Segmentation, Personas and Consumer Targeting

Introduction and Context

Brand management depends on making choices about whom a brand will serve, how those consumers differ, and why the brand should be meaningful to them. A market is rarely homogeneous. Consumers vary in age, income, location, identity, lifestyle, motivations, purchase situations, digital behaviour, cultural context, and expected value. Treating everyone as one undifferentiated audience can produce vague positioning, inefficient media spending, weak product design, and exclusionary communication.

Segmentation, persona development and targeting convert broad market information into actionable decisions. Segmentation identifies groups with meaningful similarities and differences. Personas translate evidence about a selected group into a vivid but disciplined representation of a typical user or buyer. Targeting evaluates which segments the organisation should prioritise, considering attractiveness, strategic fit, feasibility, profitability, risk and ethics.

The central managerial challenge is balance. Segments must be sufficiently different to justify different marketing actions, sufficiently large or valuable to support investment, and sufficiently reachable for the organisation to serve. They must also avoid reducing people to stereotypes. A strong segmentation system therefore combines quantitative evidence, qualitative understanding, business judgement and ethical governance.

Learning Objectives

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

1. Explain the purpose and logic of market segmentation.
2. Distinguish demographic, geographic, psychographic, behavioural, needs-based and value-based variables.
3. Construct actionable segments using multiple variables.
4. Create evidence-based personas without confusing them with stereotypes.
5. Evaluate segment attractiveness, strategic fit and organisational capability.
6. Recognise bias, privacy, discrimination and representation risks in targeting.
7. Apply a structured segmentation and targeting process to a brand-management problem.

1. The Logic of Segmentation

Segmentation is the process of dividing a broader market into relatively distinct groups whose members share relevant characteristics, needs or responses. It is not simply the classification of people. Its purpose is to improve decisions about value propositions, products, pricing, distribution, communication and customer experience.

A useful distinction is between a descriptive variable and a strategic variable. Age, household size and location describe consumers. They become strategically useful only when they help explain a difference in needs, behaviour or response. For example, age may correlate with media use in one market but be a poor predictor of skincare needs in another. Similarly, income may indicate purchasing capacity but not willingness to pay.

Segmentation supports three linked questions:

• Who is different? This is the identification question.
• Different in what way? This is the needs, behaviour or value question.
• What should the organisation do differently? This is the action question.

A segment is therefore useful only when a difference can lead to a meaningful managerial response. If two groups have different ages but require the same product, respond to the same message and generate comparable value, separating them may add complexity without improving performance.

Segmentation, Targeting and Positioning

Segmentation identifies possible groups. Targeting selects the groups the organisation will prioritise. Positioning defines the distinctive value and meaning the brand intends to occupy in the minds of those selected consumers. These activities are analytically connected but not identical.

A brand may identify six segments but target two. It may target one segment with a premium proposition and another with a simplified proposition. Alternatively, it may use one core positioning while adapting execution for different contexts. Targeting should not automatically mean excluding everyone else; it means allocating priority, resources and design attention.

2. Major Segmentation Variables

2.1 Demographic Variables

Demographic segmentation uses observable population characteristics such as age, gender, household composition, education, occupation, income and life stage. These variables are widely available and useful for market sizing, media planning and distribution decisions.

Demographics are especially useful when they have a plausible connection with consumption. Household composition may affect food-pack sizes, housing services and insurance needs. Life stage may affect financial products, travel and childcare. Income may affect affordability and payment preferences.

However, demographics are often overused. They are proxies, not explanations. A demographic group can contain substantial differences in attitudes and behaviour. Gender, for example, should not be treated as a reliable substitute for identity, need or product usage. Responsible practice uses demographic variables where relevant and tests whether they actually predict the outcome of interest.

2.2 Geographic Variables

Geographic segmentation divides markets by country, region, city, neighbourhood, climate, urbanisation or travel distance. It is valuable when availability, culture, regulation, infrastructure or climate affects demand.

Geography can also operate as a proxy for social conditions. A retailer may identify differences in shopping missions across neighbourhoods, but it must avoid assuming that all residents share one lifestyle. Local research and community consultation are important when geography intersects with ethnicity, deprivation or political sensitivity.

