Digital Marketing Environment and Research

Introduction

Digital marketing decisions are made within an environment that is dynamic, interconnected and measurable—but never perfectly observable. Customers change their expectations, competitors alter their offers, platforms revise algorithms and policies, and wider forces such as economic conditions, regulation and cultural change influence demand. Effective digital marketers therefore do more than create content or purchase advertising. They gather evidence, interpret uncertainty and translate environmental insight into responsible decisions.

This chapter presents a practical research system for assessing customers, competitors, digital platforms and the wider environment. It integrates market research, social listening, keyword research, SWOT analysis and PESTLE analysis. It also examines the limitations of digital data, including bias, privacy concerns, measurement error and the danger of confusing online activity with genuine customer value.

Learning Objectives

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

1. Explain the role of environmental and marketing research in digital decision-making.
2. Assess customers using primary, secondary, qualitative and quantitative evidence.
3. Apply segmentation, personas and customer-journey analysis appropriately.
4. Conduct basic competitor, platform, social-listening and keyword research.
5. Use SWOT and PESTLE to structure strategic analysis.
6. Select suitable digital metrics and interpret them cautiously.
7. Recognise ethical, legal, methodological and practical limitations of digital data.
8. Convert research findings into clear recommendations and testable actions.

1. The Digital Marketing Environment

The marketing environment consists of forces that affect an organisation’s ability to create, communicate and deliver value. It can be considered at three connected levels:

The environment is not simply a list of influences. It is a system. For example, a privacy regulation may change platform tracking, which reduces the availability of advertising data, which affects targeting and measurement, which may encourage investment in first-party data and customer research.

Digital environments have several distinctive characteristics:

• Speed: Trends, complaints and competitive responses can develop within hours.
• Scale: A message can reach a large audience, while a small mistake can also spread widely.
• Interactivity: Customers can comment, review, share, remix and challenge brand communications.
• Traceability: Many actions produce digital records, although those records are incomplete and unevenly accessible.
• Platform dependence: Organisations often rely on search engines, social networks, marketplaces and advertising systems that they do not control.
• Personalisation: Messages and offers may be adapted using behavioural, demographic or contextual signals.
• Volatility: Platform algorithms, formats, costs and policies can change without an organisation’s consent.

Research is consequently not a one-off activity completed before a campaign. It is a continuing cycle of question formation, evidence collection, interpretation, action and evaluation.

2. Research as a Decision Process

Marketing research is the systematic design, collection, analysis and communication of information relevant to a marketing decision. Good research begins with a decision problem rather than with a fascination for available data.

A useful research sequence is:

1. Define the decision: What must the organisation decide?
2. Specify the information need: What must be known to make that decision responsibly?
3. Choose evidence sources: Which primary, secondary, qualitative or quantitative sources are suitable?
4. Collect and prepare evidence: Record methods, dates, samples, definitions and limitations.
5. Analyse patterns and alternatives: Look for relationships, differences, explanations and contradictory evidence.
6. Make a bounded recommendation: State what the evidence supports and what remains uncertain.
7. Act and evaluate: Implement, monitor outcomes and revise the understanding.

A research question should be specific enough to guide method selection. “What do people think of our brand?” is broad. “Which barriers prevent first-time urban customers aged 18–30 from completing an online purchase, and which messages might reduce those barriers?” is more useful because it identifies an audience, behaviour and decision context.

Primary and Secondary Research

Primary research is collected for the current problem. Examples include interviews, surveys, usability tests, observation, experiments and focus groups. It can be tailored to the decision but may require time, money and research skill.

Secondary research already exists and is reused for the current problem. Examples include government statistics, trade reports, company reports, academic studies, platform documentation, search trends and historical performance data. It is often efficient, but its definitions, date, sample and purpose may not match the current decision.

Qualitative and Quantitative Evidence

Qualitative research explores meaning, motivation, language and experience. Interviews and open-ended community discussions can reveal why customers abandon a process or distrust a claim. Qualitative findings are usually rich but are not automatically representative of a wider population.

Quantitative research measures frequency, distribution or relationships. Surveys, experiments and analytics can estimate how common a behaviour is or compare outcomes between groups. Quantitative precision does not eliminate bias: a large sample can still be misleading if it excludes important customers or measures the wrong thing.

A strong project often combines methods. This is called methodological triangulation. For example, analytics may show where users leave a website, a usability test may reveal what confuses them, and a survey may estimate how widespread the problem is.

