Segmentation, Targeting and Customer Insights

Introduction and Context

Entrepreneurs rarely fail because no one could possibly use their product. They more often fail because they address an imprecise market, misunderstand customer priorities, or allocate scarce resources to customers who are unlikely to buy. Segmentation, targeting and customer insight provide a disciplined way to decide whom a venture should serve, what value it should create, and how its offer should be communicated.

Segmentation divides a broad market into relatively homogeneous groups. Targeting evaluates these groups and selects one or more for focused attention. Customer insight goes further: it explains the motivations, constraints, decision processes and unarticulated needs that make a segment commercially meaningful. Together, these activities transform descriptive market data into strategic choices.

For an entrepreneurial venture, the objective is not to identify the largest possible audience. It is to identify an attractive group whose needs the venture can serve better, more economically or more distinctively than alternatives.

Learning Objectives

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

1. Explain the relationships among segmentation, targeting, positioning and customer insight.
2. Construct actionable segments using demographic, geographic, behavioral, psychographic and needs-based variables.
3. Evaluate segment attractiveness and the venture’s ability to serve a segment.
4. Develop evidence-based customer personas and distinguish them from stereotypes.
5. Apply qualitative and quantitative research to discover customer needs.
6. Recognize ethical, analytical and operational limitations in segmentation decisions.
7. Select target markets and translate insights into entrepreneurial action.

1. The Strategic Logic of STP

The STP framework consists of three connected decisions:

• Segmentation: Which meaningful groups exist within the market?
• Targeting: Which group or groups should receive priority?
• Positioning: What distinctive value should be associated with the venture in the minds of those customers?

Customer insight is the evidence base connecting the three. A segment is not simply a demographic category. “Women aged 25–34” may be easy to describe, but it does not necessarily imply common needs or purchasing behavior. A stronger segment might be “urban professionals who need nutritious meals that can be ordered and consumed within a short working lunch.” The latter suggests a problem, context, benefit and potential solution.

A useful segment should be:

• Identifiable: its members can be described and recognized.
• Measurable: its approximate size, value and behavior can be estimated.
• Substantial: it is sufficiently valuable to justify attention.
• Accessible: the venture can reach and serve its members.
• Differentiable: it responds differently from other groups to an offer or marketing action.
• Actionable: the venture can design a feasible proposition for it.
• Stable enough: its needs and boundaries are not changing so quickly that research becomes obsolete.

These criteria are not absolute tests. A small niche can be attractive to a specialized venture, while a large segment may be unattractive because competition is intense or access costs are prohibitive.

2. Bases of Market Segmentation

Demographic and firmographic segmentation

Demographic variables include age, gender, income, education, occupation, household structure and life stage. In business-to-business markets, corresponding variables are called firmographic variables and include industry, organization size, annual revenue, location, ownership and technology usage.

These variables are useful because they are often available in secondary data and can support media selection, sales planning and market sizing. However, they are usually proxies rather than explanations. Income may correlate with purchasing power, but it does not reveal whether a customer values convenience, status, environmental impact or technical performance.

Geographic segmentation

Geographic segmentation distinguishes customers by country, region, climate, urban or rural location, neighborhood, travel distance or service territory. It is particularly important when delivery costs, regulation, infrastructure or cultural context affect demand. Digital ventures may appear geographically unconstrained, but language, payment systems, time zones, logistics and legal jurisdiction still create geographic differences.

Behavioral segmentation

Behavioral segmentation uses observable actions, such as purchase frequency, spending level, product usage, channel preference, adoption stage, response to promotions and loyalty. For a subscription venture, useful groups might include trial users, habitual users, dormant users and high-value advocates.

Behavioral data are often more predictive than demographic data because they reflect actual choices. Yet behavior must be interpreted carefully. A customer who buys infrequently may be dissatisfied, may have a seasonal need, or may be highly loyal to a premium product purchased only occasionally.

