Marketing decisions are made in an environment that firms do not control completely. Customers, competitors, regulators, technologies, cultural expectations, economic conditions and ecological constraints continuously alter the attractiveness of markets. Consequently, effective marketing management requires more than creativity or promotional skill. It requires disciplined environmental scanning, competitor intelligence and research capable of producing valid, ethical and actionable evidence.

This chapter presents an integrated approach. PESTLE analysis helps managers understand the macro-environment; Porter’s Five Forces evaluates industry structure; competitor analysis examines rival capabilities and strategic intentions; and marketing research converts managerial problems into defensible evidence. The chapter also considers qualitative and quantitative designs, sampling, measurement, experiments, secondary data and marketing intelligence systems.

Learning Objectives

1. The Marketing Environment as a System

The marketing environment is the set of external and internal forces that influence an organization’s ability to create, communicate, deliver and exchange value. It can be considered at four interconnected levels:

Level	Main concern	Typical evidence
Macro-environment	Broad forces affecting many industries	Inflation, legislation, demographic change, climate policy
Industry environment	Profitability and competitive structure	Rivalry, substitutes, supplier power, entry barriers
Market environment	Customers, channels and immediate partners	Needs, demand, distributors, influencers
Internal environment	Resources and capabilities under managerial influence	Brand equity, data, skills, finance, processes

The levels should not be treated as isolated. For example, a privacy regulation (macro) may increase compliance costs and alter competitive advantage (industry), change customer trust expectations (market), and require new data-governance capabilities (internal).

A useful distinction is between trend, event and weak signal. A trend is a persistent direction of change, such as population ageing. An event is a discrete occurrence, such as a regulatory ruling. A weak signal is an early, ambiguous indication of a possible future development. Intelligence work combines all three, while recognizing that extrapolating a trend is not the same as forecasting its consequences.

2. PESTLE Analysis

PESTLE examines Political, Economic, Social, Technological, Legal and Environmental forces. It is a scanning framework, not a predictive model. Its value depends on identifying implications, uncertainties and managerial responses rather than merely listing facts.

Political

Political factors include public policy, trade relations, taxation priorities, government stability, procurement rules and geopolitical risk. Marketing implications may include altered market access, restrictions on claims, public-sector demand or supply-chain disruption.

Economic

Economic analysis considers income, unemployment, interest rates, inflation, exchange rates, credit availability and consumer confidence. Managers should distinguish nominal from real growth. If sales increase by 8% while prices increase by 6%, the approximate real volume growth is not automatically 2% because the exact calculation depends on the price and volume indices. Using index values, real growth is:

Here growth rates are expressed as decimals. At 8% nominal growth and 6% inflation, real growth is 1.89% approximately: 1.08 ÷ 1.06 − 1 = 0.0189. The calculation assumes that the relevant inflation measure represents the prices affecting the sales measure.

Social

Social forces include demographic structure, lifestyles, identity, values, education, family patterns, health concerns and cultural norms. Segmentation based only on age or income can be misleading when attitudes, usage occasions and social identities explain behaviour more effectively.

Technological

Technology affects products, channels, data collection, production and customer experience. Artificial intelligence, automation, connected devices and retail-media platforms create opportunities but also risks involving bias, explainability, cybersecurity and surveillance.

Legal

Legal factors include consumer protection, competition law, intellectual property, product safety, employment law, accessibility and data-protection requirements. A marketing action can be technically feasible but legally impermissible or ethically unacceptable.

Environmental

Environmental forces include climate risk, resource scarcity, pollution controls, biodiversity concerns and consumer expectations concerning sustainability. Environmental claims require substantiation. Vague claims such as “green” may mislead if the basis, scope and comparison are unclear.

From scanning to strategic interpretation

Each component must be defined on a stated ordinal scale, such as 1–5. The resulting product is a prioritization aid, not a probability estimate or a scientifically validated forecast. It should not conceal disagreement or create false precision.

3. Industry Analysis: Porter’s Five Forces

Porter’s framework assesses the structural determinants of industry profitability. The five forces are rivalry among existing firms, threat of new entrants, bargaining power of buyers, bargaining power of suppliers and threat of substitutes.

Rivalry among existing firms

Rivalry tends to intensify when competitors are numerous or similar, industry growth is slow, products are weakly differentiated, fixed costs are high, switching costs are low and exit barriers are substantial. Price competition can reduce margins, while differentiation, service, ecosystems and innovation can shift competition away from price.

Threat of new entrants

Entry is easier when capital requirements, regulation, brand loyalty, access to distribution, network effects and scale economies are weak. Digital channels may lower some entry barriers while increasing the importance of data, trust, algorithms and installed user bases.

