🎯 Market Segmentation

Definition, purpose, and economic rationale of segmentation, plus its criticisms
Segmentation strategy: undifferentiated vs. differentiated approaches
The S-T-P framework (Segmentation, Targeting, Positioning) and its three variations
Identifying the market: market potential and the Bass diffusion model
Bases for consumer markets: demographic, geographic, psychographic, behavioural, hybrid, generational, cultural, online
Selecting target markets and evaluating segment attractiveness
Marketing program and positioning, including perceptual mapping
Business markets (firmographics) and customer retention
Algorithms, statistical techniques, and data sources

📌 What Market Segmentation Is

Market segmentation (or customer segmentation) is the process of dividing a consumer or business market into meaningful sub-groups of current or potential customers, known as segments.

Objective: identify profitable and growing segments that a company can target with tailored marketing strategies.
Researchers examine shared needs, interests, lifestyles, or demographic profiles to find high-yield segments, which are prioritized as target markets.
It assumes different segments need different marketing programs: different offers, prices, promotions, distribution, or some combination.
It also builds profiles of key segments to understand their needs and purchase motivations, which support marketing strategy and planning.

Market type	Typical segmentation criteria
B2B	Company type, industry, geographic location
B2C	Demographic, behavioral, lifestyle, or socioeconomic criteria

Rationale

To achieve competitive advantage and superior performance, firms should:

Identify segments of industry demand
Target specific segments of demand
Develop specific "marketing mixes" for each targeted segment

From an economic perspective, segmentation assumes that heterogeneity in demand allows demand to be disaggregated into segments with distinct demand functions.

Criticisms

It does not, in fact, segment markets, and is no better than mass marketing at building brands.
In competitive markets, segments rarely differ much in how they use brands, and it fails to identify sufficiently narrow clusters.
Geographic/demographic segmentation is overly descriptive and lacks insight into motivations needed to drive communications strategy.
Market dynamics: segments are unstable over time; structural change causes segment creep and membership migration.
Segments are categories marketers create; consumers do not self-identify with them.

Despite its limitations, segmentation remains one of the enduring concepts in marketing. One American study suggested almost 60 percent of senior executives had used it in the past two years.

🧭 Market Segmentation Strategy

Approach	Description
Undifferentiated	The marketer ignores segmentation and develops a product meeting the needs of the largest number of buyers.
Differentiated	The firm targets one or more segments and develops separate offers for each.

Example: even once-commodity goods are now differentiated. Salt comes as cooking, table, sea, rock, kosher, mineral, herbal, iodized salt and substitutes. Sugar comes as cane, beet, raw, white refined, brown, caster, lumps, icing, syrup, invert sugar and substitutes. Invert sugar and syrups are marketed to food manufacturers (conserves, chocolate, baked goods); refined sugar is for the table, while caster and icing sugar suit home baking.

Factors affecting the strategy

Factor	Implication
Company resources	When resources are restricted, a concentrated strategy may be more effective.
Product variability	Uniform products (sugar, steel) suit undifferentiated marketing; differentiable products (cars) suit differentiated or concentrated approaches.
Product life cycle	A new product may launch with one version, then become more segmented; more competitors may force differentiation.
Market characteristics	If buyers have similar tastes or will not pay a premium for different quality, undifferentiated marketing is indicated.
Competitive activity	If competitors use differentiated or concentrated segmentation, undifferentiated marketing may prove fatal.

🔁 Segmentation, Targeting, Positioning (S-T-P)

Stage	What it comprises
Segmentation	Identifying the market to segment; identifying, selecting, and applying bases; developing profiles
Targeting	Evaluating each segment's attractiveness and selecting the segments to target
Positioning	Identifying optimal positions and developing the marketing program

A market segment is a portion of a larger market whose needs differ somewhat from the larger market. A firm focused solely on one segment's needs can meet them better than a firm trying to serve several segments.

Three variations in practice

Ad-hoc segmentation: firms independently run their own project, starting from a category of offerings and validating distinct consumption profiles. Profiles are often trade secrets.
Syndicated segmentation: firms buy commercially available frameworks from specialized firms that use data science to generate consumer profiles; clients consult them for a fee.
Feral segmentation: cultural intermediaries coin, circulate, and validate consumer categories that emerge unsolicited in popular culture, outside managerial control and without prescribed market research techniques.

📏 Identifying the Market to Be Segmented

The market for a product is its market potential or total addressable market (TAM). Estimating it is easy for existing products but very challenging for new ones with no historical data.