Psychographic segmentation

Psychographics describe values, attitudes, lifestyles, interests, personality orientations and identity-related motivations. Two customers with similar income and age may differ substantially in their concern for sustainability, risk tolerance or desire for social recognition.

Psychographic research commonly uses interviews, observation, diary studies and structured scales. It should not be confused with unsupported labels such as “busy people” or “health-conscious consumers.” A psychographic segment becomes useful when its characteristics are linked to a decision, problem or response to an offer.

Needs-based and job-based segmentation

Needs-based segmentation groups customers according to the outcomes they seek and the problems they are trying to solve. Jobs-to-be-done thinking asks what customers are trying to accomplish in a particular situation. The “job” may be functional, emotional or social.

For example, a commuter may hire a mobility service to reach work reliably, reduce travel anxiety, and appear responsible to colleagues. A single product category can therefore contain several jobs and several segments.

Needs-based segmentation is especially valuable for innovation because it can reveal non-consumers and substitute solutions. Customers may solve a need with a competitor, a manual workaround, postponement or doing nothing. The relevant competitive set is therefore defined by the job, not only by product category.

Multi-variable segmentation

The strongest segmentation often combines variables. A venture might identify “small independent restaurants in metropolitan areas that have high delivery demand, lack data expertise, and are willing to pay for simple forecasting.” This combines firmographics, behavior, needs and willingness to adopt.

However, adding variables does not automatically improve a segment. Excessive subdivision produces segments that are too small, difficult to reach or impossible to distinguish operationally. The test is whether each additional variable improves prediction, service design or resource allocation.

3. From Data to Actionable Segments

Segmentation is an analytical process rather than a single demographic table. A practical workflow is as follows.

Step 1: Define the market and decision

Specify the product category, customer problem, geography, time period and decision to be made. “The market for education” is too broad. “Paid online preparation for postgraduate entrance examinations among final-year students in India” is more useful.

Step 2: Generate hypotheses

Use existing research, competitor analysis, expert interviews, customer complaints, search behavior and frontline observations to propose possible differences among customers. At this stage, hypotheses are provisional.

Step 3: Collect evidence

Combine qualitative and quantitative methods. Interviews explain meaning and language; surveys estimate prevalence; transaction data reveal behavior; observation identifies workarounds; experiments test response to an offer.

Step 4: Identify meaningful differences

Look for differences in needs, willingness to pay, purchase process, channel use, retention, service cost and response to benefits. A difference matters only when it changes a business decision.

Step 5: Construct and profile segments

Give each segment a descriptive name based on its defining problem or behavior, not a catchy stereotype. Describe its size, needs, triggers, barriers, alternatives, buying process and expected economics.

Step 6: Validate

Test whether members recognize the problem, whether the segment can be reached, and whether behavior differs in practice. Validation may use pilot campaigns, landing pages, prototypes, interviews or small-scale sales.

Step 7: Translate into an operating choice

Specify the product features, price logic, channels, messages, partnerships, service model and metrics associated with the chosen segment.

A segment profile should answer five questions:

1. Who belongs to the segment?
2. What situation or problem creates demand?
3. What outcome does the customer value?
4. How does the customer currently solve the problem?
5. Why can this venture serve the segment profitably and distinctively?

4. Customer Insight: Beyond Description

A customer insight is an evidence-supported interpretation that explains why a customer behaves as observed and indicates an opportunity for action. It is not merely a fact. “Forty percent of surveyed users abandon the application” is a finding. “Users abandon because the first-use process requires information they do not yet trust the venture to store” is a possible insight requiring verification.

A useful insight generally contains:

• a customer situation or tension;
• an underlying motivation or barrier;
• evidence from more than one source where possible; and
• a clear implication for value creation or communication.

Qualitative methods

Depth interviews explore experiences, language, emotions and decision criteria. Questions should invite concrete stories rather than opinions alone: “Tell me about the last time you solved this problem.”