Bargaining power of buyers

Buyer power increases when customers are concentrated, purchases are large, offerings are standardized, switching is easy and buyers possess good information. A business-to-business buyer may have considerable negotiating power even when end consumers are fragmented.

Bargaining power of suppliers

Supplier power is stronger when suppliers are concentrated, inputs are differentiated, switching is costly, substitutes are limited and the supplier can integrate forward. Dependence on a single data platform or advertising intermediary can create strategic vulnerability.

Threat of substitutes

A substitute performs a similar customer job through a different solution. Streaming substitutes for cinema attendance; public transport substitutes for private car use; and do-it-yourself software substitutes for professional services. Substitution is about customer needs, not merely product categories.

The Five Forces model is most useful when managers specify the relevant industry and time period. It is less effective when used as a checklist detached from complementors, ecosystems, regulation or rapid technological discontinuity. In platform markets, indirect network effects and governance rules may need explicit analysis.

4. Competitor Analysis and Intelligence

Competitor analysis asks four questions: Who competes for the customer’s choice? What are competitors trying to achieve? What are they capable of doing? How are they likely to respond?

Competitors should be defined broadly. Direct competitors offer similar solutions; indirect competitors address the same need differently; potential competitors may enter from adjacent industries; and “share of wallet” competitors compete for the customer’s budget even when products are unrelated.

Competitive intelligence is the lawful and ethical collection, analysis and communication of information for decision-making. It differs from industrial espionage. Acceptable sources include public filings, published prices, websites, patents, job advertisements, trade publications, customer research and authorized channel observations. Deception, unauthorized access, bribery and misuse of confidential information are unacceptable.

A useful output is a competitor response matrix linking a firm’s move to possible rival reactions, their probability and the indicators that would reveal preparation. Intelligence becomes valuable when it is timely, sourced, triangulated and connected to a decision; a large undigested database is not intelligence.

5. Marketing Research: From Management Problem to Evidence

Marketing research is the systematic design, collection, analysis and communication of information relevant to a specific marketing decision. The process normally involves:

A management problem might be “sales are declining.” A research problem is more precise: “Which changes in perceived value, distribution availability and competitor switching are associated with declining repeat purchase among urban customers?” The distinction prevents data collection from becoming a substitute for managerial diagnosis.

A hypothesis is a testable proposition, for example: “Higher perceived delivery reliability is associated with higher repurchase intention.” Association does not establish causation. A causal claim requires temporal ordering, covariation and credible elimination of alternative explanations.

6. Research Paradigms and Design Choices

A qualitative design explores meanings, experiences, processes and context. Interviews, focus groups, ethnography, online communities and observation can reveal unmet needs and the language customers use. Samples are usually purposive or theoretical rather than statistically representative. Analysis involves coding, categorization, comparison and interpretation; researchers should document reflexivity and alternative explanations.

A quantitative design measures variables and estimates patterns in a population or tests specified relationships. Surveys, structured observation, experiments and analytics are common. Quantitative rigor depends on sampling, measurement quality, statistical assumptions and transparent analysis—not merely on a large sample.

A mixed-methods design combines approaches deliberately. Exploratory sequential research may use interviews to develop a survey instrument. Explanatory sequential research may use a survey followed by interviews to explain surprising results. Integration should occur in the research questions, sampling logic, analysis or interpretation, rather than by placing two unrelated studies side by side.

Sampling

Probability sampling gives units a known, non-zero selection probability and supports statistical generalization when implemented appropriately. Simple random, systematic, stratified and cluster sampling are common. Nonprobability methods include convenience, quota, purposive and snowball sampling; these may be practical or appropriate for exploratory work but do not automatically support population estimates.

For a proportion estimated with a simple random sample, a commonly used approximate sample-size formula is:

where n₀ is the initial sample size, z is the standard-normal critical value for the desired confidence level, p is the anticipated population proportion, and e is the maximum tolerable margin of sampling error. The formula assumes a probability sample, independent observations and a sufficiently large population relative to the sample. If no prior estimate of p is available, 0.5 gives the largest required sample for a fixed z and e.

For a finite population of size N, a finite-population correction can be applied:

These formulas address sampling error under stated assumptions. They do not correct nonresponse bias, coverage error, poor wording or dishonest answers. If the design uses clusters or unequal selection probabilities, a design effect and weighting strategy may be required.

Measurement and scales

Constructs such as satisfaction, trust and perceived value are not directly observed. Researchers operationalize them through indicators. Nominal scales classify categories; ordinal scales rank categories; interval-like scales represent ordered differences under stronger assumptions; ratio scales have a meaningful zero.

