge. Before committing resources, the insight lead reframed the problem: the team needed to understand where setup failed, what customers believed would happen, and which concerns were functional, emotional or procedural.

Analysis and approach

The team combined four sources of evidence:

1. Funnel analytics to locate abandonment points.
2. Remote interviews with customers who completed and abandoned setup.
3. A usability test of the setup flow.
4. A controlled experiment comparing the existing flow with a simplified version.

Analytics showed that abandonment was concentrated on a screen requiring customers to choose a transfer rule. Interviews revealed that customers did not primarily fear fraud; they were uncertain about how the rule would interact with irregular income and existing bills. Usability sessions showed that the interface used financial terminology unfamiliar to many participants.

The resulting insight was: Customers valued automatic saving, but uncertainty about control and timing made the feature feel risky; the design asked them to configure a financial system before they understood its consequences.

The team redesigned the flow around a plain-language preview, a flexible pause control and examples using realistic income patterns. It also allowed customers to begin with a conservative default and change it later.

Outcome

In the experiment, the redesigned flow produced a higher setup completion rate than the original flow. The team continued monitoring cancellation, overdraft incidents and customer complaints rather than treating completion alone as success. The redesign was therefore evaluated both for adoption and for unintended financial harm.

Learning points

The case illustrates that:

• Behavioural data identifies where a problem occurs but may not explain why.
• Customers’ first stated concern may not be the underlying barrier.
• Qualitative evidence can improve the interpretation of quantitative patterns.
• A product change should be tested against both primary and unintended outcomes.
• Insight is valuable when it changes design and clarifies the next experiment.

Contemporary Best Practices

A high-quality insight function should:

• Begin with decisions and uncertainties rather than preferred methods.
• Combine behavioural, attitudinal and contextual evidence.
• Make population, sample and data limitations explicit.
• Include diverse and accessible participant experiences.
• Involve product, design, analytics, legal and customer-facing teams early.
• Use small, rapid studies when learning is reversible, and stronger designs when consequences are material.
• Maintain a searchable repository of methods, findings, decisions and outcomes.
• Treat contradictions as learning opportunities.
• Separate facts, interpretations, hypotheses and recommendations.
• Measure whether insight changes decisions and improves outcomes.

Chapter Summary

Market research is the systematic generation and interpretation of evidence for market, customer and product decisions. Consumer insight goes beyond reporting data: it explains a meaningful human or market truth, establishes why it matters and identifies an opportunity for action.

The field has evolved from periodic measurement and opinion research toward integrated, continuous and cross-functional learning. Its industry includes client organisations, internal insight teams, full-service and specialist agencies, sample providers, technology firms, analytics specialists, professional bodies and research participants.

Exploratory, descriptive, causal, evaluative, primary, secondary, qualitative and quantitative approaches answer different questions. No method is inherently best; quality depends on fit with the decision, population, evidence requirement and consequences of error. Effective insight processes distinguish data, information, findings, insight and action, and they triangulate evidence without concealing uncertainty.

Insight maturity progresses from ad hoc research to a learning organisation in which evidence shapes strategy, product development, governance and resource allocation. Ethical practice requires informed participation, privacy protection, transparency, methodological honesty and responsible use of technology. For product managers, the central lesson is that evidence is not a substitute for judgement: it is a disciplined input that makes judgement more explicit, testable and accountable.

Key Terms

Market research: Systematic design, collection, analysis and interpretation of information relevant to a market or business decision.

Consumer research: Study of consumers’ needs, perceptions, experiences and behaviour.

Customer research: Research concerning existing or prospective customers and their relationship with an organisation.

Consumer insight: Evidence-based interpretation that reveals a meaningful truth and indicates an action or opportunity.

Market: A set of actual and potential exchanges defined by a category, need, solution or context.

User research: Investigation of how people use, experience or interact with a product or service.

Voice of the customer: Structured representation of customer needs, expectations and feedback.

Primary research: New data collected for a specific current question.

Secondary research: Analysis or use of information previously collected for another purpose.

Exploratory research: Research used to discover meanings, problems, motivations or hypotheses.

Descriptive research: Research that measures characteristics, distributions or prevalence.

Causal research: Research designed to assess whether one factor produces a change in another.

Evaluative research: Research that assesses an idea, product, service or experience against criteria.

Qualitative research: Research focused on meanings, experiences, processes and interpretations.

Quantitative research: Research involving measurement and numerical analysis across cases.

Triangulation: Combining multiple methods, sources or perspectives to strengthen interpretation.

Evidence: Information that supports or challenges a claim.