3. Understanding Customers

Customer research should move beyond demographic description. Age, gender, location and income can be useful, but they rarely explain behaviour on their own. Digital marketers should examine needs, goals, constraints, triggers, perceived risk, information sources and the context in which decisions occur.

Segmentation

Segmentation divides a broad market into groups with meaningful similarities and differences. Common bases include:

• Geographic: country, region, climate or urbanisation.
• Demographic: age, household, occupation or income.
• Psychographic: values, interests, identity and lifestyle.
• Behavioural: purchase frequency, product usage, loyalty, channel activity or response to offers.
• Needs-based: problems customers are trying to solve and benefits they seek.
• Firmographic: industry, size and buying structure in business-to-business markets.

A segment is strategically useful when it is identifiable, substantial enough to matter, reachable, responsive to a differentiated offer and consistent with organisational capability. Digital targeting options do not automatically create meaningful segments. A platform category may be convenient for advertising while being a poor representation of customer need.

Personas and Jobs to Be Done

A persona is a research-informed representation of a significant customer type. It should include evidence, not invented decoration. A useful persona may describe the customer’s situation, objective, barriers, decision criteria, preferred information sources and likely objections.

The jobs-to-be-done perspective asks what progress a customer is trying to make in a particular circumstance. A customer may not be “buying project-management software”; they may be trying to coordinate a distributed team with less uncertainty. This perspective helps marketers connect content and offers to outcomes rather than to product features alone.

Personas are hypotheses, not facts. They should be reviewed when research, behaviour or market conditions change. Stereotypical personas can exclude minority users and encourage unsupported assumptions.

Customer Journey Analysis

A customer journey describes stages through which a person may move, such as awareness, consideration, purchase, use, retention and advocacy. Real journeys are not always linear. Customers may compare alternatives repeatedly, ask peers for advice, return after a long delay or begin with a marketplace rather than a brand website.

For each stage, investigate:

• customer goal and question;
• touchpoints and channels;
• content or evidence required;
• emotional and practical barriers;
• organisational action;
• outcome metric and research method.

A journey map is valuable when it identifies decisions and friction. It is less valuable when it merely lists channels without showing customer experience or evidence.

4. Competitor and Market Research

Competitor analysis examines organisations that compete for the same customers, budget, attention or problem-solving role. Direct competitors offer similar products to similar customers. Indirect competitors solve the same problem differently. Substitute behaviours—such as doing nothing, delaying purchase or using an informal alternative—can also compete with a brand.

A practical competitor audit may compare:

Competitor research must distinguish observation from inference. “The brand publishes three videos weekly” is an observation. “Video is its most profitable channel” is an inference that requires evidence not normally available from public inspection.

Avoid copying competitors without understanding the reason for their actions. A competitor’s high posting frequency may reflect a different budget, audience, product cycle or measurement system. The strategic question is not simply “What are they doing?” but “Which customer problem are they addressing, how effectively, and where is there credible opportunity to differ?”

5. Social Listening and Online Conversation

Social listening is the systematic monitoring and interpretation of public or authorised digital conversations about a brand, category, issue or competitor. It differs from social media monitoring. Monitoring often focuses on direct mentions and immediate response; listening seeks broader themes, sentiment, language, needs and emerging issues.

A basic social-listening process is:

1. Define the purpose, such as reputation management, product insight or issue detection.
2. Establish a query set containing brand names, product terms, common misspellings, competitors, category terms and relevant issues.
3. Select appropriate sources and record their coverage and access conditions.
4. Remove duplicates, spam, automated content and irrelevant uses of terms.
5. Code themes such as benefit, complaint, barrier, recommendation, misinformation or emerging need.
6. Examine volume, context, source, audience and change over time.
7. Compare findings with customer-service, sales and research evidence.
8. Escalate safety, legal, discrimination, privacy or crisis issues through approved procedures.

Sentiment classification can be useful for screening large volumes, but sentiment labels are not objective facts. Sarcasm, humour, dialect, multilingual expression and context can produce errors. A positive mention may concern a competitor or be ironic. A negative mention may be a valuable product insight rather than a reputational crisis.

Social conversation is also socially uneven. Highly active users are not necessarily representative customers. Public posts may overrepresent extreme experiences, organised campaigns or communities with reliable internet access. Researchers should report source coverage and avoid claiming that online discussion represents the whole market.

