Before you start
You will be able to
- Describe psychographic bases (lifestyle, values, personality) and the VALS framework in outline, and say how they are measured.
- Distinguish the behavioural bases: benefits sought, usage rate, occasions, user and loyalty status, readiness stage and attitude.
- Calculate the share of volume or spending that comes from the heaviest buyers, and test the 80/20 claim against data rather than assuming it.
- Explain why benefit and needs segmentation explain choices better than demographics, and why such segments are harder to reach.
- Use a multi-item attitude scale as a segmentation base, and judge which behaviours it does and doesn’t predict.
- Identify the bases used to segment business markets beyond firmographics.
Review first
- Demographic and geographic segmentation — Demographic and geographic bases, cross-tabs with chi-square and Cramér’s V, and the limits of “who” variables.
- What influences buying — Lifestyle, attitudes, motives and the other psychological influences on buying were defined there.
- Questionnaire design — Psychographic and attitude bases are measured with Likert items and scales, checked with Cronbach’s alpha.
- How consumers make buying decisions — Involvement and the buying decision process, which readiness stages describe.
Fictional case A fictional coffee business founded by Ana Lim. It starts as a catering and barista-training service, opens its own cafés and an online shop for beans, and plans a canned cold brew.
The idea
Fictional case
Ana’s head barista has a theory: “Our customers aren’t an age group. They’re coffee people. They drink a lot of it, they care where it comes from, and they want to support places like ours.” The theory makes three claims about kinds of people, how much they drink, what they value and how loyal they feel, that the needs study can test. It recorded cups of coffee a week, monthly spend and a three-item scale of loyalty to local, independent businesses for each of the 640 respondents.
The previous lesson showed that demographics describe who buys far better than why. This lesson turns to bases closer to the why. Psychographic segmentation divides a market by how people live and what they are like: their lifestyle (activities, interests and opinions, from M01), their values and their personality. Behavioural segmentation divides it by what people do and seek in the category: the benefits they want, how much they buy, when, how loyally, and how ready they are to buy.
Psychographics: how people live and what they value
Two people with the same age and income can lead very different lives: one cycles to work, cooks from scratch and reads labels; the other works shifts and grabs whatever is quickest. Psychographics captures that difference. It is measured with statements people rate, such as “I try new foods whenever I can” (an interest) or “Protecting the environment matters more to me than convenience” (a value), combined into scales as M04’s Questionnaire design taught and checked for reliability with Cronbach’s alpha.
A long-established commercial psychographic system is VALS, launched by SRI International in 1978, relaunched in a new version in 1989 and offered today by Strategic Business Insights, an SRI spin-off. In SRI’s words, VALS “uses psychology to describe the dynamics underlying consumer preferences and choices”. OpenStax lists its eight US types: Innovators, Thinkers, Achievers, Experiencers, Believers, Strivers, Makers and Survivors. OpenStax’s descriptions place the types by the resources they have, from Innovators with the most to Survivors with the fewest, and by what drives them: knowledge for Thinkers, excelling at work and in the family for Achievers, being different for Experiencers.
Behaviour: what people seek and do
Behavioural bases come closest to the purchase itself. The most influential is benefit segmentation, named by Russell Haley in 1968: divide buyers by the benefits they seek from the product, because those benefits are the reasons a segment exists. Haley’s illustration was toothpaste, where he described a sensory segment (flavour and appearance), sociables (bright, white teeth), worriers (preventing decay) and an independent, price-led segment. Measuring how important a list of needs is to each buyer and clustering them, as Ana’s study does with eight coffee needs, is the modern form: needs-based segmentation.
Other behavioural bases describe the buying itself. Usage rate separates non-users from light, medium and heavy users. Occasions divide buying by when and why it happens: the commuter’s morning cup, the weekend treat, the office meeting.
User status separates non-users, former users, first-time users and regular users; loyalty status how firmly customers stick to one brand. The buyer-readiness stage places people from unaware, through aware and interested, to intending to buy. And attitude, how people feel about the product or the brand, can be a basis too.
Heavy and light coffee drinkers: does 80/20 hold?