Insight maturity: The capability to convert evidence into consistent, high-quality decisions and organisational learning.

Sampling bias: Systematic difference between those included in research and the relevant population.

Data minimisation: Collecting only the personal information necessary for a legitimate purpose.

Review Questions

Short Answer Questions

1. What is the difference between a decision problem and a research problem?
2. Define consumer insight and explain how it differs from a finding.
3. Distinguish exploratory research from descriptive research.
4. Why is triangulation useful in consumer research?
5. Identify two responsibilities of an internal insight team.

Long Answer Questions

1. Explain the evolution of market research from periodic measurement to contemporary insight practice.
2. Compare qualitative, quantitative and mixed-method research, including situations in which each is appropriate.
3. Discuss the role of evidence in product management and explain why evidence should not be confused with data volume.
4. Describe the five levels of insight maturity and the organisational changes required to progress between them.
5. Explain the principal ethical and governance obligations involved in conducting consumer research.

Analytical / Application Questions

1. A product team observes that many users abandon a new feature during setup. Design a research programme that combines analytics, qualitative research and experimentation. Explain what each method would contribute.
2. A survey reports that 70% of respondents intend to purchase a new service, but early sales are weak. Analyse at least four possible explanations and propose evidence that would distinguish among them.
3. An organisation has many dashboards and research reports, but product teams rarely change their decisions. Diagnose the likely insight-maturity problems and recommend a practical improvement plan.

Multiple Choice Questions with Answers

Question 1

Which statement best defines market research?

A. A process of collecting only customer complaints

B. The systematic collection, analysis and interpretation of information for a market or business decision

C. A creative exercise for generating advertising ideas

D. A financial forecast based only on historical sales

Correct Answer: B. The systematic collection, analysis and interpretation of information for a market or business decision

Question 2

What distinguishes a consumer insight from a simple research finding?

A. An insight always comes from a very large sample

B. An insight is always expressed as a percentage

C. An insight interprets why evidence matters and indicates an opportunity for action

D. An insight excludes qualitative evidence

Correct Answer: C. An insight interprets why evidence matters and indicates an opportunity for action

Question 3

Which type of research is primarily intended to generate hypotheses about poorly understood problems?

A. Exploratory research

B. Descriptive research

C. Causal research

D. Evaluative research

Correct Answer: A. Exploratory research

Question 4

Which method is most directly suited to testing whether a change in an onboarding flow causes higher activation?

A. An unstructured stakeholder discussion

B. A randomised A/B experiment

C. A literature review

D. An ethnographic observation with no comparison condition

Correct Answer: B. A randomised A/B experiment

Question 5

What is the principal purpose of triangulation?

A. To replace research design with more data

B. To combine evidence from multiple sources or methods to strengthen interpretation

C. To ensure that every study uses quantitative methods

D. To guarantee that findings are statistically significant

Correct Answer: B. To combine evidence from multiple sources or methods to strengthen interpretation

Question 6

Which example is secondary research?

A. Interviewing prospective customers for a new study

B. Running an experiment on two product designs

C. Analysing a previously published government market report

D. Conducting a new usability test

Correct Answer: C. Analysing a previously published government market report

Question 7

Which statement about insight maturity is most accurate?

A. Maturity is measured only by the number of research reports produced

B. A mature organisation connects evidence to decisions, outcomes and organisational learning

C. Maturity requires using artificial intelligence in every study

D. A mature organisation avoids qualitative research

Correct Answer: B. A mature organisation connects evidence to decisions, outcomes and organisational learning

Question 8

A sample excludes people without reliable internet access, although they are part of the target market. What risk is most evident?

A. Sampling or coverage bias

B. Random assignment

C. Causal validity

D. Data minimisation

Correct Answer: A. Sampling or coverage bias

Question 9

Which practice best reflects responsible research governance?

A. Collecting all available personal information in case it becomes useful

B. Hiding sponsorship to prevent participants from changing their answers

C. Using clear consent, limiting data collection and protecting participant information

D. Allowing automated summaries to replace human review

Correct Answer: C. Using clear consent, limiting data collection and protecting participant information

Question 10

A dashboard shows that feature abandonment occurs on one screen, but interviews reveal that users are uncertain about the feature’s consequences. What does this combination illustrate?

A. Behavioural data can locate a problem while qualitative research can help explain it

B. Qualitative evidence makes analytics unnecessary

C. Interviews establish population prevalence more accurately than all surveys

D. A correlation automatically proves the cause of abandonment

Correct Answer: A. Behavioural data can locate a problem while qualitative research can help explain it