Reliability concerns consistency. Internal consistency is sometimes summarized with Cronbach’s alpha:

Here k is the number of items, σ²_i is the variance of item i, and σ²_t is the variance of the total score formed from the items. Alpha assumes a particular form of measurement equivalence and is influenced by the number of items; a high value does not prove validity or unidimensionality. Researchers should consider construct validity, content coverage, convergent and discriminant evidence, wording, response styles and measurement invariance across groups.

Validity asks whether the evidence supports the intended interpretation and use of scores. Threats include leading questions, acquiescence, recall error, common-method bias, poor translations and construct underrepresentation. Pretesting and cognitive interviewing help identify whether respondents interpret questions as intended.

Experiments and causal inference

Experiments manipulate a treatment, compare outcomes and use random assignment where feasible. In a randomized A/B test, the estimated average treatment effect is:

where Y(1) is the potential outcome under treatment, Y(0) is the potential outcome under control, and E denotes the population average. For each individual only one potential outcome is observed; random assignment allows the unobserved counterfactual average to be estimated under assumptions including correct implementation, no interference that undermines the design and adequate compliance.

A practical experiment requires a clearly defined unit of randomization, pre-specified primary outcomes, a control condition, sufficient duration, sample-size planning, treatment integrity and safeguards against repeated testing or selective reporting. Statistical significance alone does not establish managerial importance. Managers should examine effect size, confidence intervals, cost, heterogeneity and external validity.

Observational data can reveal associations and support prediction, but confounding may explain observed relationships. Before claiming that a campaign caused sales growth, analysts should consider seasonality, distribution changes, competitor actions, macroeconomic conditions and regression to the mean.

7. Secondary Data, Analytics and Insight Systems

Secondary data were collected for another purpose. Internal sources include transactions, CRM records, web analytics, service logs and loyalty data. External sources include official statistics, syndicated research, panels, industry reports, social listening and public filings.

Large datasets are not necessarily representative. Digital trace data often overrepresent active users, omit offline behaviour and change when platforms alter algorithms. Data linkage can increase analytical value while also increasing re-identification risk.

A marketing insight system integrates people, processes, technology and governance. It may contain a data warehouse or lake, identity and consent controls, dashboards, research repositories, experimentation tools and decision routines. A mature system distinguishes descriptive reporting (“what happened”), diagnostic analysis (“why might it have happened”), predictive analysis (“what is likely”) and prescriptive analysis (“what action is recommended”). Prediction is not explanation, and automated recommendations require human accountability.

8. Ethics, Governance and Responsible AI

Ethical research respects autonomy, beneficence, non-maleficence and justice. Participants should receive understandable information about purpose, procedures, risks, data use and withdrawal rights. Consent must not be coerced, and incentives should not unduly pressure vulnerable participants.

Researchers should minimize data collection, protect confidentiality, restrict access, define retention periods and report breaches appropriately. Sensitive data require heightened safeguards. Children and vulnerable groups may require additional consent and institutional review. Deception is permissible only when scientifically justified, proportionate, approved where applicable and followed by debriefing.

Marketing intelligence must comply with applicable privacy, consumer-protection, competition and intellectual-property law. Ethical practice exceeds legal compliance: a technically lawful use of location or inferred health data may still violate reasonable expectations.

AI systems introduce risks of biased training data, proxy discrimination, hallucinated evidence, opaque profiling, automated exclusion and uncontrolled disclosure of confidential information. Best practice includes documented data provenance, representative evaluation, human review, explainable decision rules where feasible, audit logs, security controls and a clear process for contesting consequential decisions. Generative AI may assist coding, summarization or questionnaire drafting, but researchers must verify outputs, protect confidential data and disclose material uses.

9. Integrated Applied Case: FreshRoute Grocery Delivery

Background

FreshRoute, a regional grocery-delivery service, has experienced falling repeat orders despite continued growth in online grocery adoption. Management suspects that discounting by large platforms, delivery unreliability and concern about data use are affecting retention.

Problem and analysis

A PESTLE scan identifies inflation, tighter privacy expectations, urban traffic policy, increased use of automated fulfillment and sustainability pressure concerning packaging. Five Forces suggests intense rivalry, moderate entry threats because technology is accessible, strong buyer power due to low switching costs, supplier power in specialized fresh produce, and strong substitutes including store pickup and direct retailer delivery.

Competitor analysis shows that one rival competes on low fees, another on reliability guarantees, and a third on local sourcing. FreshRoute’s internal data show that repeat ordering is lowest in districts with late deliveries, but this association may reflect difficult geography or lower-income segments rather than delivery performance alone.

Research approach

The firm first conducts 24 purposive interviews with recent lapsed customers, including varied age groups, order sizes and districts. Thematic analysis identifies four possible mechanisms: unreliable delivery windows, substitution of unavailable products, unclear refund processes and discomfort with personalized promotions.

