Segmentation, Targeting and Customer Insight

Introduction

Brand strategy begins with a deceptively difficult question: for whom is the brand creating value? Markets are rarely homogeneous. Customers differ in their resources, identities, situations, expectations, behaviour and desired outcomes. A brand that attempts to appeal to everyone often becomes indistinct, while a brand that understands a specific group can develop sharper value propositions, more relevant communications and stronger customer relationships.

Segmentation, targeting and positioning (STP) provide a structured way to connect market diversity with brand strategy. Segmentation divides a broad market into meaningful groups; targeting evaluates those groups and selects one or more for strategic attention; positioning defines the distinctive value and meaning the brand should own in the target customer’s mind. Customer insight supplies the evidence that makes these decisions credible.

This chapter focuses on four major segmentation approaches—demographic, psychographic, behavioural and needs-based segmentation—and then examines segment attractiveness, target-market selection, personas and jobs-to-be-done (JTBD) analysis. The central principle is that a segment is not valuable merely because it can be described. It is valuable when it represents a sufficiently distinctive, reachable and commercially or strategically important group whose needs the organisation can serve better than alternatives.

Learning objectives

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

1. Explain the role of segmentation, targeting and customer insight in brand strategy.
2. Distinguish demographic, psychographic, behavioural and needs-based segmentation.
3. Assess segment attractiveness using structured qualitative and quantitative criteria.
4. Select an appropriate targeting strategy and justify it.
5. Construct evidence-based personas without confusing stereotypes with insight.
6. Apply the jobs-to-be-done framework to understand customer choice.
7. Identify ethical, methodological and technological risks in customer research and targeting.

1. The logic of STP

Segmentation

Segmentation is the process of identifying groups of customers who share relevant characteristics or problems and who may respond differently to a marketing strategy. Effective segments have within-group similarity and between-group difference in characteristics that matter to the brand decision. Age alone, for example, may not explain why people purchase a product; a shared need for convenience may be more strategically useful.

Segmentation is not the same as classification. Classification describes people; segmentation supports a decision. A useful segment should change the way the organisation designs its offer, sets its price, distributes it or communicates it.

Targeting

Targeting is the selection of segments to serve. A firm may choose one segment, several segments, or a broad market. Targeting requires a comparison between external opportunity and internal capability. A large segment may be unattractive if it is dominated by powerful competitors, difficult to reach or poorly matched with the firm’s resources.

Positioning

Positioning is the deliberate effort to establish a distinctive and valuable meaning for the brand in the target market. Segmentation and targeting answer “who?” Positioning answers “why should this group choose us rather than an alternative?” Positioning will be examined in greater detail in later chapters, but insight from this chapter provides its foundation.

From data to decision

The STP process should be treated as iterative rather than purely linear:

market and customer data → possible segments → segment evaluation → target selection → value proposition and positioning → market response → learning and refinement

A segment definition should therefore be treated as a strategic hypothesis. It must be tested against actual behaviour, customer language, commercial outcomes and competitive responses.

2. Bases of segmentation

2.1 Demographic segmentation

Demographic segmentation uses observable or reported population characteristics such as age, gender identity, income, occupation, education, household composition, life stage and nationality. Business-to-business markets may use firm size, industry, location, revenue or procurement structure.

Demographics are useful because they are often relatively easy to measure and purchase data may be available. A student housing brand, for example, may need to understand age, university location and household status. However, demographics are usually descriptive rather than explanatory. Two people with the same income and age can have entirely different motivations, values and behaviour.

Good practice is to use demographic variables as context or as a practical proxy only when evidence shows that they relate to a relevant outcome. Avoid statements such as “all older consumers dislike technology” or “all high-income consumers prefer premium products.” Such claims convert averages into stereotypes.

2.2 Psychographic segmentation

Psychographic segmentation considers values, attitudes, interests, lifestyles, identities, personality orientations and opinions. A psychographic segment might consist of customers who value environmental responsibility, seek status through design, prioritise personal independence or enjoy experimentation.