Observation and contextual inquiry reveal what customers actually do, including workarounds they may forget or fail to report.

Diary studies capture behavior over time and are useful for recurring or episodic problems.

Focus groups are useful for exploring language and reactions, but group conformity and dominant participants can distort results. They should not be treated as reliable estimates of market size.

Customer journey mapping organizes stages such as awareness, evaluation, purchase, use, support and renewal. At each stage, the researcher can identify goals, pain points, evidence requirements and moments of abandonment.

Quantitative methods

Surveys estimate the distribution of attitudes, behaviors and stated intentions. Good survey design requires clear constructs, neutral wording, appropriate scales, screening questions and a sample relevant to the decision. Stated willingness to buy commonly exceeds actual purchase behavior, so intention data should be triangulated with observed behavior or experiments.

Customer and transaction data can reveal frequency, recency, monetary value, retention and conversion patterns. These metrics are descriptive and should not automatically be interpreted as causal. For example, a correlation between app engagement and renewal may reflect the fact that satisfied customers engage more, rather than engagement causing satisfaction.

Triangulation

Triangulation compares evidence from different methods or sources. If interviews identify a delivery-time problem, transaction data show abandonment at delivery selection, and an experiment shows improved conversion after clearer delivery commitments, confidence in the insight increases.

5. Personas as Decision Tools

A persona is a research-based representation of a strategically important customer pattern. It is not a fictional individual intended to replace evidence. A strong persona includes:

• a concise identity and context;
• relevant goals and jobs;
• triggers and desired outcomes;
• barriers, anxieties and objections;
• current alternatives and decision criteria;
• preferred channels and information sources;
• buying authority and constraints;
• representative evidence, such as quotations or behavioral data; and
• implications for product, price, promotion and distribution.

Personas should be limited in number. If a team creates ten or fifteen, prioritization becomes difficult. Each persona should correspond to a meaningful segment and should be updated when evidence changes.

Avoid demographic caricatures. A persona named “Tech-Savvy Tara” is weak if it merely lists age and hobbies. A better persona explains that Tara manages a small design studio, needs rapid project-cost visibility, distrusts complex enterprise software, and will adopt a tool if setup takes less than an hour and data can be exported.

6. Selecting Target Markets

Targeting is a strategic choice under resource constraints. A venture may use one of four broad approaches:

A segment attractiveness assessment should consider market demand, growth, competitive intensity, customer economics, access cost, fit with capabilities, strategic learning and risk. A simple weighted score can structure discussion, but it should not create false precision.

Here, S_j is the attractiveness score for segment j, w_i is the importance weight assigned to criterion i, and r_ij is the venture’s rating of segment j on criterion i. Weights should sum to 1, and ratings must use a documented scale. The score is a decision aid, not an objective measure of market truth; results depend on criterion choice, evidence quality and managerial judgment.

For example, suppose a team uses three criteria: growth weighted 0.30 and rated 4, access weighted 0.25 and rated 5, and strategic fit weighted 0.45 and rated 3. The score is:

S = (0.30 × 4) + (0.25 × 5) + (0.45 × 3) = 1.20 + 1.25 + 1.35 = 3.80

on a five-point scale. The calculation does not prove that the segment should be selected. The team must still examine competitors, unit economics, legal constraints and execution capacity.

Market size and commercial attractiveness

Entrepreneurs often distinguish total addressable market (TAM), serviceable available market (SAM) and serviceable obtainable market (SOM). TAM describes the broad theoretical demand, SAM the part compatible with the venture’s offer and geography, and SOM the realistic near-term share given resources and competition.

Market size should be built from transparent assumptions. A bottom-up estimate may use:

where N is the number of potential customers, p is average revenue per purchase, and f is expected purchases per customer per year. Units must be consistent. This estimate assumes that the customer count, price and purchase frequency are meaningful averages and that double counting is controlled. It is a scenario, not a guaranteed forecast.