A rule of thumb says that 80% of a business’s revenue comes from the top 20% of its customers; OpenStax states it as the Pareto principle. In Ana’s study, the 133 respondents who drink 16 or more cups a week, 20.8% of them, account for 45.7% of all the cups the 640 drink. Ranked by monthly spend instead, the top fifth (the 128 who spend most, each at least $70 a month) account for 38.6% of spending. Concentrated, yes; 80/20, no. Treat the rule as a pattern to check against data, not as a law.
| Usage band (cups a week) | Respondents | Share of respondents | Share of cups | Share of spend | Spend per head (month) |
|---|---|---|---|---|---|
| Light (1–4 cups) | 151 | 23.6% | 6.2% | 23.1% | $49.25 |
| Medium (5–15 cups) | 356 | 55.6% | 48.0% | 56.1% | $50.72 |
| Heavy (16 or more) | 133 | 20.8% | 45.7% | 20.8% | $50.34 |
| All | 640 | 100% | 100% | 100% | $50.29 |
Percentages are rounded, so they may not add to exactly 100.
The table holds a surprise. Heavy drinkers account for 45.7% of the cups but only 20.8% of the spending, the same as their share of respondents: they spend about the same per head as light drinkers ($50.34 against $49.25 a month; a two-sample t test gives t = 0.31, p = 0.76). Across all respondents, cups and spend are uncorrelated (r = −0.01). Heavy drinkers get many cheap cups at home, from supermarkets and from office machines; light drinkers buy fewer, dearer cups, 39.1% of them mainly at cafés against 20.3% of heavy drinkers. Whether “heavy” means volume or value changes who the best customers are.
In depth
Measuring psychographics
Lifestyle is commonly measured with “AIO” statements about activities, interests and opinions, as OpenStax describes. For segmentation the items work best when they are few and targeted at the category: for coffee, attitudes to independent shops, interest in new flavours, concern about sugar. Each scale should hold several items that measure one idea, reach an acceptable Cronbach’s alpha and relate to the behaviour it is meant to explain. Segments are then formed either a priori, by cutting a scale (high, middle and low), or post hoc, by clustering several scales together.
Values are not the same as the opinions in an AIO battery. An opinion is a view on a particular topic, such as “organic coffee is overpriced”; a value is a lasting belief about what matters in life, such as “protecting the environment comes first”, which shapes many opinions and choices at once. Questionnaires usually keep the two in separate scales.
Psychographics has two known weaknesses. Labels such as “free spirits” are vivid but can promise more than the data show, a risk module 2 returns to when naming segments. And broad personality traits are weak predictors of what people buy: Lastovicka and Joachimsthaler (1988) note that reviews have generally found weak evidence for the belief that consumers choose products to fit their personalities, and argue that part of the reason is unreliable measures of behaviour. Category-specific attitudes and values sit closer to the purchase than general traits, which makes them a more promising place to start.
VALS illustrates both the appeal and the limits. Its types are memorable and come with descriptions of how each group shops and what it reads, which helps creative teams. But a VALS type is a classification of a person’s general motivations, not of their needs in a particular category: two Experiencers may want different coffee. Like any psychographic system, it is most useful when the types are shown, with data, to differ in the category being marketed.
The behavioural bases in more detail
Kotler’s Principles of Marketing (3rd ed., 1986) listed the behavioural variables with their usual breakdowns, and the table follows it, with a coffee example for each.
| Variable | Typical breakdown | Coffee example |
|---|---|---|
| Occasion | Regular occasion, special occasion | The morning commute; a weekend treat; meetings at work |
| Benefits sought | The benefit that matters most, such as quality, service or economy | Taste, a low price, a caffeine kick, low sugar |
| User status | Non-user, ex-user, potential user, first-time user, regular user | Has never bought Lantern; bought once; buys weekly |
| Usage rate | Light, medium, heavy user | Cups a week, or dollars a month |
| Loyalty status | None, medium, strong, absolute | Always buys at the same café; switches freely |
| Readiness stage | Unaware, aware, informed, interested, desirous, intending to buy | 86.1% of Ana’s respondents hadn’t heard of Lantern |
| Attitude towards the product | Enthusiastic, positive, indifferent, negative, hostile | Loves canned coffee; would never drink it |
Readiness stage matters because each stage needs a different job from marketing. Of Ana’s 640 respondents, 551 (86.1%) were at the first stage, unaware of Lantern. Of the 89 who knew it, 50 hadn’t bought it in the past year and 39 had, a split by user status rather than readiness. Intent to buy Cold Brew (the top-two-box score: definitely or probably would buy) was 27.4% among the unaware against 40.0% among aware non-buyers and 41.0% among recent buyers: awareness first, then trial, is the obvious order of tasks, though the people who know Lantern may also differ in other ways.