Psychographics can explain why customers with similar demographics behave differently. They are especially valuable for brands whose meaning, identity or community is part of the purchase decision. The main challenge is measurement. Attitudes stated in a survey may not predict behaviour, and broad labels such as “ethical consumer” can hide substantial variation.

Psychographic research should therefore combine self-reported attitudes with observed behaviour, purchase history, qualitative interviews or experiments. Researchers should also distinguish a customer’s personal value from the specific situations in which that value influences choice.

2.3 Behavioural segmentation

Behavioural segmentation groups customers according to what they do. Common variables include purchase frequency, recency, monetary value, usage occasion, channel preference, loyalty, response to promotions, category involvement and stage in the decision process.

A retailer may distinguish first-time buyers, occasional buyers, frequent buyers and lapsed customers. A software company may distinguish users who explore advanced functions from those who use only basic features. Behavioural segments are often actionable because communication, service and retention activities can be adapted to observed conduct.

A commonly used customer-value approach is RFM analysis: recency, frequency and monetary value. It is a managerial classification tool rather than a universal theory of customer value. It can identify commercially important customers, but it may undervalue new customers, future potential, referrals and customers who purchase infrequently but strategically.

2.4 Needs-based segmentation

Needs-based segmentation groups customers according to the outcomes they seek or the problems they are trying to solve. Examples include:

• reducing the time required to prepare a meal;
• feeling confident when choosing a financial product;
• maintaining social connection while travelling;
• obtaining reliable performance in difficult conditions;
• expressing an identity through consumption.

Needs-based segmentation is often strategically powerful because it links customer understanding directly to value creation. It can reveal competitors outside the product category. Someone seeking “a quick, healthy lunch” may compare a meal-kit service not only with other meal kits but also with restaurants, supermarket food and skipping lunch.

Needs may be functional, emotional or social. Functional needs concern practical performance; emotional needs concern feelings such as reassurance or enjoyment; social needs concern how the customer wishes to be perceived or how they wish to relate to others. A strong insight often combines all three.

2.5 Combining segmentation bases

The four bases are complementary. Consider a possible segment for a cycling brand:

The combined description is more useful than any single variable. Nevertheless, adding variables does not automatically improve a segment. Excessively narrow definitions can produce segments too small to reach, measure or serve profitably. The objective is strategic relevance, not maximum complexity.

3. Segment quality and attractiveness

A segment should be assessed using both market evidence and organisational judgement. Classic criteria include the following.

Identifiability and measurability

The organisation should be able to estimate the segment’s size, characteristics, behaviour and potential value. Perfect measurement is rarely possible, but decisions should identify the evidence base and its limitations.

Substantiality

The segment must be sufficiently valuable to justify a tailored proposition. Substantiality does not necessarily mean large volume. A small segment may be attractive if it has high margins, strong growth, strategic importance or influence over other customers.

Accessibility and reachability

The organisation must be able to reach the segment through suitable media, channels, distribution and sales processes. A segment that can be described but not reached is not actionable.

Responsiveness and differentiation

Members of the segment should respond differently to a meaningful marketing intervention. If two groups respond in the same way, treating them as separate segments may add complexity without benefit.

Growth and trajectory

Current size is only one consideration. The analyst should examine adoption trends, demographic change, category development, technology, regulation and cultural change. Growth forecasts must be treated as assumptions, not facts.

Competitive intensity

A segment may attract many competitors, strong channel partners or powerful substitute offerings. Competitive intensity should be assessed alongside customer demand rather than inferred from market size alone.

Strategic fit and capability

The organisation needs relevant assets, skills, partnerships, reputation and operational capacity. A segment that requires a completely different business model may be attractive in theory but unsuitable in practice.

Economic attractiveness

Managers may estimate expected contribution using a simple planning relationship:

Here, N is the number of potential customers reached or served during the planning period, p is the expected purchase rate or conversion probability, m is contribution margin per customer after variable costs, and F is incremental fixed cost. The terms must refer to the same period, customer population and currency. This is not a complete profit model: acquisition cost, retention, referrals, cannibalisation, capacity constraints and uncertainty may need separate treatment.