Basic approach: assess the broad population, estimate the percentage likely to use the product, then estimate revenue potential.
Historical analogy: e.g., an HDTV maker assumes adoption will resemble color TV adoption. Analogous rates provide benchmarks and can cross-validate other methods, though every market is unique.
Bass diffusion model: a more robust technique.

$N(t) - N(t-1) = \left[p + q\,\frac{N(t-1)}{m}\right]\left[m - N(t-1)\right]$

Symbol	Meaning
$N(t)$	Adopters in the current period $t$
$N(t-1)$	Adopters in the previous period
$p$	Coefficient of innovation
$q$	Coefficient of imitation (social contagion influence)
$m$	Estimated number of eventual adopters

The major challenge is estimating $p$ and $q$ . Values for more than 50 consumer and industrial categories are published; the average $p$ is 0.037 and $q$ is 0.327.

🧩 Bases for Segmenting Consumer Markets

Marketers seek internal homogeneity (similarity within segments) and external heterogeneity (differences between segments), and the base must be meaningful to the problem. Hair color suits a shampoo maker, not a seller of financial services.

A base works if it is identifiable, substantial, responsive, actionable, and stable:

Identifiability: managers can recognize distinct groups.
Substantiality: the segment is big enough (people or purchasing power) to be profitable.
Accessibility: marketers can reach it with promotion or distribution.
Responsiveness: members respond to offers targeted at them.
Actionable: the segment guides marketing decisions.

The most common bases are geographics, demographics, psychographics, and behavior. Marketers normally pick one, though bases can be combined with care (hybrid segmentation). Even dress size has been used successfully by some fashion houses.

👥 Demographic segmentation

Uses variables such as age, income, family size, and socio-economic status. It assumes similar demographic profiles mean similar purchasing patterns, motivations, interests, lifestyles, and brand preferences. Any census variable can be used.

Variable	Example descriptors
Age	Under 5, 5–8, 9–12, 13–17, 18–24, 25–29, 30–39, 40–49, 50–59, 60+
Occupation	Professional, self-employed, clerical, sales, trades, student, home duties, retired, etc.
Socio-economic	A–E or I–V (normally quintiles)
Family life-stage	Young single; young married no children; young family with children under 5; older married with children; older married with no children at home; older living alone
Education	Primary school up to postgraduate or higher
Home ownership	Renting, own with mortgage, owned outright
Others	Gender, marital status, family size, income, ethnicity, religion

Most use a combination of variables, requiring sophisticated analysis (cluster analysis, principal components analysis) and very large samples. Because collection is expensive, many firms buy data from commercial research firms.

Popular labels from the 1980s:

Label	Meaning
DINK	Double Income, No Kids; discretionary spending on luxury goods, entertainment, dining out
GLAM	Greying, Leisured and Moneyed; retired, asset rich; spend on recreation, travel, entertainment
GUPPY	Gay, Upwardly Mobile, Prosperous, Professional
MUPPY	Mid-aged, Upwardly Mobile, Prosperous, Professional
Preppy	(American) well-educated, well-off, upper-class young person from an expensive school
SITKOM	Single Income, Two Kids, Oppressive Mortgage; little discretionary income
Tween	About 9–12 years; between child and teenager
WASP	(American) White, Anglo-Saxon Protestant; high-status and influential
YUPPY	Young, Urban/Upwardly-mobile, Prosperous, Professional; career-minded, affluent, free spenders

🗺️ Geographic segmentation

Divides markets by geography, from continents to neighborhoods or postal codes. Variables: country, region, population density (CBD, urban, suburban, rural), city or town size, and climatic zone (Mediterranean, Temperate, Sub-Tropical, Tropical, Polar).

Geo-cluster (geodemographic) segmentation combines demographic and geographic data, classifying residential regions or postcodes by census and lifestyle characteristics.
It may be the first step in international marketing, deciding whether to adapt products to distinct geographic markets. Tourism boards often segment visitors by country of origin.
Used in direct marketing (personal selling, letter-box distribution, direct mail) and by governments and public bodies such as urban planning, health, police, telecommunications, and water boards.

🧠 Psychographic segmentation

Also called psychometric or lifestyle segmentation, it studies customers' activities, interests, and opinions (AIOs): how they spend leisure time and which influences they respond to.