Usage rate is the base where the 80/20 claim lives, and it deserves scepticism. Sharp, Romaniuk and Graham (2019), researchers at the Ehrenberg-Bass Institute and London South Bank University, report that a consumer brand’s heaviest 20% of buyers generally contribute not much more than half its sales: in an earlier analysis they cite, the average across categories was 59%, ranging from 44% for hair conditioner to 68% for dog food. The share also depends on the period: in one European panel study it was about 40% over a quarter, 50% over a year and 60% over five years, because a longer window counts more light buyers. The practical lesson is that light buyers usually matter more than the rule suggests, so check the concentration in your own data before deciding to market only to heavy users.
Usage also depends on what is counted. Lantern’s heavy drinkers by cups are not heavy by spend, because a cup at home costs a fraction of a café cup. For a café chain, spend is the measure that matters; for a bean supplier, volume might be. The course M06 Customer Analytics and RFM builds this idea into RFM scores and customer lifetime value, which combine how recently, how often and how much customers buy.
Why needs explain more, and why they are harder to reach
Benefits sought and needs are close to the reason for choice: someone who rates a low price 7 and local 2 will choose differently from someone who rates them the other way round, whatever their age. Haley’s argument was exactly this: segment first on what causes buying, then describe the segments. The cost is that needs are invisible. They exist only in survey answers, so a needs-based segment can’t be found in a census, bought as a media audience or recognised in a shop.
Two routes bridge the gap. The first is profiling: once needs-based segments are formed, describe them with descriptor variables that can be reached (age, region, where they buy, which media they use) and target through those, accepting some waste. The second is a typing tool, a few survey questions that assign a new customer to a segment, which module 2 builds for Lantern. For a business’s own customers there is a third: behaviour in purchase records (usage, occasions, loyalty) is both close to why and directly reachable, which is why loyalty-card and online data have made behavioural segmentation so common.
An attitude as a base: Lantern’s local-loyalty scale
The study measured loyalty to local businesses with three Likert items on a 1–5 scale (“I prefer to buy from local, independent businesses”, “I will pay a little more to support a local business” and “I tell friends about local businesses I like”). Their Cronbach’s alpha is 0.86, so the items hang together well, and each respondent’s score is the mean of the three, averaging 3.05. The score correlates at 0.52 with the separate rating of how important it is that coffee comes “from a local, independent business”: two measures of related ideas agreeing, a piece of evidence for the scale’s validity.
As an a priori base, cut the score into high loyalty (205 respondents scoring 3.67 or more, at least 11 of a possible 15 points), low loyalty (193 scoring 2.33 or less) and the middle. The attitude predicts the behaviour that expresses it: 47.3% of the highly loyal mainly get coffee from cafés against 17.6% of the least loyal (χ² = 40.9, p < 0.001, V = 0.25), and they spend $59.95 a month against $43.91. It barely predicts interest in a new canned product: 32.2% of the highly loyal would definitely or probably buy Lantern Cold Brew against 26.4% of the least loyal (p = 0.45, V = 0.05).
The lesson is general: an attitude base predicts behaviour in proportion to how closely the behaviour expresses the attitude. Loyalty to local businesses is a strong guide to who visits independent cafés and a weak guide to who buys a can in a supermarket, which is exactly the tension Lantern faces in deciding how to sell Cold Brew. Stated attitudes also run ahead of action, the gap M04 warned about with purchase intentions.
Business markets
In business markets the same ideas work inward from firmographics. OpenStax lists needs-based segmentation (what the customer needs from the product), value-based segmentation, also called tiering, by the value a customer brings from how much, how often and at what value it buys, and behavioural segmentation by how the customer engages with the seller, such as website visits and e-mail responses, often combined with value. In Shapiro and Bonoma’s nested approach, these sit in the inner rings: how the customer organises its buying, the situation of each purchase, and the personal characteristics of the people in the buying centre.