For example, suppose a campaign can reach 20,000 relevant prospects, the expected conversion rate is 8%, contribution margin per new customer is £35, and incremental fixed cost is £12,000:

N × p × m = 20,000 × 0.08 × £35 = £56,000

Therefore, estimated incremental contribution is £56,000 − £12,000 = £44,000, before any additional acquisition or servicing costs not included in the assumptions.

A useful decision score can be constructed when managers need to compare alternatives:

S_j is the weighted attractiveness score for segment j, w_i is the importance weight assigned to criterion i, and r_ij is the rating of segment j on that criterion. Ratings must use the same scale, such as 1–5. The score is a decision aid, not an objective truth. Weights and ratings reflect assumptions and should be tested through sensitivity analysis.

4. Targeting strategies

Undifferentiated targeting

An organisation offers one broad proposition to the market. This may be efficient when needs are relatively similar, the product is standardised and scale economies are important. It risks weak relevance when meaningful differences exist.

Differentiated targeting

The organisation selects several segments and develops distinct propositions, products or communications for each. This can increase relevance and coverage but also raises research, operational and brand-management costs. The propositions must remain coherent with the parent brand.

Concentrated or niche targeting

The organisation focuses resources on one segment or a small number of closely related segments. Niche targeting can be effective for smaller firms or specialist brands because it permits deep expertise and distinctive positioning. Its risks include dependence on a narrow market and vulnerability to changes in that segment.

Micromarketing and personalised targeting

Micromarketing adapts an offer to local areas, communities or individuals. Digital systems make personalisation technically easier, but technical feasibility does not guarantee customer value, consent or trust. Personalisation should be proportionate, transparent and based on legitimate data practices.

Selecting among strategies

The choice depends on segment heterogeneity, resources, competitive advantage, brand architecture, legal constraints and customer expectations. Targeting should not be reduced to “choose the biggest group.” A strategically suitable target is one where the brand can create distinctive value and sustain it.

5. Customer insight and research design

A customer insight is more than an interesting observation. It is an evidence-supported interpretation that explains a customer tension, motivation or unmet need and suggests an opportunity for action. “Customers use mobile phones” is data. “Time-pressured commuters use their phones to regain a sense of control during unpredictable journeys” is a possible insight, provided research supports it.

Research methods

Secondary research includes market reports, census data, academic studies, search trends, category data, social listening and competitor analysis. It is efficient but may have been collected for another purpose.

Qualitative research includes depth interviews, observation, diary studies, ethnography, focus groups and customer journey research. It reveals language, context and meaning. Small qualitative samples are not statistically representative, but they can generate hypotheses and explain behaviour.

Quantitative research includes surveys, experiments, panel data, transaction analysis and controlled tests. It estimates prevalence and tests relationships, but results depend on sampling, measurement quality and question design.

Behavioural and digital data can reveal actual actions, sequences and friction points. However, clicks are not automatically motives, and algorithmic records may exclude people with limited digital access.

The strongest research designs often triangulate methods. For example, transaction data can identify a pattern, interviews can explain it, and a controlled test can evaluate an intervention.

From observation to insight

A disciplined insight process asks:

1. What is happening?
2. For whom and in what context?
3. What tension or barrier may explain it?
4. What evidence supports that explanation?
5. What alternative explanations exist?
6. What could the brand do differently if the insight is true?

Researchers should avoid confirmation bias, leading questions and overgeneralisation from vivid anecdotes. Customer language should be preserved where it clarifies meaning, but quotations should never substitute for analysis.

6. Personas: useful representations, not fictional stereotypes

A persona is a research-based representation of a strategically important customer type. It usually includes a name or label, context, goals, behaviours, barriers, decision criteria, channels, relevant quotations and implications for the brand.

