In most companies, price is still treated as a number. A margin lever. A spreadsheet cell. A negotiation endpoint. A commercial decision that happens somewhere between finance, sales and the annual planning cycle.
That view is no longer good enough.
In a market defined by inflation fatigue, discount dependency, fragile loyalty and algorithmic comparison, price is not just what a brand charges. It is what a customer believes the brand is worth. It is a signal of quality, a test of trust, a behavioural trigger, a competitive weapon and, increasingly, a strategic expression of positioning.
That makes pricing one of marketing’s most consequential decisions. It also makes it one of the most misunderstood.
Marketers often say they are customer-centric. Pricing is where that claim is tested. Because the moment a price changes, the brand discovers what customers really value, which features matter, which competitors are credible, and how much equity the brand has actually built.
This is where individual choice models, conjoint analysis and Brand Price Trade-Off, or BPTO, become commercially useful. They help managers move beyond the blunt question “what price can we get away with?” toward a better one: “how do customers make trade-offs between brand, product attributes and price?”
That distinction matters. In mature markets, growth is rarely won by simply being cheaper. It is won by understanding what customers are willing to pay for, what they will compromise on, and where price becomes a reason to switch.
The pricing problem has changed
For years, many brands lived in a relatively forgiving pricing environment. Low inflation, predictable supply chains and expanding digital acquisition allowed businesses to manage price with a mix of category benchmarks, promotions and margin targets.
That world has changed. Consumers have become more price-aware, but not always more price-led. They compare more, switch more easily, and scrutinise value more closely. Yet the same customer who hunts for discounts in one category may pay a premium in another. Price sensitivity is not a personality trait. It is contextual.
A shopper may trade down on household basics and still pay for an expensive skincare brand. A commuter may reject a streaming price increase but accept a premium coffee subscription. A business buyer may resist a SaaS renewal increase while paying more for an implementation package that reduces risk.
This is why average elasticity is an insufficient guide. Managers need to understand how individual customers and segments choose. They need to know not only whether demand falls when price rises, but why, for whom, and against which alternatives.
That is the terrain of choice modelling.
From price elasticity to preference architecture
Traditional price elasticity tells a useful story: when price changes, demand changes. But it is often too aggregated to support modern marketing decisions. It can show that sales declined after a price rise, but not necessarily whether customers left because of the price itself, the brand, the bundle, the features, the competitor response, or the perceived fairness of the move.
Individual choice models approach the problem differently. They treat customer decisions as trade-offs. A consumer chooses between offers with different combinations of brand, price, features, pack size, design, service level, sustainability credentials, delivery speed or warranty.
The idea is grounded in random utility theory, associated with Daniel McFadden’s work on discrete choice analysis. In this framework, a consumer chooses the option that provides the highest utility, with some part of that utility observable to the researcher and some part unobserved. Marketing researchers then estimate how much each attribute contributes to choice.
That logic is powerful because it turns preference into a managerial map. It can show whether customers care more about brand than price, whether a feature justifies a premium, whether a cheaper competitor is genuinely threatening, or whether a new bundle could lift revenue without damaging share.
Conjoint analysis is one of the most widely used methods in this family. Since Paul Green and V. Srinivasan’s foundational reviews of conjoint analysis in marketing research, especially Conjoint Analysis in Consumer Research: Issues and Outlook (1978), the technique has become a core method for understanding how people value product and service attributes.
The enduring appeal is simple: conjoint analysis makes trade-offs visible.
What conjoint analysis actually does
Conjoint analysis asks respondents to evaluate product or service concepts made up of different attribute levels. Instead of asking “how important is price?” in isolation, it presents realistic alternatives and observes the choices or ratings people make.
A mobile plan might vary by brand, monthly price, data allowance, contract length and roaming benefits. A cereal product might vary by brand, price, health claim, pack size and flavour. A SaaS offer might vary by product tier, number of seats, support level, integrations and monthly fee.
From those responses, the researcher estimates part-worth utilities: the contribution each attribute level makes to preference. Price is included as one of the attributes, which allows managers to infer willingness to pay, price sensitivity and the value customers place on non-price features.
Choice-based conjoint, or CBC, is especially relevant because it asks respondents to choose among alternatives, more closely resembling market behaviour. Louviere and Woodworth’s work on choice experiments, and later advances in discrete choice modelling, helped establish the logic of experimentally designed choice sets as a way to estimate preferences and market shares.
In practice, conjoint analysis can answer questions such as:
Which features drive choice?
How much value does the brand name add?
What price premium can a stronger brand sustain?
Which customer segments are more price-sensitive?
Would a stripped-down cheaper offer grow the market or cannibalise the core product?
Which bundle maximises revenue rather than preference alone?
Where does the product become too expensive relative to perceived value?
That last point is central. Pricing is not just a demand problem. It is a value communication problem.
BPTO: a more direct test of brand-price switching
Brand Price Trade-Off, or BPTO, is a more focused pricing research method. It is designed to understand how consumers switch between brands as prices change.
A typical BPTO exercise presents consumers with a set of competing brands at defined prices. The respondent selects the brand they would buy. The price of the chosen brand is then increased, or competing prices are adjusted, and the respondent chooses again. Across repeated choices, the researcher observes at what point the consumer switches, to whom they switch, and how much price movement a brand can absorb before losing preference.