2.3 Psychographic Variables

Psychographic segmentation concerns values, attitudes, interests, personality, identity, aspirations and lifestyles. It attempts to explain why consumers make choices rather than merely describing who they are.

A psychographic segment might consist of consumers who value convenience and control, consumers who seek novelty and social recognition, or consumers who prioritise environmental responsibility and durability. These descriptions are more useful than labels such as “young professionals” when they are supported by research and connected to a purchase decision.

Psychographic measures require careful design. Respondents may express socially desirable attitudes that do not predict behaviour. Researchers should compare stated values with observed actions, purchase records, interviews and contextual evidence.

2.4 Behavioural Variables

Behavioural segmentation uses what consumers do. Common variables include:

• usage frequency and volume;
• purchase recency and monetary value;
• category involvement;
• brand loyalty and switching;
• benefits sought;
• purchase occasion;
• channel preference;
• response to promotion;
• digital engagement; and
• adoption stage.

Behaviour is often closer to commercial outcomes than demographics. A brand can distinguish heavy users, occasional users, lapsed users, trial users and non-users. A digital service may distinguish consumers who browse extensively from those who complete purchases quickly.

Behavioural data also has limitations. Past behaviour may reflect previous availability, pricing or habit rather than future potential. A consumer who has never bought a product may be highly attractive if a barrier is removed. Conversely, high frequency may be unprofitable if it depends on costly discounts.

2.5 Needs-Based Variables

Needs-based segmentation groups consumers according to the problem they are trying to solve or the benefit they seek. It is often one of the strongest foundations for brand strategy because it links consumer insight to value creation.

For example, users of a meal-delivery service may seek:

• speed on busy weekdays;
• predictable nutrition;
• family variety;
• low cost; or
• exploration of new cuisines.

These consumers may differ demographically but share a relevant consumption problem. Needs-based segments should be investigated through interviews, observation, diary studies, customer-service data, reviews, surveys and usage research.

A need is not always an articulated preference. Consumers may say they want “more choice” but actually need confidence that selecting an option will be easy. Insight work must identify functional, emotional and social dimensions of the problem.

2.6 Value-Based Variables

Value-based segmentation evaluates the economic and strategic value of customers to the organisation. Variables can include contribution margin, retention, lifetime potential, cost to serve, referral behaviour and strategic influence.

A simple conceptual expression for customer lifetime value is:

Here, R_t is revenue from the customer in period t, C_t is variable cost to serve in that period, p_t is the probability of retention or survival to that period, d is the discount rate, T is the time horizon, and A is acquisition cost. Monetary quantities must use the same currency and time basis. This is a planning model, not a precise prediction. It assumes that the organisation can estimate contribution, retention and the relevant time horizon; it should be adapted for referrals, uncertainty, changing costs and differences between accounting profit and cash flow.

Value-based segmentation should not become a justification for denying essential service to less profitable customers. In regulated or public-interest contexts, fairness, universal access and duty of care may outweigh commercial value. Value is therefore one criterion among several, not an ethical ranking of human worth.

3. Designing Actionable Segments

A practical segmentation project usually follows the stages below.

Stage 1: Define the Decision Problem

Begin with the decision the segmentation must support. “Understand our customers” is too broad. Better questions include: Which users should the new subscription target? Why are first-time buyers not returning? Which needs should guide a product redesign? What audience should receive a premium proposition?

The decision determines the unit of analysis. A buyer, user, household, account, visit or purchase occasion may be the appropriate unit. In business-to-business markets, the buying centre may include users, technical evaluators, procurement officers and economic decision-makers.

Stage 2: Assemble Evidence

Combine multiple forms of evidence:

• customer and transaction data;
• market and category research;
• surveys and usage-and-attitude studies;
• interviews, ethnography and observation;
• social listening and review analysis;
• customer-service interactions;
• channel and media data; and
• competitor and environmental analysis.

Each source has limitations. Transaction data shows behaviour but often lacks motives. Interviews reveal meaning but use small samples. Surveys provide scale but may suffer from recall and response bias. Triangulation improves confidence.

Stage 3: Select Relevant Variables

Variables should be selected because they explain differences that matter to the decision. Avoid collecting every possible characteristic. A useful test is: if this variable differs between groups, will the product, message, channel, price or experience change?