A strong persona is:

• based on multiple sources of evidence;
• focused on behaviour and needs rather than appearance;
• specific enough to guide decisions;
• connected to a segment and purchase context;
• explicit about uncertainty;
• updated when evidence changes.

A weak persona is a decorative character created from assumptions. Labels such as “Busy Beth” or “Tech-Savvy Tom” can encourage stereotyping if they are not supported by research. Demographic details should be included only when they affect the decision or help teams understand context.

Persona template

Personas should be tested by asking whether a proposed decision would change if the persona were absent. If not, the persona may be too generic.

7. Jobs to be done

The jobs-to-be-done perspective asks what customers are trying to accomplish in a particular situation. Customers “hire” products, services or brands to make progress. The job is not identical to the product category. A customer does not necessarily want a drill; they want a hole, a completed project or confidence that the task will be done correctly.

A job statement can be structured as:

When [situation], I want to [motivation or progress], so I can [desired outcome].

For example: “When I arrive home late on a weekday, I want to prepare a nutritious meal with minimal planning, so I can eat well without sacrificing time to rest.”

Jobs may be:

• functional: complete a practical task;
• emotional: feel reassured, energised or in control;
• social: signal competence, belonging or responsibility.

JTBD analysis commonly examines the job map: define, locate, prepare, confirm, execute, monitor, modify and conclude. The precise stages vary by context. Mapping the process can reveal underserved points, such as uncertainty before purchase or difficulty maintaining a product after purchase.

JTBD is not a licence to ignore demographics or behaviour. Demographics can influence access and context; behaviour can show current solutions. JTBD adds a focus on progress and alternatives. Its limitation is that jobs can become so broad that they cease to guide strategy. A useful job must be situated in a recognisable context and connected to observable consequences.

8. Applied case: FreshRoute meal service

The following is a fictional illustrative case.

Background

FreshRoute is a regional meal-kit brand offering pre-portioned ingredients delivered to households. It initially targeted “young professionals” with digital advertising. Awareness was reasonable, but repeat purchase was weak.

Problem

Management assumed that customers wanted convenience. However, cancellation data showed several different reasons: preparation took longer than expected, recipes created unwanted leftovers, and some customers disliked deciding what to order each week. Treating all customers as one convenience segment concealed different needs.

Analysis

The company combined order records, cancellation reasons, 24 interviews, an online survey and two weeks of customer diaries. The research identified three provisional groups:

1. Explorers: enjoy cooking and seek variety; they tolerate preparation time but want novel recipes.
2. Routine optimisers: want predictable, healthy weekday meals and minimal planning.
3. Occasion hosts: purchase irregularly when entertaining and need confidence, presentation and quantity assurance.

The groups differed in behaviour and need, not simply age. Routine optimisers had the highest repeat potential, while occasion hosts had higher basket value but lower frequency.

Approach

FreshRoute selected routine optimisers as its primary target and occasion hosts as a secondary seasonal segment. It introduced a “weekday essentials” plan with fewer decisions, preparation-time guarantees and adjustable portions. For occasion hosts, it developed a separate entertaining range rather than attempting to force one proposition on everyone.

An evidence-based persona for the primary target was constructed as follows:

Maya, the weekday planner: Maya works irregular hours and wants healthy dinners, but her central problem is decision fatigue after work. She compares services by reliability, preparation time and flexibility more than by recipe novelty. Her job is: “When my workday is unpredictable, I want a dependable dinner plan requiring little mental effort, so I can maintain healthy routines without feeling organised around cooking.”

Outcome

In the first test region, the weekday plan increased repeat ordering and reduced cancellations attributed to planning difficulty. The result should not be interpreted as proof that the segment will perform identically elsewhere; the test involved one region and a limited period. FreshRoute continued monitoring contribution margin, customer satisfaction and whether the new plan attracted existing customers away from higher-margin products.

Learning points

The case illustrates five principles. First, a broad demographic label did not explain retention. Second, cancellation behaviour revealed friction but interviews explained its meaning. Third, needs-based segments supported product design. Fourth, the target was selected for strategic fit and repeat potential rather than size alone. Fifth, the persona and job statement generated specific decisions that could be tested.