Formulas
Share of volume (or spend) from the heaviest buyers
- What buyer i bought in the period: units, cups, visits or dollarsunits or $
- The share of buyers counted as heaviest, after ranking all buyers from heaviest to lightest (q = 20% for the 80/20 check)%
- Share of the total bought by the heaviest q of buyers%
Assumes. Every buyer’s purchases are counted over the same period; buyers who bought nothing are left out (or the share is stated as of all customers, including them).
Excel, with purchases in B2:B641 and a cut-off in F1: =SUMIFS(B2:B641,B2:B641,">="&F1)/SUM(B2:B641). The share depends on the period: a longer window counts more light buyers and raises S_q.
Worked examples
Fictional figures
Cairnhaven Bakery has 2,000 loyalty-card holders. Ranked by loaves bought last year, the top 400 bought 18,400 loaves, the next 600 bought 13,800 and the remaining 1,000 bought 7,800. What share of loaves did the heaviest 20% buy?
- Find the heaviest 20%: 2,000 × 0.20 = 400 card holders, the top group.
- Total loaves: 18,400 + 13,800 + 7,800 = 40,000.
- Share: S = 18,400 ÷ 40,000 × 100 = 46.0%.
- Compare: the heaviest fifth buy 46.0% of loaves, not 80%; the lightest half buy 19.5%.
Answer. The heaviest 20% of card holders bought 46.0% of the loaves: concentrated, but far from 80/20.
What it means. This is close to the pattern Sharp, Romaniuk and Graham report for consumer brands, about half from the heaviest fifth over a year. A bakery that marketed only to its heaviest buyers would ignore customers who bring over half its sales.
Fictional figures
Pinemoor Leisure surveyed 800 members about a new evening social swim club. Members gave the main benefit they seek from membership and their age band, and said whether they would join. Which base, benefit sought or age, separates likely joiners better, and how should Pinemoor reach them?
| Main benefit sought | Would join | Would not | Members | Share who would join |
|---|---|---|---|---|
| Fitness | 45 | 255 | 300 | 15.0% |
| Relaxation | 80 | 170 | 250 | 32.0% |
| Meeting people | 105 | 45 | 150 | 70.0% |
| Family activities | 20 | 80 | 100 | 20.0% |
| All members | 250 | 550 | 800 | 31.3% |
- By benefit: the share who would join runs from 20.0% (family activities) and 15.0% (fitness) to 70.0% (meeting people). χ² = 147.7, 3 df, p < 0.001, Cramér’s V = 0.43.
- By age (18–34: 34.0% of 250; 35–54: 28.8% of 330; 55 and over: 31.8% of 220): χ² = 1.8, 2 df, p = 0.40, V = 0.05. Benefit sought is far more strongly associated with joining than age is, and age tells Pinemoor almost nothing.
- Size the segment: the “meeting people” members are 150 of 800, but they supply 105 of the 250 likely joiners (42.0%).
- Reach: a benefit can’t be seen on a membership card, and the social seekers are spread across ages, so age-based promotion would waste most of its reach. But they are members: Pinemoor can ask the benefit question at renewal, or use behaviour it already records, such as attendance at group classes, as a descriptor.
Answer. Benefit sought separates likely joiners far better than age (V = 0.43 against 0.05). Promote the club to members who seek to meet people, found through a renewal question or class attendance, not by age.
What it means. This is Haley’s argument in miniature: the benefit people seek is close to why they would join, and age is not. The work that remains is the one every needs-based segment needs, finding a reachable descriptor, and here a business’s own records make it unusually easy.
Fictional figures
Test the head barista’s theory that Lantern’s customers “drink a lot, care where it comes from and want to support places like ours”, using cups a week, spend and the local-loyalty scale.
- Heavy drinkers: the 20.8% who drink 16 or more cups a week drink 45.7% of the cups but account for only 20.8% of spending, and spend about the same per head as light drinkers ($50.34 against $49.25; t = 0.31, p = 0.76).
- What heavy drinkers want: they rate convenience (5.17 against 4.08 for light drinkers) and a caffeine kick (4.89 against 3.68) higher, and local lower (3.01 against 4.19). That is not a craft café’s customer.
- Loyal locals: the highly loyal (205 respondents) use cafés more (47.3% against 17.6%) and spend more ($59.95 against $43.91) than the least loyal.