Stage 4: Identify Patterns

Qualitative analysis may reveal recurring needs, tensions and contexts. Quantitative analysis may use cross-tabulation, clustering or latent-class methods. Advanced statistical techniques require appropriate sample size, valid measurement, treatment of missing data and validation on separate data where possible. A mathematical cluster is not automatically a meaningful market segment.

Stage 5: Describe and Name Segments

A segment description should include its defining need, context, behaviour, barriers, desired outcome, economic implications and reachable channels. Names should be neutral and memorable. “Confidence-seeking first-time investors” is more informative than “Segment 3.”

Stage 6: Validate and Activate

Segments should be tested for stability, distinctiveness and actionability. Pilot campaigns, concept tests, customer interviews and sales-team feedback can expose weak assumptions. A segment is not complete until the organisation knows what it will do differently.

Criteria for Effective Segments

4. Personas: Turning Evidence into Usable Understanding

A persona is a research-based representation of a meaningful user or buyer pattern. It is not a real individual and should not be presented as a statistically exact average. Its purpose is to help teams remember the consumer context when making decisions.

A robust persona normally includes:

• a descriptive name and role;
• relevant context and life situation;
• goals and desired outcomes;
• functional, emotional and social needs;
• behaviours and decision journey;
• barriers, anxieties and objections;
• triggers and information sources;
• current alternatives and workarounds;
• accessibility or service requirements;
• evidence base and confidence level; and
• implications for product, experience and communication.

A persona should state what is known, inferred and uncertain. For example, “interviews suggest that delivery reliability is a major concern” is stronger than an unsupported quotation. A memorable quote may be used only when it is genuine, anonymised and representative of the evidence.

Persona Construction Process

1. Define the segment boundary. Specify which consumers the persona represents and who it does not represent.
2. Synthesize evidence. Compare interviews, behavioural data, surveys and service records.
3. Identify patterns and tensions. Look for differences between stated intention and actual action.
4. Write the decision journey. Describe trigger, consideration, purchase, use and post-purchase stages.
5. Test with stakeholders and consumers. Ask whether the persona is recognisable without becoming stereotypical.
6. Link to decisions. Identify product, content, channel, service and measurement implications.
7. Review regularly. Personas can become obsolete as technology, culture and category conditions change.

A persona is useful when it changes a decision. If a team can read it and still make the same product and communication choices, it is probably decorative rather than strategic.

Personas and Customer Journeys

Personas should be connected to journeys because needs change across stages. A consumer may seek reassurance during research, simplicity at checkout and support after purchase. The same person may also occupy different personas in different contexts. A professional buyer can be a time-pressed user, a risk-conscious evaluator and a budget-accountable approver.

5. Evaluating Target Attractiveness

Target selection combines external attractiveness with internal fit. Common criteria include market size, growth, profitability, competitive intensity, accessibility, unmet need, strategic fit, capability requirements and risk.

A useful decision matrix scores each candidate segment against agreed criteria. Scores should not create false precision. They structure discussion and make assumptions visible.

For example, a weighted score can be represented as:

Here, S_j is the total score for segment j, w_i is the weight assigned to criterion i, and r_ij is the rating of segment j on that criterion. If weights sum to 1 and ratings use a common scale, the score is a weighted average on that scale. This method assumes criteria can be meaningfully compared and that the ratings are sufficiently reliable. It should be supplemented by threshold tests; a segment with an unacceptable legal or ethical risk should not be rescued by a high commercial score.

Illustrative criteria might be:

The calculation for Segment A is (0.25×5)+(0.25×4)+(0.15×3)+(0.20×5)+(0.15×4)=4.25. The result supports prioritising Segment A in this illustration, but management should test how the conclusion changes if the assumptions or weights change.

Targeting Strategies

• Undifferentiated targeting: one broad proposition for a mass market. It can be efficient where needs are similar but may be weak in diverse markets.
• Differentiated targeting: distinct propositions or executions for multiple segments. It can improve relevance but increases complexity and cost.
• Concentrated targeting: substantial focus on one or a few segments. It can support specialist expertise but increases dependence on those segments.
• Micromarketing or personalisation: adaptation to individuals or very small groups. It may improve relevance but creates data, privacy, operational and fairness risks.