The team then designs a probability-based survey of eligible customers, stratified by district and usage frequency. Measures include delivery reliability, perceived fairness, privacy concern, satisfaction and repurchase intention. Items are pretested through cognitive interviews. The questionnaire separates predictors and outcomes, uses balanced wording and records nonresponse.

Finally, FreshRoute conducts a randomized experiment among consenting active customers. Customers are assigned to either a standard delivery message or a transparent reliability message that provides a narrower window and automatic compensation for qualifying delays. The primary outcome is completed repeat order within 30 days; cost per incremental order is a secondary outcome. The analysis plan is registered internally before examining results.

Outcome and managerial response

Suppose the experiment produces a modest increase in repeat orders but compensation costs exceed the contribution margin in low-value orders. FreshRoute therefore targets the reliability guarantee to high-frequency segments, improves inventory substitution controls and revises privacy explanations rather than applying a universal discount. It also establishes a monthly intelligence review combining regulatory signals, competitor monitoring, customer research and experiment results.

Learning points

The case demonstrates that environmental frameworks generate hypotheses but do not prove causes. Internal analytics identified a pattern, qualitative research clarified possible mechanisms, survey research measured prevalence, and experimentation tested an intervention. Ethical consent, transparent communication and economic evaluation prevented the company from treating a statistically positive result as automatically profitable.

10. Limitations and Best Practices

Frameworks simplify reality. PESTLE categories overlap; Five Forces can understate ecosystems and complementors; competitor intelligence can be distorted by public posturing; and research can be weakened by biased samples or measures. Managers should therefore triangulate sources, state assumptions, distinguish facts from interpretations, use alternative scenarios and invite disconfirming evidence.

Best practice is to begin with a decision and success criterion, not a favorite method. Use qualitative work to understand meaning and generate constructs; use quantitative work to estimate prevalence and relationships; use experiments when a causal decision is feasible; and use secondary data when definitions and provenance are adequate. Document sampling, exclusions, missing-data treatment, coding decisions and analytic changes. Communicate uncertainty in plain language and connect every recommendation to evidence, risk and implementation responsibility.

Chapter Summary

Marketing analysis connects environmental understanding with evidence-based decision-making. PESTLE identifies broad political, economic, social, technological, legal and environmental forces, while Five Forces assesses industry profitability and competitor analysis examines rival objectives, capabilities and likely responses. These frameworks are analytical starting points rather than mechanical forecasting tools.

Marketing research requires a clear distinction between the management problem and the research problem. Qualitative, quantitative, mixed-methods and experimental designs answer different questions. Sampling determines what can be generalized; measurement determines whether constructs are represented consistently and validly; experiments strengthen causal inference but require careful design. Secondary data and intelligence systems are valuable only when their provenance, relevance, bias, privacy and governance are understood. Ethical practice, responsible AI and transparent communication are central to credible marketing intelligence.

Key Terms

Marketing environment: Internal and external forces influencing marketing decisions.

Macro-environment: Broad political, economic, social, technological, legal and environmental context.

Five Forces: Framework analyzing rivalry, entrants, buyers, suppliers and substitutes.

Competitive intelligence: Lawful collection and analysis of information about competitive conditions.

Marketing research: Systematic process of generating information for a marketing decision.

Quantitative research: Research measuring variables and estimating patterns numerically.

Mixed methods: Deliberate integration of qualitative and quantitative approaches.

Probability sampling: Sampling in which units have known non-zero selection probabilities.

Validity: Support for the intended interpretation and use of evidence or scores.

Experiment: Design involving manipulation and comparison to assess causal effects.

Secondary data: Data originally collected for a purpose other than the current study.

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

Marketing intelligence system: Integrated people, processes, data and technology supporting marketing decisions.

Review Questions

Short Answer Questions

Long Answer Questions

Analytical / Application Questions

Multiple Choice Questions with Answers

Question 1

B. It scans broad external forces and helps identify implications and uncertainties.

Correct Answer: B. It scans broad external forces and helps identify implications and uncertainties.

Question 2

Question 3

A research project begins with interviews and then develops a survey from the themes discovered. What design is this?

Question 4

Using n₀ = (z² × p × (1 − p)) ÷ e², what is the approximate initial sample size when z = 1.96, p = 0.5 and e = 0.05?

Question 5

Question 6

Question 7

Question 8

Correct Answer: B. A cinema and a streaming service addressing entertainment needs

Question 9

Question 10

Why should an observed association between campaign exposure and sales not automatically be called causal?

B. Confounding factors such as seasonality or self-selection may explain the association.

Correct Answer: B. Confounding factors such as seasonality or self-selection may explain the association.

Chapter 2 Marketing Environment, Intelligence and Research

Marketing Environment, Intelligence and Research

Introduction