Segmentation, Targeting, Personas, and Market Sizing

Introduction

A startup rarely fails because no one could possibly use its product. More often, it fails because the founders address an overly broad market, misunderstand the customer’s problem, or spend scarce resources acquiring customers who are difficult to serve profitably. Startup marketing therefore begins not with advertising, but with disciplined market definition.

Segmentation, targeting, personas, ideal customer profiles, beachhead-market selection, and market sizing are connected decisions. Segmentation identifies meaningful differences among potential customers. Targeting selects the groups the venture will prioritize. Personas and ideal customer profiles make those groups operational for product, sales, and communication decisions. Beachhead selection identifies the initial market in which the startup can establish traction. Market sizing estimates the scale of the opportunity and tests whether it can support the venture’s ambitions.

These tools do not eliminate uncertainty. They make uncertainty explicit, testable, and manageable.

Learning Objectives

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

1. Distinguish market segmentation, targeting, positioning, personas, and ideal customer profiles.
2. Apply relevant B2C and B2B segmentation variables.
3. Use jobs-to-be-done to understand customer motivations and desired outcomes.
4. Construct evidence-based personas and ideal customer profiles.
5. Select a beachhead market using strategic and operational criteria.
6. Calculate and interpret TAM, SAM, and SOM using top-down and bottom-up methods.
7. Evaluate market attractiveness while recognizing uncertainty, bias, ethics, and data limitations.

1. From a Broad Market to a Focused Opportunity

A market is not simply everyone who might theoretically benefit from a product. It is a set of buyers and users connected by a problem, need, transaction, or category of solution. A startup’s market definition should specify:

• the customer or buying organization;
• the problem or job being addressed;
• the alternative solutions currently used;
• the geographic and regulatory context;
• the economic mechanism through which value is created and captured.

Four concepts should be distinguished.

Segmentation divides a heterogeneous market into groups whose members share relevant characteristics or responses. Targeting evaluates those groups and chooses which to serve. Positioning defines how the offering should be understood relative to alternatives. A persona is a representative, research-based depiction of a user or buyer. An ideal customer profile (ICP) describes the type of account most likely to generate value and succeed with the offering, especially in B2B markets.

A segment is useful only when differences within it are smaller, or more strategically manageable, than differences between it and other segments. Age or industry alone does not make a useful segment. A useful segmentation variable predicts a meaningful difference in needs, willingness to pay, buying behavior, acquisition cost, or service requirements.

2. Segmentation Variables in B2C Markets

Business-to-consumer markets involve individuals or households, although the user, purchaser, influencer, and payer may be different people. Common variables include the following.

Demographics are often easy to measure but can be weak predictors. Two consumers of the same age may have entirely different motivations, routines, and purchasing power. Behavioral and benefit-based variables are frequently more actionable because they connect directly to product design and marketing activity.

A startup may combine variables. For example, “urban professionals aged 25–40” is broad and descriptive, whereas “professionals who purchase prepared meals at least four times per week because they value predictable nutrition and delivery reliability” is more behaviorally meaningful.

Segmentation must also avoid discriminatory or exploitative practices. Sensitive attributes, such as health status, ethnicity, religion, or inferred financial vulnerability, require careful legal and ethical treatment. A segment should not be pursued merely because it is vulnerable or easy to manipulate.

3. Segmentation Variables in B2B Markets

B2B markets are more complex because the customer is usually an organization and purchasing involves several participants. Important variables include:

• Firmographic: industry, revenue, employee count, location, ownership, growth stage.
• Operating characteristics: technology infrastructure, production process, capacity, distribution model.
• Use case: the specific workflow, department, or business problem.
• Buying behavior: procurement process, contract length, approval requirements, preferred suppliers.
• Need intensity: urgency, severity of the problem, regulatory exposure, operational impact.
• Economic potential: budget, expected account value, expansion potential, cost to serve.
• Readiness: digital maturity, change capability, executive sponsorship, integration capacity.