Very widely used because it identifies tightly defined segments and clarifies motivations for product or brand choice.
Most studies are custom-designed for a product at a specific time, and commonly use quirky names for segments.

🛒 Behavioural segmentation

Groups consumers by observed behaviours. Many marketers consider behavioural variables superior to demographics and geographics.

Variable	Descriptors
Purchase/usage occasion	Regular, special, festive, gift-giving
Benefit sought	Economy, quality, service level, convenience, access
User status	First-time, regular, non-user
Usage rate	Light, moderate, heavy
Loyalty status	Loyal, switcher, non-loyal, lapsed
Buyer readiness	Unaware, aware, intention to buy
Attitude	Enthusiast, indifferent, hostile; price or quality conscious
Adopter status	Early adopter, late adopter, laggard

Descriptors are examples and are customized. In the health industry, planners segment by "health consciousness" (low, moderate, high), a behavioural segmentation using attitude.

Purchase/usage occasion: analyzes occasions of purchase or consumption using customer-level and occasion-level models. Unlike traditional models, it assigns more than one segment to each customer depending on current circumstances.
Benefit sought (needs-based segmentation): developed by Grey Advertising in the late 1960s. Products with different quality, performance, service, or features are pitched at each segment. Widely used for motor vehicles, fashion, furniture, consumer electronics, and holidays. Loker and Purdue found naturalists, pure excitement seekers, and escapists in pleasure travel.
Attitudinal segments: reveal the attitudes and beliefs driving decisions. The UK Department of Environment split the population into six segments:

Segment	Characteristics
Greens	Believe protecting the environment is critical; conserve whenever they can
Conscious with a conscience	Aspire to be green; concerned with wastage; unaware of broader issues such as climate change
Currently constrained	Aspire to be green but cannot afford organic products; pragmatic realists
Basic contributors	Skeptical about behaviour change; aspire to conform to social norms
Long-term resistance	Serious life priorities come first; low-impact behaviour for reasons other than conservation
Disinterested	See greenies as an eccentric minority; no interest in changing behaviour

🔀 Hybrid segmentation

Single-base segmentations serve only specific functions. A needs-driven one ("I want very quiet home appliances") guides product development but says little about brand positioning, database targeting, or messaging tonality. Hybrid segmentation combines two or more variable bases into one segmentation, driven by:

More powerful AI and machine learning to attribute segmentations to customer databases
Rapid growth in the breadth and depth of available data
Increasing prevalence of customer databases within companies

Example: TUI built a 2018 hybrid segmentation from geo-demographics, category attitudes, and holiday-related needs, credited with an incremental £50 million of UK revenue in just over two years before Covid-19 travel restrictions.

🧬 Other consumer segmentation

Generational segments

A generation is "a cohort of people born within a similar period (15 years at the upper end) who share a comparable age and life stage and who were shaped by a particular period."
It assumes values and attitudes shaped by key events translate into brand preferences. Dates are approximations: Australian Boomers 1945–1960; American and European Boomers 1946–1964, since population peaks differ by country.

Generation	Born
Builders	1920 to 1945
Baby boomers	about 1946–1964
Generation X	about 1965–1980
Generation Y (Millennials)	about 1981–1996
Generation Z (Zoomers)	1997–2012

Cultural segmentation

Classifies markets by cultural origin, enabling communications crafted for cultural communities and measurement of market penetration by product, brand, and channel.
Can combine with geographics to map segments by state, region, suburb, and neighborhood (choosing premises, territories, local marketing).
Census data is valuable but cannot meaningfully apply to individuals. Name analysis (onomastics) is the most reliable way to infer individuals' origin: 80 to 85% accuracy in Australia, with at least 99% of individuals coded.

Online customer segmentation

Segments should be identifiable, substantial, accessible, stable, differentiable, and actionable. Data comes from systems such as a CRM or DMP.
Forsyth et al. grouped active online consumers into six groups, differing in time online, pages and sites accessed, time per page, and kinds of sites visited:

Group	Characteristics
Simplifiers	Over 50% of online transactions; need one-click access and quick service; dislike pop-ups meant to encourage impulse buys (Amazon suits them)
Surfers	Spend a lot of time online; need variety and constant updates
Bargainers	Seek the best price
Connectors	Like to relate to others
Routiners	Want content
Sportsters	Like sport and entertainment sites

🎯 Selecting Target Markets

Marketers decide what criteria evaluate markets, how many markets to enter, and which segments are most valuable.

