An average can describe a market without describing anyone’s actual requirements particularly well. Some buyers want simplicity, others want extensive support, and others need a product for an unusual application. Combining those preferences into one specification may produce an offer that fits none of them closely. The history of market segmentation concerns the strategic importance of recognizing such differences.
In July 1956, Wendell R. Smith published “Product Differentiation and Market Segmentation as Alternative Marketing Strategies” in the Journal of Marketing. The six-page article appeared in volume 21, issue 1, on pages 3–8. Its recorded affiliation was Alderson & Sessions, placing the contribution in a setting connected with applied marketing work (Smith, 1956).
The importance of the article lies in conceptual formulation. Smith did not invent the observation that people buy different things. He helped establish differences in demand as an explicit problem of marketing strategy. His contribution became a reference point for later research, while the practices it addressed had a much longer history.
That distinction makes 1956 a meaningful milestone without turning it into a universal beginning. Earlier businesses had developed offers for different buyers. Later researchers would investigate benefits, statistical grouping and implementation in much greater detail. Smith belongs between those histories: a significant interpreter of a problem, rather than the inventor of every subsequent solution.
A Contribution Between Economics and Management
The article appeared as marketing was increasingly articulated as a management responsibility. John A. Quelch and Katherine E. Jocz place Smith among influential contributors to that development. Their historical account describes the growing attention given to customer problems and differing benefits in the decades after the Second World War (Quelch and Jocz, 2008).
This does not mean that earlier firms operated without understanding buyers. It concerns a change in systematic explanation and teaching. A familiar business observation can acquire a different role when it is expressed as a concept that managers can discuss across products and industries.
The underlying decision is practical. Should a firm attempt to bring many buyers towards a limited offer, or recognize different requirements and cater more precisely for selected groups? The answer can affect production, distribution, costs and the allocation of resources, as well as the wording of advertising.
The historical contribution is therefore more substantial than a new label for a mailing list. It connects the structure of demand with the design of an offer. The relationship has to be reconsidered whenever a firm decides which differences it can serve and which it will leave unresolved.
A concept of this kind can be influential because it organizes existing experience. Managers may already have encountered contrasting requirements but treated each case separately. A shared framework makes it possible to compare those cases, state assumptions and ask whether an established way of doing business remains appropriate.
What Market Segmentation Means
In contemporary terms, market segmentation groups buyers according to selected characteristics relevant to an offer. Members of a group should be similar on the criteria being used, while differences between groups should matter for the decision. Sara Dolnicar, Bettina Grün and Friedrich Leisch emphasize that the criteria define the relevant similarity; the concept does not imply that people are alike in every respect (Dolnicar, Grün and Leisch, 2018, ch. 1).
This qualification prevents a common misunderstanding. Two customers may both require reliable technical support while differing greatly in age, income or interests. Conversely, two people of the same age may seek entirely different benefits from the same product category.
The task is to develop a useful representation of a particular market problem. A hotel operator, an industrial supplier and a publisher may need different distinctions. Segmentation takes its meaning from the decision it is intended to support, rather than from a universal list of categories.
The unit of analysis also needs attention. A study might concern people, organizations, purchases or situations of use. Someone buying an everyday item may apply different criteria when choosing a gift. An account of a segment should make clear whether it describes a relatively enduring preference or behaviour under specific circumstances.
This is one reason segments should not be treated as complete portraits of human beings. They are selective models. A good model makes the relevant differences easier to understand while remaining explicit about what it leaves out.
Smith’s Contrast with Product Differentiation
Smith contrasted bringing demand towards an offer with adjusting more precisely to divergent requirements. His contemporary illustration included a refrigerator without a frozen-food compartment for households already equipped with a separate freezer. The relevant distinction concerned an existing provision that changed the usefulness of a product feature (Smith, 1956).
The example is revealing because it does not begin with an arbitrary demographic label. A feature useful to one household may be less valuable to another because of the surrounding situation. The product is being assessed in relation to what its user already has and needs.