The appropriate strategy depends on resources, category structure, brand meaning, operational flexibility and the consequences of inconsistency across touchpoints.

6. Bias, Inclusion and Governance

Segmentation can reproduce social inequality when data and assumptions reflect historical discrimination. Bias may enter through sampling, measurement, labels, algorithms, proxy variables, exclusion rules, creative representation or unequal access to offers.

Important risks include:

• Representation bias: the research sample excludes relevant populations.
• Measurement bias: a survey or dataset measures one cultural interpretation of a concept.
• Proxy discrimination: a seemingly neutral variable, such as postcode, reproduces protected-group disadvantage.
• Allocation bias: high-value customers receive better treatment in ways that deny fair access to essential services.
• Stereotyping: personas reduce complex people to simplistic cultural or demographic traits.
• Privacy risk: data is collected or combined without valid consent, transparency or appropriate security.
• Manipulation risk: targeting exploits vulnerability, urgency, addiction or limited financial literacy.

Good governance requires a clear purpose, data minimisation, lawful and transparent collection, access controls, retention limits, human review and documented decision rules. Organisations must follow applicable privacy, consumer-protection, equality and advertising laws. Legal compliance is a minimum standard; ethical marketing also asks whether a practice is respectful, explainable and socially defensible.

Teams should conduct an inclusion review before launch. Questions include: Who is missing from the evidence? Could the targeting rule disadvantage a protected group? Is the message accessible? Are prices, credit terms or recommendations appropriate for vulnerable consumers? Can a person challenge or correct an automated decision?

7. Technology and AI in Segmentation

Customer-data platforms, predictive analytics and artificial intelligence can identify patterns across transactions, content engagement, service interactions and location. Machine learning may support propensity modelling, recommendation and churn prediction. These tools can improve scale and speed, but they do not eliminate the need for theory, judgement or accountability.

AI-generated segments may be statistically coherent but strategically meaningless. They can also amplify historical patterns, infer sensitive characteristics, or produce unequal outcomes. Managers should require:

1. a clear business and consumer-benefit purpose;
2. documented data provenance and variable definitions;
3. testing for accuracy, stability and disparate impact;
4. human review of consequential decisions;
5. explainable rules where practical;
6. privacy and security controls; and
7. ongoing monitoring after deployment.

Personalisation should be proportionate to the context. A recommendation for entertainment is not equivalent to automated decisions about insurance, credit, employment or healthcare. The higher the potential harm, the stronger the governance and human oversight required.

8. Illustrative Case: TerraSip Reusable Beverage System

Background

TerraSip, a fictional regional beverage brand, sells reusable bottles and refill subscriptions. Initial research showed strong environmental concern but weak repeat purchase. Management was considering a single campaign aimed at “eco-conscious adults.”

Problem

The label was broad and potentially stereotypical. It did not explain why consumers purchased, what prevented refilling, or whether environmental concern translated into willingness to pay. TerraSip needed a target strategy for a new insulated bottle and digital refill programme.

Analysis

The team combined 1,200 survey responses, purchase and subscription records, 24 interviews, customer-service logs and two weeks of usage diaries. Four patterns emerged:

1. Routine optimisers wanted reliable hydration and easy refills; sustainability was a supporting benefit.
2. Impact advocates valued waste reduction and product durability and were willing to share brand stories.
3. Style-led adopters wanted attractive design and social visibility but were sensitive to price.
4. Occasional experimenters liked the idea but encountered inconvenience and forgot to carry the bottle.

The groups differed more by need and behaviour than by age. The research also showed that “environmentally concerned” respondents frequently abandoned refills when locations were unclear.

Persona

The primary persona was Maya, the routine optimiser. Maya works across several locations, carries water inconsistently, values products that reduce daily friction and is willing to pay more when reliability is clear. Her main barrier is not lack of environmental concern; it is uncertainty about where and how quickly she can refill. She compares practical reviews, notices battery-free or low-maintenance features and responds to reminders only when they are useful rather than intrusive.

This persona was based on repeated interview and diary patterns, not on one individual. The team recorded medium confidence for motivations and high confidence for the refill-location barrier because it appeared in service and usage data.