B2B segmentation should distinguish at least four roles:

1. User: interacts with the product.
2. Buyer: controls or authorizes the purchase.
3. Economic decision-maker: evaluates financial consequences.
4. Influencer or gatekeeper: shapes the decision or controls access.

For example, an inventory-forecasting startup may be used by a planner, purchased by an operations director, approved by a finance executive, and constrained by an information-technology team. Treating “the business” as a single customer conceals the actual adoption process.

Segment Quality Criteria

A practical segment should be:

• Identifiable: its members can be described and located.
• Measurable: size, behavior, and economics can be estimated.
• Substantial: it is large or valuable enough to justify attention.
• Accessible: the startup can reach and serve it.
• Responsive: members react differently to a relevant marketing offer.
• Compatible: the segment fits the venture’s capabilities, ethics, and strategy.

These criteria are not absolute. A small segment may be attractive if it has high urgency, low competition, and strong referral potential.

4. Jobs-to-be-Done: Understanding the Underlying Progress

The jobs-to-be-done (JTBD) perspective asks what progress a customer is trying to make in a particular situation. Customers do not simply buy products; they “hire” solutions to accomplish functional, emotional, and social outcomes.

A job statement should describe the situation, motivation, and desired outcome without embedding a particular product. A useful structure is:

When [situation], I want to [motivation], so that [desired outcome].

For example: “When I arrive home late on a workday, I want to obtain a healthy meal with minimal decision effort, so that I can eat well without delaying sleep.” The job is not “buy a meal-delivery subscription.” That is one possible solution.

JTBD analysis commonly examines:

• Functional job: what must be accomplished?
• Emotional job: how does the customer want to feel?
• Social job: how does the customer want to be perceived?
• Struggling moment: what triggers the search for a solution?
• Current workaround: what is being used now?
• Desired outcome: what does “better” mean?
• Switching forces: what pushes the customer away from the current solution and pulls them toward a new one?

JTBD prevents a startup from defining competitors too narrowly. A digital budgeting application competes not only with other applications, but also with spreadsheets, bank alerts, memory, and doing nothing.

Jobs should be validated through interviews, observation, diary studies, transaction data, and experiments. Founders must avoid leading questions such as “Would you use our app?” Better questions investigate recent behavior: “Tell me about the last time this problem occurred. What did you do next? What did it cost in time, money, or stress?”

5. Personas and Ideal Customer Profiles

A persona translates research into a memorable decision tool. It is not a fictional character created from imagination. It should synthesize patterns from interviews, behavioral data, support records, analytics, and observed workflows.

A useful B2C persona includes:

• context and life situation;
• relevant behavior and frequency;
• primary job and desired outcomes;
• barriers, anxieties, and objections;
• current alternatives;
• triggers and information sources;
• buying criteria;
• representative quotations supported by evidence.

A persona should not become a stereotype. Details such as favorite music or invented hobbies add narrative color but may not improve decision quality. Every included attribute should answer: “What decision will this information change?”

A B2B ICP is account-level rather than individual-level. It may include:

• industry and business model;
• employee or revenue range;
• relevant workflow and pain intensity;
• technology environment;
• buying authority and procurement complexity;
• expected annual value;
• implementation requirements;
• likelihood of retention, expansion, and referral.

The ICP differs from a buyer persona. The ICP says which organizations are attractive; buyer personas describe the people within those organizations who influence adoption and purchase.

Evidence-Based Persona Process

1. Define the decision the persona must support.
2. Collect evidence from multiple sources.
3. Identify repeated patterns rather than exceptional stories.
4. Separate observations from interpretations.
5. Create a small number of distinct personas.
6. Test them against new interviews and behavior.
7. Retire or revise them when evidence changes.

Personas can cause harm when they encode stereotypes, use personal data without consent, or obscure important diversity within a group. They should guide empathy and prioritization, not justify exclusion without evidence.

6. Targeting and Beachhead-Market Selection

Targeting is a resource-allocation decision. A startup should not ask only, “How large is this segment?” It should ask whether the segment is sufficiently attractive and reachable given the venture’s current capabilities.