The contrast should not become an inflexible instruction. A supplier can serve a particular group and also distinguish itself from rivals within that group. It can adapt several offerings without giving each one a wholly unrelated identity. Segmentation and differentiation can operate together in practical strategy.
Peter R. Dickson and James L. Ginter revisited the conceptual confusion in a 1987 Journal of Marketing article. Their work shows that widespread use of both expressions had not produced a uniformly clear understanding of their meanings and relationships (Dickson and Ginter, 1987).
That later discussion is part of the significance of Smith’s article. An influential formulation can organize a field while still leaving questions for subsequent research. Historical recognition does not require treating every original expression as the final version of a concept.
Why the Average Offer Can Miss the Market
Consider a hypothetical market for business software. Some organizations want an immediately usable standard service. Others require extensive adaptation and specialist guidance. A compromise may be too complicated for the first group and insufficiently capable for the second.
Segmentation initially makes those differences visible. Whether the firm should respond with two products, a basic product plus services, or concentration on one group is a further decision. The description does not determine the best commercial response by itself.
This separates strategy from simple list management. Giving two names to otherwise identical offers may accomplish little. A consequential segmentation asks whether the differences warrant changes that buyers will find meaningful and that the firm can actually deliver.
Averages remain useful for summarizing scale and change. The problem arises when a summary is turned into an assumption of uniform requirements. Market segmentation addresses that gap between a convenient aggregate and the variety relevant to an offer.
It can also reveal a poor compromise. A firm may have accumulated features to satisfy separate requests without deciding whom the whole product is meant to serve. Understanding demand differences offers a way to reconsider the combination, rather than continuing to add complexity indefinitely.
Segmentation, Targeting and Positioning
The later STP framework distinguishes grouping the market, selecting target segments and establishing an intended position. Dolnicar, Grün and Leisch treat these as connected strategic decisions. They also caution against interpreting the sequence too rigidly: evaluating possible targets may require revisiting the initial segmentation (Dolnicar, Grün and Leisch, 2018, ch. 11).
For historical purposes, that later framework should not be read in its entirety into a title published in 1956. Smith became an important reference point. It does not follow that he invented every subsequent textbook stage in its present form.
The distinctions are straightforward. Segmentation develops a meaningful grouping. Targeting determines which groups the organization will prioritize. Positioning concerns the way its offer should be understood relative to alternatives. A group can be clearly described without being an appropriate target for a particular supplier.
An attractive segment might require capabilities that the firm lacks. Another supplier may possess exactly those capabilities. Target selection therefore connects an understanding of buyers with an understanding of the organization. Statistical analysis can inform that judgment but does not make it automatically.
Positioning adds a further question. A firm needs an intelligible reason for its chosen audience to consider the offer. Merely identifying the group does not establish that reason, particularly where competitors are trying to satisfy similar requirements.
The Move Towards Benefit Segmentation
Russell I. Haley’s “Benefit Segmentation: A Decision-oriented Research Tool” appeared in 1968. It questioned reliance on variables that merely describe buyers and proposed grouping them around the benefits they sought. The approach directed attention towards reasons for product use rather than convenient demographic descriptions alone (Haley, 1968).
This was a consequential extension of the discussion. Age or location may provide clues without explaining the expected benefit. People from different backgrounds may share a problem they want a product to solve, while apparently similar people may attach importance to different outcomes.
A hypothetical training service illustrates the distinction. Some participants seek a recognized certificate, others want help with an immediate workplace task, and others need preparation for changing occupations. Those objectives may suggest different formats, support and messages, even though all the customers appear within one general course category.
The historical contribution should not be inflated into proof that a stated benefit explains every purchase. Description, explanation and prediction remain different ambitions. A benefit-based approach gives research a more decision-related direction; it still needs suitable evidence.
It also changes the role of a segment label. Calling a group “practical problem solvers” is not sufficient. Researchers need to establish what problems matter, how the proposed offer addresses them and whether the grouping helps the organization choose an effective response.