Target Evaluation

Impact advocates had strong brand-affinity potential, but the primary growth opportunity was routine optimisers because they had a frequent use case, an identifiable service barrier and strong fit with TerraSip’s operational capability. Style-led adopters were selected as a secondary communication audience, not as the main product-design segment.

Approach and Solution

TerraSip introduced a bottle with a simple refill-location feature, clearer availability information, workplace refill partnerships and messaging focused on dependable everyday use. Environmental benefits were communicated as proof of value rather than as a moral test. The company tested reminders with explicit consent and offered an easy opt-out.

Outcome

In the pilot region, repeat refill participation increased, customer-service complaints about locating refill points declined, and qualitative feedback indicated improved understanding of the product’s practical benefit. The case does not claim that segmentation alone caused the results; pricing, distribution and product changes were also involved. The pilot provided stronger evidence for scaling than a broad “eco-conscious adults” campaign would have provided.

Learning Points

• A broad attitude is not necessarily an actionable segment.
• Needs, context and barriers can reveal opportunities hidden by demographics.
• A persona should guide product and service design, not only advertising.
• Target selection requires capability and operational fit.
• Sustainability communication is stronger when connected to a concrete consumer benefit.
• Pilot testing is necessary before treating a segment hypothesis as fact.

9. Best-Practice Checklist

Before approving a segmentation and targeting plan, managers should confirm that:

• the decision problem is explicit;
• the unit of analysis is appropriate;
• evidence includes both behaviour and motivation;
• variables explain meaningful differences;
• segments are distinct but not artificially precise;
• each segment can be reached and served;
• personas state evidence and uncertainty;
• target criteria include strategic fit and capability;
• financial value is not confused with human worth;
• privacy, inclusion and legal risks have been assessed;
• AI or automation is explainable and monitored where used; and
• performance measures are defined for both commercial and consumer outcomes.

Chapter Summary

Segmentation divides a heterogeneous market into groups that differ in ways relevant to managerial action. Demographic and geographic variables help describe markets, while psychographic, behavioural, needs-based and value-based variables provide deeper explanations of motivation, action and commercial contribution. Effective segmentation begins with a clear decision problem, uses triangulated evidence, applies relevant variables and validates whether each segment is identifiable, substantial, distinctive, reachable, actionable, monitorable and ethical.

Personas convert evidence into practical representations of consumer situations, goals, barriers and journeys. They should be bounded, evidence-based and linked to decisions rather than built from stereotypes. Targeting evaluates potential segments according to need, commercial potential, reachability, strategic fit, capability, risk and ethics. Weighted matrices can structure judgement but should not conceal uncertainty or override legal and ethical thresholds.

Technology and AI can support segmentation at scale, but data quality, privacy, explainability, bias and human accountability remain essential. The best brand decisions integrate consumer understanding with operational capability and responsible governance.

Key Terms

Segmentation: Dividing a broad market into groups with relevant similarities and differences.

Targeting: Selecting and prioritising segments for organisational attention and resources.

Positioning: Defining the distinctive value and meaning a brand seeks to occupy in consumers’ minds.

Demographic segmentation: Grouping consumers by characteristics such as age, income or household structure.

Geographic segmentation: Grouping consumers by location or environmental context.

Psychographic segmentation: Grouping consumers by values, attitudes, lifestyles, interests or identity.

Behavioural segmentation: Grouping consumers according to actions, usage, purchase or response patterns.

Needs-based segmentation: Grouping consumers by the problems, benefits or outcomes they seek.

Value-based segmentation: Grouping or prioritising customers according to economic and strategic contribution.

Actionability: The extent to which a segment supports a meaningful managerial intervention.

Reachability: The extent to which a segment can be accessed through channels and touchpoints.

Persona: An evidence-based representation of a meaningful consumer or buyer pattern.

Customer journey: The sequence of experiences and decisions across pre-purchase, purchase and post-purchase stages.

Customer lifetime value: The estimated present value of future customer contribution, adjusted for costs, retention and acquisition.

Proxy variable: A variable used as an indirect substitute for another characteristic.

Representation bias: Distortion caused by the under-representation or exclusion of relevant groups.

Personalisation: Adapting content, offers or experiences to individuals or small groups.