The primary target market is the main focus. The secondary target market is likely smaller but has growth potential, or is a small group of purchasers accounting for high sales volume through purchase value or frequency.
Three core considerations: segment size and growth; structural attractiveness; company objectives and resources.

There are no formulas for attractiveness, so judgment is needed. Guiding questions:

Dimension	Questions
Size and growth	How large is it? Substantial enough to be profitable (sales value or volume are superior measures)? Growing or contracting? Is growth sustainable? Stable over time?
Structural attractiveness	How many competitors target it? Do buyers have bargaining power? Are substitutes available? Can we differentiate? How responsive is it? Is it reachable?
Company objectives and resources	Aligned with our philosophy? Do we have the resources, prior experience, and know-how?

📣 Marketing Program and Positioning

The marketing program (marketing mix) must resonate with target markets and rests on deep knowledge of purchasing habits, preferred retail outlets, media habits, and price sensitivity.
Positioning is the final S-T-P step: how to present the offer so it resonates with the target market, informed by what motivates consumers.

David Ogilvy: "Positioning is the act of designing the company's offering and image to occupy a distinctive place in the minds of the target market. The goal is to locate the brand in the minds of consumers to maximize the potential benefit to the firm."

Perceptual mapping

Used to understand consumers' mental representations of brands in a category, traditionally with two variables (often price and quality).
Target-market respondents place brands on these variables; results are averaged and plotted, showing how the average person views each brand relative to others. Multi-dimensional maps are also used.

Approaches to positioning

Against a competitor
Within a category
According to product benefit
According to product attribute
For usage occasion
Along price lines (luxury or premium brand)
For a user
Cultural symbols (e.g., Australia's Easter Bilby as an alternative to the Easter Bunny)

🏢 Segmenting Business Markets

Easier than consumer markets. Bases include industry, size, location, turnover, number of employees, technology, and purchasing approach; the most widely used are geographics and firmographics.

Geographic: find promising regions using government business census products.
Firmographics (emporographics, feature-based segmentation) is business's answer to demographics; an estimated 81% of B2B marketers use it. Variables: standard industry classification (SIC), company size (revenue or employees), industry sector or location, usage rate, purchase frequency, years in business, ownership factors, buying situation.

🔒 Use in Customer Retention

A company tags each active customer on axes:

Risk of cancellation: a drop in usage is a common indicator (e.g., declining card spending in credit cards).
Risk of switching to a competitor: data analysis of why preferences changed supports win-back strategies.
Retention worthiness: is predicted post-retention profit greater than the cost of retaining the customer, including customer lifecycles? Tactics range from discounts to messages reinforcing the value proposition.

📊 Algorithms and Approaches

Method choice depends on the broad approach (a-priori or post-hoc), data availability, time, skill, and resources.

Aspect	A-priori	Post-hoc
Starting point	Theoretical framework developed before the research	No assumed framework; empirical data drives selection
Typical analysis	Cross-tabulations, frequency distributions, occasionally logistic regression	Clustering analysis or structural equation modeling
Data needs	Attitudinal data on a pre-chosen group	Rich datasets with very many cases
Drawback	Does not explore possibly more meaningful segments	Needs sophisticated algorithms

A-priori example: studying light and moderate users (a behavioural variable, user status) to increase usage rates.

Statistical techniques

Clustering algorithms (overlapping, non-overlapping, fuzzy), e.g., K-means
Conjoint analysis
Ensemble approaches such as random forests
Chi-square automatic interaction detection (a decision tree)
Factor analysis or principal components analysis
Latent class analysis
Logistic regression
Multidimensional scaling and canonical analysis
Mixture models, e.g., EM estimation, finite-mixture models
Model-based segmentation with structural equation modeling, e.g., LISREL
Artificial neural networks

Firms often hire commercial research firms or consultancies, especially for post-hoc analysis.

Data sources

Internal	External
Customer transaction records	Commissioned research (exclusive rights, typically most expensive)
Patron membership records	Data-mining techniques
CRM databases	Census data
In-house surveys	Observed purchase behaviours
Self-completed questionnaires or feedback forms	Government agencies, statistics, and surveys
	Omnibus surveys (regular standard survey where firms add questions; lower cost)
	Professional, industry, and employer associations
	Proprietary or syndicated surveys (buy access to part of a data set)
	Proprietary databases and software

Proprietary segmentation databases: Acorn and Claritas Prizm, both geo-demographic.

Market Segmentation: Marketing Study Notes