Beyond Age, Income and Location
Daniel Yankelovich’s “New Criteria for Market Segmentation” was published in 1964. The publisher’s description identifies attitudes, motives, values, usage patterns and aesthetic preferences as perspectives beyond conventional demographics, applying the discussion to consumer and industrial markets (Yankelovich, 1964).
The implication is that segmentation should not be reduced to a fixed catalogue of variables. Relevant information depends on the product and decision. A delivery service may need to understand time sensitivity and reliability requirements more than the private lifestyles of purchasing staff.
Easily measured characteristics can still be useful. They may help locate or communicate with an already understood group. But the feature that distinguishes a need and the feature that makes the group reachable are not necessarily the same.
For example, research may identify users requiring unusually intensive advice. Another observation may help select contact channels through which those users can be reached. The communication plan uses a characteristic that does not necessarily explain the original requirement.
Distinguishing those roles matters increasingly as information becomes plentiful. Availability can make a variable tempting, but it does not establish relevance. A useful analysis explains why the information belongs in the model rather than relying on the ease with which it can be collected.
Earlier Practice in Britain and the German Book Trade
Ronald A. Fullerton’s 2015 CHARM abstract documents earlier segmentation practices in Britain, Germany and the United States. His examples include publishers, fashion and ceramic producers, and nineteenth-century manufacturers of bicycles and fountain pens. The central historical point is that businesses acted on such distinctions long before marketing theory formally discussed them (Fullerton, 2015).
The evidence requires interpretation. Multiple products do not by themselves establish segmentation. Variety can arise from technical limitations, craft practices or separate inventions. A stronger claim needs evidence that differences in buyers or requirements were recognized and addressed deliberately.
Wedgwood offers an informative British setting. The Victoria and Albert Museum describes relatively affordable creamware, a royal order in 1765 and a wide range of fashionable designs. The business operated across differing possibilities of expenditure and display (Victoria and Albert Museum, 2024). Alongside Fullerton’s research, this supports discussion of the older relationship between offers and contrasting consumption expectations.
A royal commission alone remains evidence of a particular customer and order. Further records are needed to explain how prestige and broader demand interacted. Separating an object, an individual transaction and a sustained strategy prevents every historical luxury product from being labelled a complete modern marketing model.
Fullerton’s study of German book marketing covers 1815–1890. The published introduction describes publishers’ active development of markets and their growing understanding of readers, using extensive contemporary trade material. It also identifies international interaction within the book business (Fullerton, 2016).
Books can differ by subject, price, reading requirements and intended use. The historically significant question is how that variety connected with ideas about particular readers. The book trade consequently illustrates why the history of a practice can begin well before the history of its established scientific terminology.
Fullerton’s historical overview also became a chapter in The Routledge Companion to Marketing History. The editors’ published introduction situates earlier practices across several industries within a wider history and challenges an exclusively American post-war account of marketing’s origins (Tadajewski and Jones, 2016).
These comparisons qualify Smith’s priority without diminishing his contribution. He supplied an influential formulation of a strategic issue. Earlier practitioners supplied evidence that the underlying commercial problem was neither new nor confined to one country.
General Motors Before the 1956 Formulation
The automotive industry provides another documented comparison. The Henry Ford museum holds a General Motors advertisement dated December 1925 under the slogan “A Car for Every Purse and Purpose.” Its collection description explains the range of brands and price levels associated with Alfred P. Sloan, including the possibility of moving from Chevrolet towards a more expensive marque (The Henry Ford, n.d.).
The advertisement demonstrates a public appeal to different purchasing possibilities decades before Smith’s article. It does not establish which research methods managers used or whether their reasoning matched his later terminology. Advertising documents an outward promise; reconstructing internal decision processes requires further evidence. This distinction matters whenever a historical example is recruited as an origin story.
As an analytical comparison, the case also separates purchasing power from desired benefits. Buyers with similar budgets can want different things, while a shared practical requirement can be served at different price levels. A portfolio of brands therefore raises questions about segmentation without, by its mere existence, answering all of them.