A beachhead market is the initial, narrowly defined market in which a startup seeks repeatable traction before expanding. It is not necessarily the largest market. A strong beachhead often has:

• acute and frequent pain;
• a clear trigger for purchase;
• identifiable buyers and reachable channels;
• willingness and ability to pay;
• manageable implementation requirements;
• limited competitive intensity or a distinctive opening;
• potential for references, learning, and adjacent expansion.

A founder can compare candidate segments using a weighted score. This is a decision aid, not an objective truth. Let segment score S_j be:

where w_i is the importance weight for criterion i, r_ij is the rating of segment j on criterion i, and the weights sum to 1. Ratings must use a consistent scale, such as 1 to 5. The result is meaningful only if criteria are clearly defined and scores are supported by evidence.

For example, if urgency has weight 0.30 and rating 5, accessibility weight 0.25 and rating 4, willingness to pay weight 0.25 and rating 4, and expansion potential weight 0.20 and rating 3, then:

S = (0.30×5) + (0.25×4) + (0.25×4) + (0.20×3) = 4.10

The score should be accompanied by confidence levels and a sensitivity analysis. A segment that ranks first only because of uncertain assumptions deserves validation before major investment.

7. Market Sizing: TAM, SAM, and SOM

Market sizing estimates the economic scale of a defined opportunity. It is not a prediction of revenue. The three common layers are:

• TAM (Total Addressable Market): total annual demand or revenue if the relevant market were fully served under a defined pricing and scope assumption.
• SAM (Serviceable Available Market): the portion of TAM the offering can serve given geography, category, capabilities, regulation, and channel constraints.
• SOM (Serviceable Obtainable Market): the realistic portion of SAM the startup can capture within a specified period, given competition, resources, and execution capacity.

The relationship is usually:

The inequality is conceptual, not a guarantee. Each figure must use the same time period, currency, customer definition, and revenue basis.

Top-Down Sizing

A top-down estimate begins with an external market figure and applies transparent filters:

where g is the relevant geographic proportion, c is the proportion within the applicable customer or category scope, and p is the proportion compatible with the product’s price or service level. This approach is fast but inherits the quality and definitions of the source data.

Bottom-Up Sizing

A bottom-up estimate builds from observable units:

where N is the number of relevant customers or accounts, f is average purchase frequency per year, and a is average revenue per purchase. Units must be consistent: customers × purchases/customer/year × currency/purchase = currency/year.

For a B2B subscription product, an alternative is:

where ARPA is annual revenue per account. If the product has multiple tiers, ARPA should be a weighted average based on the expected customer mix, not simply the highest listed price.

Worked Example

Suppose a startup sells scheduling software to independent dental clinics in one country. Industry research identifies 20,000 clinics. The startup’s target geography contains 40% of them, 60% use compatible cloud systems, and the annual subscription is $1,200 per clinic.

A top-down SAM estimate is:

SAM = 20,000 × 0.40 × 0.60 × $1,200 = $5,760,000 per year

Suppose the startup has a five-person sales team, expects to close 12 new clinics per salesperson per quarter, and retain an existing base. The first-year new-account capacity is:

5 × 12 × 4 = 240 accounts

If all 240 accounts are billed for a full year, a simplified first-year obtainable revenue estimate is:

SOM = 240 × $1,200 = $288,000

In practice, first-year revenue may be lower because customers sign throughout the year. If, on average, new customers are active for six months in the first year, recognized first-year subscription revenue is approximately:

240 × $1,200 × 0.5 = $144,000

The $288,000 figure represents annualized run-rate revenue from the acquired accounts, while $144,000 approximates first-year recognized revenue under the stated timing assumption. These are different measures and should not be conflated.