9. Ethics, governance and contemporary practice

Segmentation can create value, but it can also produce exclusion, manipulation and privacy harm. Organisations should obtain data lawfully, explain relevant purposes, minimise unnecessary collection, protect sensitive information and provide appropriate choices. Legal requirements vary by jurisdiction; compliance should not be treated as a substitute for ethical judgement.

Particular care is needed with health, financial, biometric, location and inferred vulnerability data. A model may identify that a person is likely to be financially distressed even when the person has never disclosed it. Using that inference to exploit urgency would be ethically problematic and may breach regulation.

Algorithmic targeting can reproduce historical discrimination. Teams should test outcomes across relevant groups, document data sources, establish human oversight and create processes for correction and appeal. “Personalised” should not mean invisible surveillance.

Artificial intelligence can help cluster text, summarise interviews, identify behavioural patterns and generate research hypotheses. It can also hallucinate evidence, encode bias, expose confidential data or confuse correlation with motivation. AI-generated personas must never be treated as research participants. Human researchers remain responsible for sampling, interpretation, consent, validation and strategic judgement.

Best practice includes:

• define the decision before collecting data;
• combine behavioural evidence with customer explanation;
• distinguish facts, interpretations and assumptions;
• test segments for reachability and commercial viability;
• disclose uncertainty and revisit assumptions;
• avoid sensitive targeting without a compelling, legitimate purpose;
• monitor outcomes, not merely campaign efficiency;
• provide mechanisms for opting out and correcting data.

Chapter Summary

Segmentation identifies meaningful customer groups, targeting selects those worth serving, and positioning translates customer understanding into distinctive brand value. Demographic variables are accessible but often descriptive; psychographics explain values and attitudes; behavioural data records what customers do; and needs-based segmentation focuses on the progress customers seek. Combining bases can improve strategic usefulness, but unnecessary complexity should be avoided.

Segment attractiveness depends on measurability, substantiality, accessibility, responsiveness, growth, competition, strategic fit and economics. Weighted scores and contribution estimates can support decisions, but they depend on assumptions and should not replace judgement. Targeting may be broad, differentiated, concentrated or personalised, depending on market heterogeneity and organisational capability.

Customer insight requires more than data collection. Triangulation, careful interpretation and attention to context help distinguish observations from explanations. Personas should represent evidence-based patterns rather than stereotypes. Jobs-to-be-done analysis complements segmentation by asking what progress customers seek in particular situations. Ethical practice requires privacy protection, fairness, transparency, human oversight and responsible use of AI.

Key Terms

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

Targeting: Selecting one or more segments for organisational attention.

Positioning: Establishing a distinctive and valuable meaning for a brand in a target customer’s mind.

Demographic segmentation: Grouping customers by characteristics such as age, income or household status.

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

Behavioural segmentation: Grouping customers by actions such as usage, purchase frequency or loyalty.

Needs-based segmentation: Grouping customers by desired outcomes or problems to be solved.

Segment attractiveness: The degree to which a segment offers opportunity and strategic suitability.

Reachability: The extent to which a segment can be accessed through communication, distribution or sales channels.

Responsiveness: The extent to which a segment reacts differently to a marketing intervention.

Strategic fit: Compatibility between a segment’s requirements and organisational capabilities.

Customer insight: An evidence-supported explanation of a customer motivation, tension or unmet need.

Persona: A research-based representation of a strategically important customer type.

Jobs to be done: The progress a customer seeks to make in a particular situation.

Functional need: A practical task or performance outcome desired by a customer.

Emotional need: A desired feeling or psychological outcome.

Social need: A desired identity, relationship or social perception.

Triangulation: Using multiple research methods or data sources to strengthen interpretation.

RFM analysis: A behavioural approach based on recency, frequency and monetary value.

Micromarketing: Adapting marketing activities to local, community or individual-level groups.