- But for Cold Brew the attitude barely matters: top-two intent is 32.2% among the highly loyal and 26.4% among the least loyal (p = 0.45).
Answer. Half right. Loyalty to local businesses does mark valuable café customers; heavy drinking does not, because heavy drinkers buy cheap, convenient cups and care less about local. Neither base on its own identifies Cold Brew’s prospects.
What it means. Usage and attitude are useful bases, but each answers a different question: who spends in cafés, who drinks volume, who would try a can. That is the case for forming segments on several needs at once, which module 2 does, rather than on any single behaviour or attitude.
Explore
Common mistakes
“80% of our sales come from 20% of our customers.”
Check it. In Lantern’s study the 20.8% who drink 16 or more cups a week account for 45.7% of cups, and the top fifth by spend for 38.6% of spending; Sharp, Romaniuk and Graham find that a consumer brand’s heaviest 20% of buyers usually give not much more than half its sales. The share also grows with the length of the period measured.
“Heavy users are our most valuable customers.”
Only if “heavy” is measured in what matters to the business. Lantern’s heaviest drinkers by cups spend about the same per head as light drinkers ($50.34 against $49.25; t = 0.31, p = 0.76), because their cups are cheap ones. Rank by spend or margin when value is the question.
“A psychographic type tells us what people want from our product.”
Systems such as VALS classify general motivations and lifestyles, not needs in a category. Show with data that the types differ in your category before building segments on them, and be wary of vivid labels that promise more than the data show.
“Customers with the right attitude will buy our new product.”
Attitudes predict the behaviours that express them. Lantern’s local-loyalty scale predicts café use and spend, but Cold Brew intent barely differs between its high and low scorers (32.2% against 26.4%). Test the attitude against the behaviour you care about.
“Needs-based segments are better, so we don’t need demographics.”
Needs explain choice better, but they can’t be seen or bought as media. Needs-based segments still need demographic, geographic and behavioural descriptors, or a typing tool, before anyone can reach their members.
Guided practice
Work through these with hints when you need them. Every hint and every worked solution is there to be used.
Loading four questions…
On your own
No hints this time. Answer, check, and read the feedback; the solution appears once you have answered or if you ask for it. Your first answers count toward mastery.
Loading five questions…
In practice
Fictional case
Bellwick Grocers, a fictional regional supermarket, runs an online delivery service with about 260,000 orders a year. Freya Lindgren, its customer-insight manager, has been asked why online sales have stopped growing and what to do about it.
She starts with behaviour she can see. Order data gives every online customer’s usage rate, timing and basket. Grouping orders by occasion shows two patterns: small, frequent top-up orders (58.0% of orders) and large weekly shops (31.0%), with the rest irregular. Growth has stalled in the weekly shops, while top-ups are rising. Occasion turns out to be a better base than any customer demographic the service holds.
She checks the 80/20 claim before acting on it. The marketing team wants to concentrate on its heaviest customers. Freya ranks customers by spend over a year and finds that the top fifth account for about half of online sales, not 80%, so she argues against abandoning lighter customers, who are also the main source of new weekly shoppers.
She adds the why with a short survey. A survey of a sample of customers asks how important each of eight things is: delivery slots, price, substitutions, freshness and so on. Clustering the answers gives needs-based segments that cut across the occasions: some weekly shoppers care most about price, others about reliable slots. Because each surveyed customer is linked to their orders, Freya can describe each needs segment by its order behaviour, which makes the segments reachable through the bank of customers Bellwick already has.
She recommends what each segment needs. Price-led weekly shoppers get a delivery pass that removes the fee; slot-led ones get guaranteed evening slots; top-up customers get a lower minimum order. Her report states the limits: the survey sample may over-represent engaged customers, and the segments must be re-checked after six months, because occasion and need patterns change with seasons and prices.
Knowledge check
Six questions, one attempt each, no hints. Score 80% or more to complete the lesson; you can take it again at any time.
Loading six questions…
Summary
- Psychographic segmentation uses lifestyle (activities, interests, opinions), values and personality, measured with rating scales; VALS, from SRI and now Strategic Business Insights, describes eight US types that differ in resources and in what drives them.
- Behavioural segmentation uses benefits sought, usage rate, occasions, user status, loyalty status, readiness stage and attitude towards the product.