6. Keyword Research and Search Intent

Keyword research investigates the words and phrases people use when searching for information, products or solutions. It supports search-engine optimisation, paid search, content planning and understanding customer language.

Important dimensions include:

• Relevance: Does the phrase relate to the organisation’s offer and customer problem?
• Intent: Is the search informational, navigational, commercial investigation or transactional?
• Demand: How frequently or consistently is the phrase searched?
• Competition: How difficult or expensive may it be to gain visibility?
• Specificity: Does a longer, more detailed phrase reveal a clearer need?
• Context: Does meaning vary by location, device, season, language or audience?

Search volume is not the same as market demand or sales potential. A broad phrase may attract many searches but weak intent, while a low-volume phrase may represent a valuable specialist need. Keyword tools provide estimates based on their own data and models; they should be treated as directional evidence.

A disciplined workflow begins with customer language gathered from interviews, sales calls, support records and reviews. These terms are expanded using search suggestions, competitor pages and keyword tools, then grouped by intent and journey stage. Content is created to answer the underlying need rather than to repeat a phrase mechanically. Search engines and users both penalise irrelevant, misleading or low-value content over time.

7. SWOT Analysis

SWOT organises findings into four categories:

• Strengths: internal capabilities or assets that may create advantage.
• Weaknesses: internal limitations that reduce performance or resilience.
• Opportunities: external conditions that the organisation may exploit.
• Threats: external conditions that may damage performance or increase risk.

SWOT is a synthesis framework, not a research method. Each statement should be supported by evidence and defined in relation to a market. “Strong social presence” is too vague; “high share of relevant category conversations among the target segment, supported by a defined sample and period” is more useful.

The value of SWOT increases when it generates strategic choices. A strength may be used to capture an opportunity; a weakness may need correction before an opportunity is pursued; a strength may help reduce a threat; and a weakness may increase vulnerability. SWOT should not become a list of optimistic adjectives or uncontrollable events.

8. PESTLE Analysis

PESTLE examines macro-environmental forces:

• Political: government priorities, public policy, trade conditions and political stability.
• Economic: income, inflation, interest rates, employment and business confidence.
• Social: demographics, values, lifestyles, culture, trust and social norms.
• Technological: platforms, connectivity, automation, artificial intelligence, cybersecurity and technical standards.
• Legal: privacy, consumer protection, intellectual property, advertising rules and accessibility requirements.
• Environmental: climate risks, resource use, supply-chain impacts and sustainability expectations.

PESTLE should be connected to marketing decisions. For example, a change in privacy law may affect consent design, measurement and audience activation. Economic pressure may increase price sensitivity and demand for flexible payment. Environmental concern may increase scrutiny of delivery claims, packaging and green advertising.

A useful PESTLE table records the force, evidence, likely effect, time horizon, degree of uncertainty and possible response. It should distinguish an external trend from an organisational response. “Use more email” is not a PESTLE factor; “reduced availability of third-party behavioural signals” may be one.

9. Digital Metrics and Basic Interpretation

Digital research often includes numerical indicators. Metrics become meaningful only when their definitions, denominator, time period, attribution rules and data source are clear.

A commonly used engagement rate is:

Engagements may include reactions, comments, shares, saves or clicks, depending on the stated definition. Reach counts unique accounts or users where the platform’s definition permits; impressions count displays and may include repeated exposure. These denominators must not be treated as interchangeable.

For example, if a post receives 240 measured engagements and 12,000 impressions:

This calculation describes the selected measurement convention. It does not prove that the post caused sales or that 2% is good without a relevant benchmark and objective.

Conversion rate can be expressed as:

The denominator might be sessions, clicks, leads or users. A report must state which one is used. Cost per acquisition is:

These formulas are descriptive, not universal laws. They assume that the numerator and denominator are measured consistently, refer to the same period and are not double-counted. Attribution models may assign one conversion to several touchpoints or to a selected final interaction; therefore CPA is partly a property of the measurement system, not only of customer behaviour.

Vanity and Value Metrics

A vanity metric appears impressive but has weak connection to the decision, such as raw follower count. A value metric is linked more directly to an objective, such as qualified leads, repeat purchase, completed onboarding or customer retention. Vanity metrics are not always useless; reach can be relevant for an awareness objective. The test is whether the metric helps explain progress toward a clearly defined outcome.