Triangulation: Combining different data sources or methods to improve confidence in an insight.

Strategic fit: The alignment between a segment’s requirements and an organisation’s objectives and capabilities.

Review Questions

5 Short Answer Questions

1. What is the difference between segmentation and targeting?
2. Why should demographic variables not automatically be treated as explanations of consumer behaviour?
3. Identify three components that should appear in an evidence-based persona.
4. What is meant by a needs-based segment?
5. Name two ethical risks associated with AI-supported targeting.

5 Long Answer Questions

1. Explain how demographic, psychographic and behavioural variables can be combined to create actionable segments.
2. Describe a complete process for developing and validating consumer personas.
3. Discuss the criteria that managers should use when evaluating target attractiveness.
4. Explain why customer lifetime value can support targeting decisions but should not determine them alone.
5. Evaluate the advantages and limitations of differentiated targeting compared with concentrated targeting.

3 Analytical / Application Questions

1. A mobile banking brand has high trial but low monthly use. Design a research-led segmentation approach that identifies needs, barriers and behavioural differences. Specify the data you would collect and how the findings could change targeting.
2. Using the weighted-score method described in this chapter, construct a target evaluation matrix for three segments of a subscription service. Choose and justify at least four criteria, assign weights, calculate scores and discuss the limitations of your conclusion.
3. Audit an existing marketing persona for possible stereotyping, proxy discrimination and privacy risks. Propose revised evidence requirements and governance controls before the persona is used in an AI-personalisation programme.

Multiple Choice Questions with Answers

Question 1

Which feature makes a market segment actionable?

A. It has a memorable name

B. It supports a meaningful managerial intervention

C. It contains consumers of the same age

D. It is the largest possible group

Correct Answer: B. It supports a meaningful managerial intervention

Question 2

Which segmentation approach groups consumers according to the outcome or problem they seek?

A. Geographic segmentation

B. Demographic segmentation

C. Needs-based segmentation

D. Channel segmentation

Correct Answer: C. Needs-based segmentation

Question 3

What is the primary purpose of a persona?

A. To represent the exact statistical average of every customer

B. To replace all quantitative research

C. To provide an evidence-based representation that guides decisions

D. To classify consumers by income only

Correct Answer: C. To provide an evidence-based representation that guides decisions

Question 4

Which statement about behavioural segmentation is most accurate?

A. It uses only stated attitudes

B. It groups consumers according to actions such as usage or purchase

C. It is always superior to needs-based segmentation

D. It excludes digital activity

Correct Answer: B. It groups consumers according to actions such as usage or purchase

Question 5

In the weighted-score formula, what does w_i represent?

A. The rating of segment j

B. The number of segments

C. The weight assigned to criterion i

D. The customer retention probability

Correct Answer: C. The weight assigned to criterion i

Question 6

A persona based on one vivid customer interview, without comparison with other evidence, is most likely to suffer from which problem?

A. Triangulation

B. Overgeneralisation

C. Reachability

D. Market growth

Correct Answer: B. Overgeneralisation

Question 7

Which is an example of proxy discrimination?

A. Using a validated need measure

B. Using a postcode variable that reproduces disadvantage associated with a protected group

C. Allowing customers to opt out of reminders

D. Testing a product concept with users

Correct Answer: B. Using a postcode variable that reproduces disadvantage associated with a protected group

Question 8

Why should customer lifetime value not be the sole basis for targeting?

A. It cannot involve costs

B. It is unrelated to retention

C. Ethical, legal, access and strategic considerations may also apply

D. It is always measured in physical units

Correct Answer: C. Ethical, legal, access and strategic considerations may also apply

Question 9

In the TerraSip case, why were routine optimisers selected as the primary growth opportunity?

A. They were defined only by age

B. They had a frequent use case and a barrier the firm could address

C. They were the most environmentally vocal group

D. They required no product changes

Correct Answer: B. They had a frequent use case and a barrier the firm could address

Question 10

Which practice best supports responsible AI-based segmentation?

A. Using every available data field

B. Removing human review from consequential decisions

C. Documenting data provenance and testing for disparate impact

D. Inferring sensitive traits whenever possible

Correct Answer: C. Documenting data provenance and testing for disparate impact