Segments Also Have Social Histories
A segment is not an unchanging natural fact. Blaine J. Branchik’s Journal of Historical Research in Marketing study examines the American elderly market, connecting its development with urbanization, retirement provision, political organization and changing marketing practices (Branchik, 2010).
The example reveals another layer of explanation. A group may be described by a seemingly simple age criterion while its economic significance changes through institutions and living conditions. Income sources, household arrangements and expectations of later life affect which offers become relevant.
The research concerns the United States. Its historical phases should not be transferred mechanically to other countries. Comparable ages do not guarantee comparable purchasing power, daily experience or access to services. International segmentation needs investigation of those conditions rather than identical headings in a spreadsheet.
Differences also remain within an age group. Health, household circumstances and interests can vary substantially. The existence of a senior market does not imply that older people form one useful segment for every product.
The historical lesson is that market categories are shaped over time. Demographic description can help identify a population, while the conditions that make it a meaningful commercial audience require a fuller account. Segmentation becomes more informative when those two questions remain connected.
From Market Knowledge to Statistical Techniques
Later research developed increasingly sophisticated methods for forming and evaluating groups. Quelch and Jocz’s retrospective points to the expanding methodological field and the 1978 discussion in the Journal of Marketing Research. It also distinguishes technical progress from the visibility of documented managerial application (Quelch and Jocz, 2008).
A mathematically defined solution is not yet an implemented strategy. An organization needs to understand how it was produced, what the groups represent and which choices it can support. Otherwise, an elaborate analytical structure may remain unused.
Dolnicar, Grün and Leisch distinguish approaches that define groups in advance from approaches that investigate patterns in data. A predetermined criterion might establish the initial division. Data-driven analysis can examine combinations of variables whose resulting profiles are not known beforehand. Mixed approaches are also possible (Dolnicar, Grün and Leisch, 2018, ch. 2).
This is not a simple replacement of experience by statistics. Experience can guide the choice of questions and help interpret findings. Data can test assumptions and reveal unexpected relationships. Either becomes problematic when treated as a complete substitute for the other.
The historical development therefore involves changes in the resources available for understanding markets. What a proprietor might learn through direct interaction can become harder to observe in a large organization. Research methods offer ways to investigate that distance, although they also introduce choices that must be understood.
A Cluster Is Not Automatically a Natural Market Segment
A clustering method groups observations by selected similarities. Its result does not prove that it has uncovered the one true structure of the market. Data selection, preparation, analytical method and the chosen number of groups can all influence the outcome.
Dolnicar, Grün and Leisch distinguish natural, reproducible and constructive approaches. They explain that real consumer data often do not contain clearly separated natural clusters. A useful segmentation can consequently be a reasoned construction, provided its limits are understood and the resulting model is evaluated appropriately (Dolnicar, Grün and Leisch, 2018, ch. 2).
Imagine willingness to pay and need for advice varying gradually across customers. Dividing that variation into three groups may help planning without demonstrating a sharp boundary at which people become fundamentally different types of buyer. The groups are then working models of a continuous pattern.
That does not make them worthless. A map also simplifies a landscape to support navigation. The relevant question is whether the simplification serves its intended purpose. For segmentation, that includes whether a firm can recognize a group, respond appropriately and observe the consequences.
This perspective also encourages modesty in naming. A vivid label may help internal discussion but should not conceal uncertainty or replace evidence. The profile should remain traceable to the information and assumptions from which it was developed.
The Economic Limits of Variety
More segments do not automatically produce better marketing. Additional variants, packaging, channels and service promises require resources. Research, data maintenance and internal coordination add costs as well. A finer division must produce enough value to justify those consequences.
Dolnicar, Grün and Leisch therefore begin their practical process by asking whether an organization should segment at all. They treat the decision as a longer-term commitment that can require changes in products, pricing, distribution and organizational arrangements (Dolnicar, Grün and Leisch, 2018, ch. 3).
That is a useful corrective to the idea that segmentation ends with a presentation. Naming three groups accomplishes little if the firm cannot or will not respond to their requirements. Analytical investment and the ability to act need to fit together.