- Benefit segmentation (Haley, 1968) groups buyers by the benefits they seek, the reasons segments exist; needs-based segmentation is its modern, survey-and-cluster form.
- S_q = the heaviest q of buyers’ purchases ÷ all purchases. Check 80/20 against data: in Lantern’s study the 20.8% who drink 16 or more cups a week drink 45.7% of cups, and consumer brands typically get about half their sales from their heaviest 20%.
- Volume and value can rank customers differently: Lantern’s heavy drinkers by cups spend about the same per head as light drinkers ($50.34 against $49.25 a month).
- Needs and benefits explain choice better than demographics but are harder to reach; profiling with reachable descriptors, typing tools and customers’ own purchase records bridge the gap.
- An attitude base predicts the behaviours that express it: Lantern’s local-loyalty scale (alpha 0.86) predicts café use and spend, not Cold Brew intent.
- Business markets add needs-based, value-based (tiering) and behavioural bases inside the firmographic rings.
Key formulas
Terms introduced
- Psychographic segmentation
- Dividing a market by how people live and what they are like: their lifestyle (activities, interests and opinions), values and personality.
- Values
- A person’s lasting beliefs about what is important or desirable in life, such as security, achievement or protecting the environment, which guide their choices.
- Personality
- The distinctive and relatively stable pattern of traits that shapes how a person thinks, feels and behaves, such as sociability or openness to new experiences.
- Behavioural segmentation
- Dividing a market by what people seek and do in the category: the benefits sought, usage rate, occasions, user and loyalty status, readiness stage or attitude towards the product.
- Benefit segmentation
- Dividing buyers by the benefits they seek from a product, on the view that those benefits are the reasons segments exist; named by Russell Haley in 1968.
- Needs-based segmentation
- Forming segments from how important different needs or benefits are to each buyer, usually measured with importance ratings and grouped by cluster analysis; a post hoc form of benefit segmentation.
- Usage rate
- How much of a product a buyer uses or buys in a period, usually banded into non-users and light, medium and heavy users.
- User status
- A buyer’s relationship to a product or brand as a user: non-user, ex-user, potential user, first-time user or regular user.
- Loyalty status
- How consistently a buyer sticks to one brand or seller, from no loyalty (switching freely) to absolute loyalty.
- Buyer-readiness stage
- How far a person has moved towards buying a product: unaware, aware, informed, interested, desirous or intending to buy.
Sources
- Gomez Albrecht, M., Green, M. & Hoffman, L. (2023). Principles of Marketing, section 5.1, “Market Segmentation and Consumer Markets” (psychographic and behavioural segmentation, VALS, usage rate and the Pareto principle). OpenStax. openstax.org/books/principles-marketing/pages/5-1-market-seg…
- SRI International. “VALS™ market research” (history: launched 1978, new VALS 1989, now offered by Strategic Business Insights). sri.com/hoi/vals-market-research
- Haley, R. I. (1968). Benefit segmentation: a decision-oriented research tool. Journal of Marketing, 32(3). doi:10.1177/002224296803200306. Haley’s toothpaste segments as summarised in Weiss, A., “How to Segment Markets”, MarketingProfs. marketingprofs.com/Tutorials/Segmentation1.asp
- Sharp, B., Romaniuk, J. & Graham, C. (2019). Marketing’s 60/20 Pareto Law. Technical report, SSRN (Ehrenberg-Bass Institute and London South Bank University). doi:10.2139/ssrn.3498097. openresearch.lsbu.ac.uk/item/88vw1
- Kotler, P. (1986). Principles of Marketing, 3rd ed., Table 10-1, “Major segmentation variables for consumer markets”, as reproduced in a University of Washington course handout. faculty.washington.edu/harryb/courses/INFO%20Course/segmenta…
- Lastovicka, J. L. & Joachimsthaler, E. A. (1988). Improving the detection of personality-behavior relationships in consumer research. Journal of Consumer Research (March 1988). doi:10.1086/209138 (abstract checked in the Synapse index). doi.org/10.1086/209138
- Gomez Albrecht, M., Green, M. & Hoffman, L. (2023). Principles of Marketing, section 5.2, “Segmentation of B2B Markets”. OpenStax. openstax.org/books/principles-marketing/pages/5-2-segmentati…