10. Limitations of Digital Data

Digital data is powerful because it can be collected rapidly and at scale. It is limited because it is generated by systems designed for interaction, delivery or platform operation—not necessarily for unbiased research.

Common limitations

1. Coverage bias: People absent from a platform or dataset are invisible.
2. Self-selection: Those who post, click or respond may differ from quieter users.
3. Survivorship bias: Analysis of visible successful content ignores content that received little exposure.
4. Measurement error: Tracking can fail because of consent choices, blocked scripts, device changes or inconsistent tagging.
5. Attribution uncertainty: Multiple influences precede many purchases, making causal credit difficult.
6. Platform opacity: Algorithms, estimated audiences and changing definitions are often not fully transparent.
7. Data duplication: One person may appear as several devices or accounts; several people may share one account.
8. Temporal instability: A trend can be seasonal, news-driven or temporary.
9. Proxy problems: A click, like or search may not represent understanding, satisfaction or purchase intention.
10. Context loss: Short posts and isolated metrics may omit irony, history and social meaning.

Correlation does not establish causation. If sales rise after an influencer post, other explanations may include seasonality, price changes, distribution improvements or unrelated publicity. Stronger causal evidence may come from controlled experiments, matched comparisons, time-series analysis or triangulation, although each has assumptions.

Researchers should maintain a data dictionary, document collection dates, preserve definitions, distinguish observed from inferred variables and report uncertainty. A responsible report says what the evidence indicates, what it cannot establish and what test would reduce uncertainty.

11. Ethics, Privacy and Governance

Digital research involves responsibilities toward customers, employees, communities and the public. Core principles include purpose limitation, data minimisation, transparency, security, fairness, accountability and respect for consent. Applicable requirements vary by jurisdiction, but organisations should not assume that publicly visible information is ethically unrestricted for any purpose.

Important governance questions include:

• Was the data collected lawfully and fairly?
• Were people told how it would be used?
• Is the proposed use compatible with the original purpose?
• Are sensitive characteristics being inferred or used unfairly?
• Can individuals access, correct or delete data where applicable?
• How long should the data be retained?
• Who can access it, and how is it secured?
• Has the organisation assessed risks to vulnerable groups?

Targeting can become discriminatory when proxies for protected characteristics produce unequal access, pricing or exposure. Personalisation can become manipulative when it exploits vulnerability, obscures commercial intent or makes meaningful choice difficult. Artificial intelligence may accelerate transcription, clustering, sentiment analysis or content testing, but it can reproduce training-data bias, invent explanations and expose confidential information. Human review, approved tools, data controls and audit trails remain necessary.

12. An Integrated Research Framework

A practical project can combine the methods in this chapter as follows:

1. Frame the decision: define the business problem, target outcome and constraints.
2. Scan the environment: use PESTLE and secondary research to identify external changes.
3. Understand customers: combine interviews, surveys, analytics, reviews and journey mapping.
4. Map competition: compare offers, positioning, content, search visibility and experience.
5. Listen to language: use social listening and customer records to identify needs and concerns.
6. Study search behaviour: group keywords by intent and customer journey stage.
7. Synthesize: use SWOT to connect internal capability with external conditions.
8. Prioritise: assess potential impact, evidence strength, feasibility, risk and learning value.
9. Test: design a campaign, content experiment, usability improvement or offer test.
10. Evaluate and update: compare outcomes with the original objective and document limitations.

This framework avoids the common error of treating a single dashboard as the market. It also makes research actionable: every insight should lead to a decision, a test, a monitoring requirement or a justified decision not to act.

Applied Case: GreenBasket Online Grocery

This is an illustrative case created for educational purposes.

Background

GreenBasket is a small online grocery retailer specialising in local and lower-packaging products. Website traffic is increasing, but first-order conversion is below management expectations. The business wants to decide whether to invest first in paid search, social content or checkout improvements.

Problem

The team initially assumed that low conversion resulted from insufficient brand awareness. However, this assumption was based mainly on impressions and follower growth. A more disciplined investigation was required.

Analysis

Customer-service records showed repeated questions about delivery windows, minimum order value and product freshness. Website analytics indicated that many users viewed delivery information and then left. A usability session with eight recruited participants found that delivery charges were difficult to understand before checkout.

Keyword research identified high-intent phrases relating to local grocery delivery and specific delivery areas. It also showed informational searches about packaging and freshness. Social listening found positive discussion about local sourcing but concerns about price and reliability. Competitor analysis showed that two competitors displayed delivery fees and time slots prominently on product and basket pages.