The underlying management problem remains recognizable across periods. The question is not how many groups a method can generate. It is which differences a business can take into account meaningfully and economically. A limited number of well-understood segments may be more useful than an elaborate classification without consequences.
Concentration also has trade-offs. A specialized offer can fit a chosen need closely while increasing dependence on that part of the market. Broader coverage can spread exposure but demand more coordination. The concept does not eliminate those choices; it makes their basis more explicit.
What Has to Change Inside the Firm
A segment decision can involve several functions. Sales staff need an intelligible account of relevant requirements. Product teams need to know which features matter. Service and distribution have to support the promised experience. Communication can express that coordination but cannot replace it.
Consider a hypothetical maintenance provider separating routine planned visits from highly time-sensitive breakdowns. Different messages would have limited value if both groups received the same response capacity. The proposed adaptation would need operational support.
The reverse problem is possible too. A valuable specialist service can remain unnoticed if buyers cannot understand its relevance. A segment-based offer needs both a genuine capability and a recognizable meaning for the intended user.
Dolnicar, Grün and Leisch explicitly connect target selection with modification of the marketing mix. Product, price, channels and communication need to be considered together, alongside positioning and competition (Dolnicar, Grün and Leisch, 2018, ch. 11).
This explains why segmentation is more than a research technique and more than copywriting. It sits at a junction between observation and action. Its usefulness depends on whether knowledge about a group is translated into a coherent offer rather than remaining an isolated departmental exercise.
Digital Personalization Is a Different Level of Decision
Digital systems can adapt messages or offers to individual contacts. Segmentation may inform that process, but it is not identical to every personalized interaction. Serving a group strategically and displaying a particular message to one person concern different levels of decision.
An organization can choose a long-term audience around shared requirements and then vary communication by situation. Equally, it can generate many individualized messages without a clear understanding of which requirements it is especially well equipped to satisfy.
A technically selectable advertising audience is not automatically a strategically meaningful market segment. A system’s filter establishes that a selection is possible. It does not establish that the selected characteristic captures a relevant need or predicts an appropriate response.
Smith’s continuing relevance should therefore not be expressed by crediting him with modern software functions. It lies in the need to connect observed differences with sensible choices about an offer. More information expands the possible analysis without removing the responsibility for that judgment.
The distinction also prevents present-day practice from becoming the measure of all earlier work. A merchant could recognize meaningful differences without a digital database. A contemporary firm can possess a large database and still misunderstand those differences. The technology available and the quality of strategic interpretation are related but separate matters.
Reviewing Segments Instead of Treating Them as Permanent
A useful classification can lose its relevance. Needs, competitors, prices and situations change. Even where a broad segment remains meaningful, its composition or expectations may shift. Segmentation is therefore not a timeless statement about a market.
Dolnicar, Grün and Leisch connect implementation with continuing evaluation and monitoring. The organization should investigate whether the strategy contributes to its objectives and whether changes require a different response. Analysis does not end when the target has been chosen (Dolnicar, Grün and Leisch, 2018, ch. 12).
Differences in a table are insufficient evidence of practical value. A statistically distinct group may be difficult to reach. A readily reachable group may require services the supplier cannot provide economically. Evaluation needs to consider the actual decisions enabled by the model.
These limits are part of the concept’s usefulness. Segmentation can make assumptions explicit enough to test and revise. It offers a basis for learning, rather than a guarantee against mistakes. A category that is never reconsidered can become a convenient label detached from the market it was intended to explain.
What the 1956 Contribution Means Today
Smith’s article deserves its position as a major contribution to the language of marketing strategy. It helped make differences in demand an explicit starting point for decisions about an offer. Later discussion and research could return to that shared question even as methods and terminology changed.
The earlier examples prevent an exaggerated origin story. Publishers, ceramic producers and other businesses responded to contrasting buyers long before 1956. Later work prevents the opposite error of treating six pages as a complete anticipation of benefit segmentation, statistical analysis and organizational implementation.