Target selection should also consider the venture’s ability to win. A large segment with powerful incumbents may be less attractive than a smaller segment where the venture has a distinctive capability, trusted access or superior learning opportunity.

7. Positioning and the Value Proposition

After selecting a target, the venture develops a position: the place it seeks to occupy in the target customer’s mental and practical choice set. Positioning should clarify the target, need, category, benefit and reason to believe.

A value proposition can be expressed as:

“For [target customer] who [important situation or need], [offering] is a [category or solution] that [primary benefit], unlike [relevant alternative], because [credible support].”

This statement is a strategic hypothesis, not advertising copy. It should be tested for relevance, clarity, distinctiveness, credibility and economic feasibility.

8. Illustrative Case: FreshRoute for Independent Restaurants

Background

FreshRoute is a fictional early-stage venture developing software that helps independent restaurants forecast ingredient demand and reduce food waste. Its founders initially define the market as “all restaurants.” Their first prototype contains complex dashboards and integrations designed for large restaurant groups.

Problem

Pilot interviews show that independent restaurants are not primarily seeking sophisticated analytics. Owners describe a more immediate job: deciding tomorrow’s purchasing quantities despite uncertain demand. They rely on memory, spreadsheets and supplier advice. They fear waste, stockouts and time-consuming software setup.

Analysis

The research team combines twelve interviews, observation of four kitchens, a short survey of 86 owners and analysis of anonymized pilot data. Three provisional segments emerge:

1. Large restaurant groups with dedicated operations teams.
2. Independent urban restaurants with repeated delivery and takeaway demand.
3. Seasonal rural restaurants with highly variable traffic.

The second segment reports the strongest combination of pain, access through restaurant associations and willingness to test a simple tool. The first segment has high revenue potential but long procurement cycles and established software competitors. The third segment has a genuine problem but insufficient year-round usage for the initial subscription model.

Approach and solution

FreshRoute selects independent urban restaurants as its beachhead market. The persona is “Owner-operator Arjun”: he manages one or two locations, reviews purchasing late in the evening, uses a point-of-sale system but lacks analytical staff, and values a recommendation he can understand in under ten minutes.

The revised proposition is a low-setup forecasting service that converts existing sales data into next-day purchasing suggestions. The team removes advanced dashboards, adds explanations for each recommendation, and provides an export for supplier orders. It tests two messages: “reduce food waste” and “order confidently for tomorrow.” The second produces more qualified trials because it addresses the immediate job rather than a broad environmental aspiration.

Outcome

In a four-week pilot, participating restaurants report that the simplified workflow is easier to adopt. The founders do not claim that the pilot proves long-term profitability; they use it to validate problem relevance, adoption barriers and product direction. Further work is required on retention, forecast accuracy across cuisines and integration costs.

Learning points

The case illustrates that a target market is selected through fit, access and ability to win—not size alone. It also shows the value of needs-based segmentation, observation, a focused persona and experimentation with alternative value propositions.

9. Ethics, Governance and Limitations

Segmentation can create value, but it can also exclude, manipulate or discriminate. Important governance questions include:

• Was personal data collected with a lawful and understandable basis?
• Is the data necessary for the stated purpose?
• Are sensitive attributes being used in ways that create unfair exclusion?
• Can customers access, correct or delete relevant data where applicable?
• Are algorithmic segments explainable and monitored for disparate effects?
• Are vulnerable customers protected from exploitative targeting?

Privacy regulation varies by jurisdiction, but responsible practice generally requires purpose limitation, data minimization, security, retention controls and meaningful transparency. Sensitive inferences—such as health status, financial distress or political affiliation—require particular caution.

Algorithmic clustering can identify patterns, but it does not reveal why they exist. A model may reproduce historical discrimination or use proxy variables for protected characteristics. Human review, bias testing, documentation and an appeals process are therefore important when segmentation affects credit, access, pricing or essential services.