PESTLE analysis identified economic pressure on household budgets, increased interest in sustainable consumption, privacy-related limits on behavioural tracking and potential environmental scrutiny of sustainability claims. SWOT synthesis identified GreenBasket’s local supplier relationships as a strength, weak delivery communication as an internal weakness, growing interest in local products as an opportunity and established delivery platforms as a threat.

Approach and Solution

The team prioritised checkout transparency before increasing acquisition spending. It redesigned delivery information, displayed the minimum order value earlier, added a delivery-area checker and created content answering freshness and packaging questions. It then planned a controlled comparison between the existing and revised checkout experience, with conversion defined as completed first orders and with the test period, traffic sources and exclusions documented.

Paid search was targeted toward high-intent local queries rather than broad sustainability terms. Social content used supplier stories and practical explanations, not unsupported environmental claims. The team also established a review process for claims and a monthly research cycle combining analytics, customer-service themes, search data and public conversation.

Outcome

The case does not claim a predetermined numerical result. Its important outcome is a change in decision quality: research redirected investment from broad awareness activity toward a documented customer barrier, while preserving later testing of acquisition channels. The business could then evaluate whether improved checkout clarity increased completed orders and whether new customers were retained.

Learning Points

• A visible metric such as traffic may conceal a customer-experience problem.
• Customer-service evidence can be as strategically useful as platform analytics.
• Competitor research should identify experience standards, not encourage imitation.
• PESTLE can reveal regulatory and economic implications of an apparently simple campaign.
• A hypothesis should be tested rather than presented as a proven causal explanation.

Professional Best Practices

• Begin with a decision and research question, not a preferred channel.
• Define every metric, denominator, time period and attribution rule.
• Combine behavioural data with customer explanations.
• Record sampling, collection and coding methods.
• Separate observation, interpretation and recommendation in reports.
• Use competitor and social data ethically and respect access restrictions.
• Treat keyword volume, sentiment and platform scores as estimates.
• Test important assumptions with experiments or additional research.
• Include underrepresented customers and assess accessibility.
• Protect personal information and minimise unnecessary collection.
• Recheck findings when platforms, regulations or market conditions change.
• Communicate uncertainty clearly to decision-makers.

Chapter Summary

Digital marketing research enables organisations to understand customers, competitors, platforms and wider environmental forces before committing resources. Effective research begins with a decision problem and combines primary and secondary, qualitative and quantitative evidence. Customer analysis should consider needs, context, behaviour and journey friction rather than relying only on demographic labels. Competitor research distinguishes observation from inference, while social listening and keyword research reveal public language, concerns and search intent.

SWOT synthesises internal strengths and weaknesses with external opportunities and threats; PESTLE structures analysis of political, economic, social, technological, legal and environmental forces. Digital metrics such as engagement rate, conversion rate and CPA require precise definitions and do not automatically establish value or causation. Digital data is affected by coverage bias, self-selection, platform opacity, tracking failure, attribution uncertainty and context loss. Ethical research therefore requires lawful and fair data use, transparency, security, minimisation, attention to discrimination and careful governance of artificial intelligence. The strongest practice is an iterative cycle of research, action, testing, evaluation and revision.

Key Terms

Attribution: The method used to assign credit for an outcome to one or more marketing touchpoints.

Competitor analysis: Systematic assessment of organisations or alternatives competing for customers, attention or resources.

Conversion rate: The percentage of defined opportunities that result in a specified conversion.

Customer journey: The sequence of experiences, decisions and touchpoints associated with a customer’s relationship with an organisation.

Engagement rate: A ratio of defined interactions to a stated reach or impression denominator.

First-party data: Data collected directly by an organisation through its own authorised customer or audience relationships.

Jobs to be done: The progress or problem-solving outcome a customer seeks in a particular situation.

Keyword research: Investigation of search terms, their meanings, demand, intent and competitive context.

Macro-environment: Wider external forces that influence markets and organisations.

Market segmentation: Division of a market into groups with meaningful similarities and differences.

Methodological triangulation: Use of multiple methods or evidence sources to improve understanding of a research problem.

PESTLE: A framework analysing political, economic, social, technological, legal and environmental forces.

Persona: A research-informed representation of a significant customer type.

Primary research: Original evidence collected for a specific research problem.

Secondary research: Existing evidence reused for a current research problem.