Common Market-Sizing Errors

1. Treating a broad industry forecast as the startup’s market.
2. Counting users when revenue comes from paying organizations, or vice versa.
3. Multiplying population by price without evidence of need, access, or purchase frequency.
4. Assuming the startup can obtain a fixed market share simply because the market is large.
5. Mixing current market revenue with projected future revenue.
6. Double-counting overlapping segments.
7. Ignoring churn, discounts, implementation capacity, seasonality, and regulation.
8. Presenting false precision when assumptions are weak.

A rigorous estimate includes a base case, a conservative case, and an optimistic case. It should also disclose source date, geographic scope, customer definition, price basis, and assumptions.

8. Evaluating Market Attractiveness

Market attractiveness is multidimensional. A large market may be unattractive if access is expensive, regulation is restrictive, or incumbent suppliers are deeply embedded. A small market may be strategically valuable if it provides a credible path to learning and expansion.

Important criteria include:

• Problem intensity: Is the problem costly, urgent, frequent, or risky?
• Willingness to pay: Does value exceed price and switching cost?
• Growth: Is demand expanding, stable, or declining?
• Competitive structure: Are substitutes and incumbents strong?
• Accessibility: Can customers be reached through affordable channels?
• Unit economics: Can contribution margin support acquisition and service?
• Retention and expansion: Will customers renew, increase usage, or refer others?
• Strategic fit: Does the market match capabilities and purpose?
• Risk: Are there legal, technological, reputational, or dependency risks?

A useful distinction is between market attractiveness and startup fit. Attractiveness concerns the market; fit concerns the venture’s ability to win there. Both must be assessed.

9. Research, Validation, and Technology

Segmentation hypotheses should be tested through mixed methods. Qualitative interviews reveal language, context, and motivations. Quantitative surveys estimate prevalence, but responses about future behavior can be unreliable. Behavioral data, landing-page tests, pricing experiments, pilot contracts, and sales-conversion data provide stronger evidence of actual commitment.

AI tools can assist with transcript coding, cluster exploration, synthetic message testing, and personalized content drafts. However, AI-generated segments may reproduce sampling bias, infer sensitive characteristics, or create persuasive but unsupported narratives. Human review, data minimization, consent, explainability, and outcome monitoring are essential. AI should accelerate analysis, not replace customer contact or managerial judgment.

Privacy laws and sector regulations may restrict collection and use of personal data. Founders should establish a lawful purpose, collect only necessary information, protect data, document retention, and provide appropriate transparency. In B2B settings, organizational data can still contain personal information about employees and decision-makers.

10. Illustrative Case: FreshRoute Kitchen

Background. FreshRoute Kitchen is a fictional startup offering scheduled delivery of refrigerated, ready-to-cook meal kits in a large metropolitan area. Its founders initially described the market as “busy people who want healthier food.”

Problem. Advertising to the entire city produced clicks but few repeat purchases. Customer interviews revealed several different jobs: parents seeking predictable weekday dinners, fitness-oriented consumers tracking nutrition, and young professionals seeking convenience. These groups differed in order frequency, price tolerance, delivery windows, and retention.

Analysis. The team reviewed 40 interviews, 800 website sessions, and 120 trial orders. The strongest repeat behavior came from dual-income households with children who ordered three or more weekday dinners per week. Their job was not simply “eat healthy”; it was to reduce weekday planning and avoid last-minute takeaway meals. The group also had a clear trigger—returning to work or school after weekends and holidays—and could be reached through local parenting communities and workplace partnerships.

Approach. FreshRoute defined this group as its beachhead. Its persona was “the weekday coordinator,” a time-constrained household member responsible for meal planning. The team changed messaging from “chef-inspired meals” to “three predictable dinners without weekday planning.” It introduced family-sized bundles, fixed delivery windows, and a pause option for holidays. The ICP was a household within the delivery radius, ordering at least three meals weekly, with sufficient disposable income and a demonstrated preference for scheduled delivery.