Between those qualifications, the achievement remains substantial. Smith stood neither outside all earlier experience nor at the end of conceptual development. He formulated a recurring problem clearly enough to become a durable reference point: when should a firm bring different requirements towards a common offer, and when should good marketing provide deliberately different answers?
References
Branchik, B. J. (2010) ‘Silver dollars: the development of the US elderly market segment’, Journal of Historical Research in Marketing, 2(2), pp. 174-197. https://doi.org/10.1108/17557501011042542. Publisher’s abstract consulted.
Dickson, P. R. and Ginter, J. L. (1987) ‘Market Segmentation, Product Differentiation, and Marketing Strategy’, Journal of Marketing, 51(2), pp. 1-10. https://doi.org/10.1177/002224298705100201. Publisher’s abstract and bibliographic information consulted.
Dolnicar, S., Grün, B. and Leisch, F. (2018) Market Segmentation Analysis: Understanding It, Doing It, and Making It Useful. Singapore: Springer. Chapters consulted: 1-3 and 11-12. https://doi.org/10.1007/978-981-10-8818-6. Open access under CC BY 4.0: https://creativecommons.org/licenses/by/4.0/. This article provides original summaries and interpretations rather than unchanged reproductions.
Fullerton, R. A. (2015) ‘Segmentation in Practice: an Historical Overview of the 18th and 19th Centuries’, Proceedings of the Conference on Historical Analysis and Research in Marketing, p. 190, abstract. https://ojs.library.carleton.ca/index.php/pcharm/article/view/1481/1336 (Accessed: 22 October 2026).
Fullerton, R. A. (2016) The Foundations of Marketing Practice: A History of Book Marketing in Germany. Abingdon: Routledge. Published preview and introduction consulted. https://api.pageplace.de/preview/DT0400.9781317536130_A26310091/preview-9781317536130_A26310091.pdf (Accessed: 22 October 2026).
Haley, R. I. (1968) ‘Benefit Segmentation: A Decision-oriented Research Tool’, Journal of Marketing, 32(3), pp. 30-35. https://doi.org/10.1177/002224296803200306. Publisher’s abstract consulted.
Quelch, J. A. and Jocz, K. E. (2008) ‘Milestones in Marketing’, Business History Review, 82(4), pp. 827-838. https://www.hbs.edu/research/RePEc/hbs/journl/2008Q4quelch.pdf (Accessed: 22 October 2026).
Smith, W. R. (1956) ‘Product Differentiation and Market Segmentation as Alternative Marketing Strategies’, Journal of Marketing, 21(1), pp. 3-8. https://doi.org/10.1177/002224295602100102. Also consulted: a later Marketing Management reprint of the original article, accessible at https://www.academia.edu/36725071/Product_Differentiation_and_Market_Segmentation_As_Alternative_Marketing_Strategies (Accessed: 22 October 2026).
Tadajewski, M. and Jones, D. G. B. (2016) ‘The history of marketing practice’, in Jones, D. G. B. and Tadajewski, M. (eds) The Routledge Companion to Marketing History. Abingdon: Routledge, pp. 1-22. Published preview consulted. https://api.pageplace.de/preview/DT0400.9781134688685_A26553252/preview-9781134688685_A26553252.pdf (Accessed: 22 October 2026).
The Henry Ford (n.d.) ‘Advertisement for General Motors, “A Car for Every Purse and Purpose,” 1925’. Digital collection entry, object 64.167.657.405. https://www.thehenryford.org/collections/explore/artifact/192114 (Accessed: 22 October 2026).
Victoria and Albert Museum (2024) ‘Wedgwood: An introduction’, updated 17 April. https://www.vam.ac.uk/articles/wedgwood-an-introduction (Accessed: 22 October 2026).
Yankelovich, D. (1964) ‘New Criteria for Market Segmentation’, Harvard Business Review, March-April. Publisher’s description of the original article consulted. https://store.hbr.org/product/new-criteria-for-market-segmentation/64213 (Accessed: 22 October 2026).