Other limitations include sampling bias, response bias, changing preferences, small-sample overinterpretation and the ecological fallacy—the error of assuming that an average group characteristic applies to every individual. Personas and segments should support judgment, not replace direct customer learning.

10. Technology, AI and Contemporary Practice

Customer-data platforms, web analytics, social listening and experimentation tools make segmentation more dynamic. Machine-learning methods can cluster customers, predict churn or estimate propensity to respond. These tools are useful when the organization has reliable data, a clear decision and adequate governance.

AI-generated personas or synthetic interview summaries should not be treated as primary evidence. They can help organize existing material, generate research questions or simulate alternative messages, but they may amplify stereotypes and invent plausible-sounding claims. Human researchers must trace conclusions to source data.

Contemporary practice is moving from static demographic segments toward dynamic, situation-based segmentation. A single customer may belong to different actionable groups at different moments: a first-time buyer, an urgent problem-solver and a loyal advocate are behavioral states rather than permanent identities. Ventures should therefore monitor movement between segments and design appropriate journeys rather than treating segments as fixed boxes.

Best-Practice Checklist

Before committing resources to a target segment, ask:

1. Is the segment defined by a meaningful difference in need, behavior or economics?
2. What evidence supports its boundaries and estimated size?
3. Can the venture reach decision-makers at acceptable cost?
4. What alternatives does the segment currently use?
5. Why will customers switch, adopt or remain loyal?
6. Can the venture serve the segment better or more distinctly than competitors?
7. What assumptions remain untested?
8. What metrics will indicate progress—conversion, retention, usage, margin, referral or learning?
9. Could the segmentation create unfair exclusion or privacy harm?
10. When will the segment definition and persona be reviewed?

Chapter Summary

Segmentation divides a broad market into groups that differ in meaningful and actionable ways. Demographic, geographic, behavioral, psychographic and needs-based variables each provide useful evidence, but no single basis is universally sufficient. Needs-based and job-based approaches are especially valuable for entrepreneurial innovation because they reveal the outcomes customers seek and the substitutes they use.

Targeting is the selection of segments that combine attractiveness with venture fit and ability to win. Market size, growth and revenue potential must be considered alongside access cost, competition, capabilities, risk and ethical responsibility. Weighted scoring and bottom-up market estimates can structure decisions, but they depend on assumptions and should not create false precision.

Customer insight explains the motivations and tensions behind observed behavior. Interviews, observation, surveys, transaction data, journey mapping and experiments become more credible when triangulated. Personas are concise, evidence-based representations of strategically important customer patterns; they should guide decisions rather than substitute for research.

Effective STP is iterative. Ventures form hypotheses, test them, select a focused market, develop a credible value proposition and monitor evidence as customers and contexts change. Responsible segmentation also requires privacy protection, bias monitoring, transparency and care with AI-generated analysis.

Key Terms

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

Targeting: Evaluating segments and selecting those to serve.

Positioning: The intended place of an offering in the target customer’s choice set.

Customer insight: An evidence-supported explanation of customer motivation, tension or behavior.

Demographic segmentation: Grouping by characteristics such as age, income or life stage.

Firmographic segmentation: Grouping organizations by characteristics such as industry or size.

Behavioral segmentation: Grouping by observed actions, usage or purchase patterns.

Psychographic segmentation: Grouping by values, attitudes, lifestyles or motivations.

Needs-based segmentation: Grouping by problems, desired outcomes or jobs.

Job to be done: The functional, emotional or social progress a customer seeks in a situation.

Persona: A research-based representation of an important customer pattern.

TAM: Total addressable market under broad theoretical demand.

SAM: Serviceable available market compatible with the venture’s offer and reach.

SOM: Serviceable obtainable market realistically achievable in the near term.

Value proposition: A statement of the target customer, need, benefit and reason to believe.

Triangulation: Comparing evidence from multiple methods or sources.

Beachhead market: A focused initial market used to establish traction and learning.