The team estimated 50,000 households in the delivery area, of which 20% met its initial behavioral criteria. At an average annual contribution of $900 per active household, the behavioral SAM was:

50,000 × 0.20 × $900 = $9,000,000

Rather than claim a percentage of this market, the founders estimated first-year capacity at 1,000 households based on kitchen output and delivery routes. Their obtainable annualized revenue was therefore approximately:

1,000 × $900 = $900,000

Outcome. Repeat ordering improved after the offer was redesigned, although the company discovered that delivery density and food waste constrained profitability. It later expanded to adjacent households only after improving route economics.

Learning points. The case illustrates that a segment becomes strategically useful when it connects a specific job, trigger, channel, offer, and operating model. It also shows why market size must be constrained by operational capacity and why revenue opportunity is not equivalent to profit opportunity.

Challenges and Limitations

Segmentation can oversimplify people, especially when categories become treated as permanent identities. Markets are dynamic: a new regulation, platform, competitor, or cultural change can alter needs and channel access. Survey respondents may overstate interest, while early adopters may not represent the mainstream market. Market-size estimates can encourage premature confidence because large numbers appear authoritative.

Founders should therefore treat segmentation and sizing as living hypotheses. They should record assumptions, assign confidence levels, seek disconfirming evidence, and update models as actual acquisition, conversion, retention, and usage data become available. The most valuable output is not a perfectly precise number, but a defensible decision about whom to serve first and what must be learned next.

Chapter Summary

Effective startup marketing begins with a precise definition of the market and the customer problem. Segmentation divides a heterogeneous market using variables that predict meaningful differences in needs, behavior, economics, or access. B2C segmentation may use demographic, geographic, psychographic, behavioral, benefit, technological, and economic variables. B2B segmentation additionally considers firmographics, workflows, buying roles, technology, procurement, and account economics.

Jobs-to-be-done analysis focuses on the progress customers seek in a particular situation and prevents founders from defining competitors too narrowly. Personas translate research into representative user models, while ICPs identify attractive organizations, especially in B2B markets. Targeting evaluates segments against attractiveness and startup fit. A beachhead market is a focused initial market selected for urgency, reachability, willingness to pay, manageable service requirements, and expansion potential.

TAM, SAM, and SOM describe progressively narrower opportunity scopes. Top-down estimates apply transparent filters to external market data; bottom-up estimates build from customers, purchase frequency, and price. All estimates require consistent units, explicit assumptions, and careful distinction between annualized revenue and recognized revenue. Market attractiveness also depends on competition, accessibility, unit economics, retention, strategic fit, and risk. Because markets and evidence change, segmentation and market sizing should be continuously validated rather than treated as permanent facts.

Key Terms

Segmentation: Dividing a heterogeneous market into meaningful groups.

Targeting: Selecting market segments for focused strategic attention.

Positioning: Defining how an offering is understood relative to alternatives.

Persona: A research-based representation of a typical user or buyer.

Ideal customer profile: A description of the type of account most likely to create and receive value.

Jobs-to-be-done: A framework describing the progress a customer seeks in a situation.

Functional job: The practical task a customer wants to accomplish.

Switching force: A factor pushing a customer away from an existing solution or toward a new one.

Firmographic segmentation: Grouping organizations by attributes such as industry, size, or location.

Buying center: The people who participate in an organizational purchase.

Beachhead market: The initial focused market selected for repeatable traction.

Market attractiveness: The overall appeal of a market based on demand, economics, competition, access, and risk.

TAM: Total addressable market under a defined scope and period.

SAM: Serviceable available market that the offering can potentially serve.

SOM: Serviceable obtainable market that the startup can realistically capture.

Top-down sizing: Estimating opportunity by filtering an external market total.

Bottom-up sizing: Estimating opportunity from observable customer and transaction units.

ARPA: Average revenue per account over a specified period.

Unit economics: Revenue and cost relationships associated with a customer or transaction.

Market hypothesis: A testable assumption about customers, needs, behavior, or opportunity size.