Sentiment analysis: Classification or interpretation of expressed attitudes or emotional direction in communications.

Social listening: Monitoring and interpreting digital conversations about brands, categories, issues or competitors.

SWOT: A framework for strengths, weaknesses, opportunities and threats.

Triangulation: Comparison of evidence from different sources, methods or perspectives.

Vanity metric: A measure that may appear impressive but has weak connection to a meaningful decision or outcome.

Review Questions

Short Answer Questions

1. Distinguish primary research from secondary research and give one digital marketing example of each.
2. Why should social-listening findings not automatically be treated as representative of all customers?
3. What is the difference between a strength and an opportunity in SWOT analysis?
4. Name the six components of PESTLE analysis.
5. Why must a conversion-rate denominator be defined clearly?

Long Answer Questions

1. Explain how an organisation could combine customer research, competitor analysis and journey mapping to improve an online conversion process.
2. Discuss the advantages and limitations of keyword research for digital marketing planning.
3. Explain how SWOT and PESTLE differ and how they can be used together in strategic planning.
4. Evaluate the claim that large volumes of digital data necessarily produce reliable marketing decisions.
5. Discuss the ethical and governance responsibilities involved in social listening and personalised targeting.

Analytical / Application Questions

1. A retailer observes high impressions, low click-through, many customer-service complaints and strong competitor visibility. Design a research plan that identifies the likely causes and proposes evidence-based actions.
2. A brand’s social posts receive highly positive sentiment, but sales and retention are declining. Analyse at least four possible explanations and specify what additional evidence should be collected.
3. Using the PESTLE and SWOT frameworks, assess how a privacy regulation that reduces third-party behavioural tracking could affect a small online retailer. Recommend two responsible responses and identify the assumptions that should be tested.

Multiple Choice Questions with Answers

Question 1

Which statement best describes the purpose of marketing research?

A. To collect the largest possible amount of data

B. To provide evidence relevant to a marketing decision

C. To guarantee that a campaign will succeed

D. To replace managerial judgement entirely

Correct Answer: B. To provide evidence relevant to a marketing decision

Question 2

Which research method is most directly designed to explore customers’ meanings, motivations and experiences?

A. Qualitative interviewing

B. Impression counting

C. Keyword-volume reporting

D. Competitor price scraping

Correct Answer: A. Qualitative interviewing

Question 3

In a SWOT analysis, an emerging customer need outside the organisation is normally classified as a(n):

A. Strength

B. Weakness

C. Opportunity

D. Internal capability

Correct Answer: C. Opportunity

Question 4

Which PESTLE category includes privacy regulation and consumer-protection rules?

A. Political

B. Technological

C. Environmental

D. Legal

Correct Answer: D. Legal

Question 5

A post receives 240 measured engagements and 12,000 impressions. Using impressions as the denominator, what is the engagement rate?

A. 0.2%

B. 2%

C. 5%

D. 20%

Correct Answer: B. 2%

Question 6

What is a major limitation of using public social-media conversations as a measure of the whole market?

A. They can never contain customer language

B. They are always collected through surveys

C. Active posters may not represent the wider customer population

D. They provide no information about issues or concerns

Correct Answer: C. Active posters may not represent the wider customer population

Question 7

Which search query most clearly indicates transactional intent?

A. What is recyclable packaging?

B. History of local grocery shops

C. Best foods for a picnic

D. Order organic groceries for delivery

Correct Answer: D. Order organic groceries for delivery

Question 8

Which statement about competitor research is most methodologically appropriate?

A. Visible posting frequency proves profitability

B. Observed activity should be distinguished from inferred performance

C. A competitor’s strategy should always be copied

D. Public content reveals every competitor capability

Correct Answer: B. Observed activity should be distinguished from inferred performance

Question 9

Why can a high-volume digital dataset still produce biased conclusions?

A. Large datasets cannot be analysed numerically

B. Data may exclude groups or systematically overrepresent active users

C. Quantitative data never contains patterns

D. Digital data is always collected illegally

Correct Answer: B. Data may exclude groups or systematically overrepresent active users

Question 10

Which practice best supports responsible digital research?

A. Collect all available personal data indefinitely

B. Treat platform sentiment labels as unquestionable facts

C. Define the purpose, minimise data and document limitations

D. Use a single metric for every marketing objective

Correct Answer: C. Define the purpose, minimise data and document limitations