Review Questions

Short Answer Questions

1. What is the difference between segmentation and targeting?
2. Why can demographic segmentation be insufficient as an explanation of customer behaviour?
3. Identify four criteria used to assess segment attractiveness.
4. What is an evidence-based persona?
5. Distinguish a functional job from an emotional job in JTBD analysis.

Long Answer Questions

1. Compare demographic, psychographic, behavioural and needs-based segmentation. Explain when each may be useful and identify a limitation of each.
2. Explain how an organisation could evaluate three potential target segments using both qualitative criteria and an economic estimate.
3. Discuss why customer insight should be developed through triangulation rather than through a single research method.
4. Explain the differences between concentrated, differentiated and personalised targeting strategies.
5. Evaluate the contribution and limitations of personas and jobs-to-be-done analysis in brand strategy.

Analytical / Application Questions

1. A university retailer observes that postgraduate students and undergraduate students purchase similar products but have different reasons for doing so. Design a segmentation and research plan that could determine whether they should be treated as separate target segments.
2. A streaming service has high trial rates but low renewal rates. Using behavioural, needs-based and JTBD analysis, diagnose possible causes and recommend evidence that would distinguish among them.
3. Construct a weighted segment-attractiveness scorecard for two hypothetical segments. Choose at least four criteria, assign weights totalling 1, rate each segment on a consistent scale, and explain the assumptions and limitations of your recommendation.

Multiple Choice Questions with Answers

Question 1

Which statement best defines market segmentation?

A. Selecting the largest available customer group

B. Dividing a market into groups with strategically relevant similarities

C. Designing a distinctive brand identity

D. Setting the price for a product

Correct Answer: B. Dividing a market into groups with strategically relevant similarities

Question 2

Which segmentation base focuses most directly on the outcomes customers seek?

A. Demographic

B. Psychographic

C. Behavioural

D. Needs-based

Correct Answer: D. Needs-based

Question 3

A segment is especially actionable when it is identifiable, reachable and what else?

A. Entirely homogeneous in every respect

B. Responsive to a meaningful marketing intervention

C. Defined only by age

D. Free from all competitors

Correct Answer: B. Responsive to a meaningful marketing intervention

Question 4

In the contribution example, 20,000 prospects convert at 8% and produce £35 contribution each. What is the contribution before fixed costs?

A. £5,600

B. £16,000

C. £56,000

D. £700,000

Correct Answer: C. £56,000

Question 5

What is the primary purpose of a persona?

A. To replace all market research

B. To represent a research-based customer type for strategic decisions

C. To describe an imaginary ideal customer using stereotypes

D. To calculate total market size

Correct Answer: B. To represent a research-based customer type for strategic decisions

Question 6

Which statement best expresses the jobs-to-be-done perspective?

A. Customers always buy products because of demographic similarity

B. Customers hire products or brands to make progress in a situation

C. Customers can be understood only through purchase frequency

D. Customers’ jobs are identical to product categories

Correct Answer: B. Customers hire products or brands to make progress in a situation

Question 7

What does triangulation in customer research involve?

A. Using one very large survey

B. Combining multiple methods or data sources

C. Targeting three demographic groups

D. Removing qualitative evidence

Correct Answer: B. Combining multiple methods or data sources

Question 8

A concentrated targeting strategy generally involves:

A. Serving the entire market with one proposition

B. Serving several unrelated segments with different propositions

C. Focusing resources on one segment or a small number of related segments

D. Personalising every message to every individual

Correct Answer: C. Focusing resources on one segment or a small number of related segments

Question 9

Which is an important ethical risk of algorithmic targeting?

A. It always produces statistically representative samples

B. It can reproduce discrimination in historical data

C. It eliminates the need for human oversight

D. It makes privacy regulation irrelevant

Correct Answer: B. It can reproduce discrimination in historical data

Question 10

In the weighted score formula, what must the criterion weights satisfy?

A. They must all equal 5

B. They must be negative

C. They must sum to 1

D. They must be measured in currency

Correct Answer: C. They must sum to 1