Review Questions

Short Answer Questions

1. What is the difference between segmentation and targeting?
2. Why may behavioral variables be more useful than demographic variables?
3. Define a jobs-to-be-done statement and give its basic structure.
4. How does an ICP differ from a persona?
5. What is the relationship among TAM, SAM, and SOM?

Long Answer Questions

1. Explain the principal B2C and B2B segmentation variables and discuss when each is useful.
2. Describe a rigorous process for developing and validating a customer persona.
3. Explain the characteristics of an attractive beachhead market.
4. Compare top-down and bottom-up market-sizing methods, including their strengths and limitations.
5. Discuss the ethical, privacy, and analytical risks involved in data-driven segmentation.

Analytical / Application Questions

1. A startup sells compliance software to small manufacturers. Develop two possible segments, identify the relevant buyer roles, and recommend one beachhead market using explicit criteria.
2. Estimate the annual SAM for a service with 30,000 potential customers, where 35% are in the launch geography, 50% meet the technical requirements, and annual revenue per customer is $800. State your assumptions and interpret the result.
3. Create a JTBD statement, persona, and ICP for a venture of your choice. Explain how each would change the product, channel, or positioning decision.

Multiple Choice Questions with Answers

Question 1

Which statement best distinguishes an ideal customer profile from a persona?

A. An ICP describes an attractive account, while a persona represents a user or buying-role individual.

B. An ICP is used only in B2C markets, while a persona is used only in B2B markets.

C. An ICP describes a product feature, while a persona describes a price.

D. An ICP is an advertising slogan, while a persona is a market-size estimate.

Correct Answer: A. An ICP describes an attractive account, while a persona represents a user or buying-role individual.

Question 2

Which question most directly applies the jobs-to-be-done perspective?

A. What age group has the largest population?

B. What progress is the customer trying to make in a particular situation?

C. Which logo color does the customer prefer?

D. How many competitors use social media?

Correct Answer: B. What progress is the customer trying to make in a particular situation?

Question 3

A market has 10,000 potential accounts. Forty percent are in the launch geography, 50% meet technical requirements, and annual revenue per account is $1,000. What is the estimated SAM?

A. $500,000

B. $2,000,000

C. $2,500,000

D. $4,000,000

Correct Answer: A. $500,000

Question 4

Which characteristic is most important when selecting a beachhead market?

A. It is the largest imaginable population.

B. It has acute need and is reachable with the startup’s current capabilities.

C. It contains customers who never change suppliers.

D. It requires the startup to serve every customer simultaneously.

Correct Answer: B. It has acute need and is reachable with the startup’s current capabilities.

Question 5

Which is a B2B firmographic variable?

A. Household mood

B. Favorite leisure activity

C. Industry and employee count

D. Individual sleep pattern

Correct Answer: C. Industry and employee count

Question 6

If TAM is $20 million and SAM is $6 million, which SOM is logically consistent with the hierarchy?

A. $25 million

B. $20 million

C. $8 million

D. $2 million

Correct Answer: D. $2 million

Question 7

Why is bottom-up market sizing often useful for startups?

A. It eliminates all uncertainty.

B. It connects the estimate to observable customers, prices, and purchase behavior.

C. It always produces a larger estimate than top-down sizing.

D. It does not require assumptions.

Correct Answer: B. It connects the estimate to observable customers, prices, and purchase behavior.

Question 8

In a B2B purchase, the person who uses a software system daily is best described as the:

A. User

B. Economic decision-maker in every case

C. Market regulator

D. Competitor

Correct Answer: A. User

Question 9

Which practice improves the reliability of a persona?

A. Inventing demographic details to make it vivid

B. Basing it on repeated patterns from interviews and behavioral evidence

C. Creating one persona for everyone in the market

D. Changing it to match every individual customer

Correct Answer: B. Basing it on repeated patterns from interviews and behavioral evidence

Question 10

A startup estimates 240 acquired accounts at $1,200 annual revenue per account. What is the annualized revenue run rate from those accounts?

A. $28,800

B. $144,000

C. $288,000

D. $2,880,000

Correct Answer: C. $288,000