Proxy variable: A measure that stands in for another characteristic and may create bias.

Data minimization: Collecting only data necessary for a defined purpose.

Review Questions

Short Answer Questions

1. Distinguish segmentation from targeting.
2. Why are demographic variables often insufficient by themselves?
3. Define a customer insight and distinguish it from a descriptive finding.
4. List four characteristics of a useful persona.
5. What are TAM, SAM and SOM?

Long Answer Questions

1. Explain the major bases of segmentation and evaluate their usefulness for an entrepreneurial venture.
2. Describe a complete process for converting customer data into actionable segments.
3. Discuss how qualitative and quantitative research can be combined to produce reliable customer insight.
4. Explain the criteria a venture should use when selecting a target market.
5. Evaluate the ethical risks of algorithmic and data-driven segmentation.

Analytical / Application Questions

1. A startup selling reusable packaging has identified households, restaurants and event organizers as possible segments. Design a segmentation and targeting analysis, including the evidence you would collect and the criteria you would use.
2. Construct a research-based persona for a customer of a service you know. State which elements are supported by evidence and which remain hypotheses.
3. Using the weighted scoring equation in this chapter, create three criteria and compare two potential target segments. Explain the assumptions, limitations and strategic implications of your result.

Multiple Choice Questions with Answers

Question 1

Which statement best defines targeting?

A. Dividing a market into demographic groups

B. Selecting segments for focused strategic attention

C. Creating a fictional customer biography

D. Estimating total industry revenue

Correct Answer: B. Selecting segments for focused strategic attention

Question 2

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

A. Geographic segmentation

B. Demographic segmentation

C. Needs-based segmentation

D. Firmographic segmentation

Correct Answer: C. Needs-based segmentation

Question 3

A persona is most useful when it is:

A. Based entirely on the founder’s intuition

B. A detailed fictional story unrelated to decisions

C. A research-based representation of an important customer pattern

D. Limited to age, gender and income

Correct Answer: C. A research-based representation of an important customer pattern

Question 4

In the equation S_j = ∑(w_i × r_ij), what does w_i represent?

A. The segment’s total revenue

B. The importance weight of criterion i

C. The number of customers in segment j

D. The venture’s conversion rate

Correct Answer: B. The importance weight of criterion i

Question 5

Which research method is especially useful for observing workarounds customers may not report in interviews?

A. Contextual observation

B. A national census only

C. A pricing slogan test only

D. A demographic profile

Correct Answer: A. Contextual observation

Question 6

A segment is actionable when:

A. It has the largest possible population

B. It can be served through a feasible offering and operating plan

C. It contains customers of exactly the same age

D. It requires no additional research

Correct Answer: B. It can be served through a feasible offering and operating plan

Question 7

Which statement about TAM, SAM and SOM is correct?

A. SOM is always larger than TAM

B. SAM is the realistic near-term sales forecast

C. TAM is broad theoretical demand, while SOM is realistically obtainable demand

D. TAM and SOM are interchangeable terms

Correct Answer: C. TAM is broad theoretical demand, while SOM is realistically obtainable demand

Question 8

If growth has weight 0.30 and rating 4, access has weight 0.25 and rating 5, and fit has weight 0.45 and rating 3, the weighted score is:

A. 2.95

B. 3.25

C. 3.80

D. 4.50

Correct Answer: C. 3.80

Question 9

Why should stated willingness to buy be interpreted cautiously?

A. It always understates actual purchases

B. It may exceed actual purchase behavior

C. It cannot be collected through surveys

D. It measures only geographic location

Correct Answer: B. It may exceed actual purchase behavior

Question 10

Which is an important governance practice in data-driven segmentation?

A. Collecting every available personal attribute

B. Using sensitive inferences without disclosure

C. Applying data minimization and monitoring for unfair effects

D. Treating algorithmic clusters as objective facts

Correct Answer: C. Applying data minimization and monitoring for unfair effects