Where conjoint analysis decomposes value across multiple attributes, BPTO concentrates on the relationship between brand and price. It is particularly useful in categories where brand choice and price points dominate the decision: packaged goods, retail, telecoms, insurance, utilities, consumer electronics, travel and other competitive markets where buyers can compare alternatives quickly.
BPTO helps managers answer a different set of questions:
How much price premium can the brand sustain before customers switch?
Which competitor is the main beneficiary when the brand raises price?
Which brand has the strongest price resilience?
Where are the psychological thresholds?
Which customers are brand-loyal and which are deal-sensitive?
What happens to share if the category leader moves price first?
The method is valuable because it reveals the commercial meaning of brand equity. Brand strength is not only whether people like the brand. It is whether they still choose it when the price changes.
Conjoint vs BPTO: the useful comparison
Conjoint analysis and BPTO are sometimes treated as alternatives. They are better understood as different lenses.
Conjoint analysis is broader. It is useful when the decision involves multiple attributes and managers need to understand the relative value of features, benefits, brand and price. It is suited to product design, packaging, innovation, bundling, service tiers and market simulation.
BPTO is narrower and sharper. It is useful when the manager’s main concern is how consumers trade brand against price. It is suited to price increases, competitive pricing, premium resilience, private-label threats and brand-switching analysis.
Conjoint asks: what combination of attributes creates the most value?
BPTO asks: how far can price move before the brand loses the customer?
Conjoint is better for designing the offer. BPTO is better for stress-testing the price.
The two methods can be complementary. A brand planning a new product architecture might use conjoint analysis to define the best feature-price combinations, then use BPTO to test whether the final price points are defensible against competitors.
The managerial mistake is to use one method to answer the other method’s question. BPTO will not tell a brand which product features to build. Conjoint may not capture the simple brutality of shelf-level brand switching if the exercise is overdesigned or too abstract.
Why individual choice models matter for managers
The strategic value of individual choice models lies in segmentation. Price sensitivity is not uniform. It varies by customer, occasion, category, need state, income pressure, brand attachment, perceived risk and competitive context.
A single price elasticity can hide several markets inside one market.
One segment may be brand-loyal and relatively insensitive to price changes. Another may be highly promotion-driven. A third may care less about price but more about convenience. A fourth may be willing to pay more for sustainability, local sourcing, safety, service or design.
Conjoint analysis can estimate these preference structures. Hierarchical Bayes methods, widely used in conjoint research, allow researchers to estimate individual-level utilities even when each respondent answers only a limited number of tasks. This is important because it moves the analysis beyond averages and enables more precise segmentation.
For managers, that matters in several ways.
First, it improves product design. A business can see which attributes create value and which merely add cost. That can prevent the expensive mistake of over-engineering products customers will not pay for.
Second, it improves pricing architecture. Brands can design good-better-best tiers with clearer value fences. Instead of guessing what belongs in a premium tier, conjoint can show which features justify a price step.
Third, it improves promotional discipline. If BPTO shows that a brand has strong price resilience among core customers, excessive discounting may be unnecessary and damaging. If it shows that switching accelerates after a certain threshold, managers can avoid crossing a dangerous line.
Fourth, it improves competitive strategy. BPTO can identify the brands customers defect to when prices rise. That is more useful than broad category share data because it shows competitive substitution under pricing pressure.
Fifth, it improves communication. If conjoint shows that customers value a specific benefit but do not understand it, the pricing problem may actually be a messaging problem.
Pricing is not just willingness to pay
One of the most common misuses of conjoint analysis is to turn it into a mechanical willingness-to-pay machine. The logic is tempting: estimate the utility of price, estimate the utility of features, convert the result into monetary value, and declare that customers are willing to pay £X for a benefit.
That can be useful, but it must be handled carefully.
Willingness to pay in research is not always the same as willingness to buy in market. Survey respondents face hypothetical tasks. Real customers face budgets, habits, salespeople, reviews, stock availability, social influence and competitor promotions. A conjoint model is a structured approximation, not a crystal ball.
This is why external validation matters. Conjoint and BPTO should be calibrated against real sales data, experiments, A/B tests, historical price moves or market outcomes where possible. The model should inform pricing decisions, not replace commercial judgement.
There is also a difference between willingness to pay and willingness to accept a price increase. Existing customers often judge price changes through fairness, trust and reference prices. A new product may support a premium that an existing product cannot suddenly claim without damaging relationships.
This is increasingly visible in consumer markets. Consumers may accept premium pricing when the value story is clear, but react strongly to perceived “greedflation”, shrinkflation or subscription creep. The Business of Fashion’s 2025 report, “Why CMOs Should Have a Seat at the Pricing Table”, reflects this wider issue: pricing is no longer only a commercial or finance decision. It affects brand perception, customer trust and long-term demand.
Marketing has to be involved because pricing changes meaning.
The brand premium question
BPTO is especially useful because it brings brand equity down to earth.
Brand equity is often discussed in soft terms: awareness, preference, salience, love, trust. These matter. But pricing reveals whether they have economic force. A brand that consumers prefer only when prices are equal may have less equity than its tracking studies suggest. A brand that can hold share at a premium has more strategic room.
This is why BPTO can be uncomfortable. It exposes the difference between claimed loyalty and paid loyalty.
A consumer may say a brand is their favourite, but switch instantly when it is 10% more expensive. Another may rarely talk about the brand, but continue choosing it even after a price rise because they trust its reliability. BPTO reveals that behavioural resilience.
For premium brands, this is essential. The goal is not simply to charge more. It is to understand the conditions under which the premium is credible. A luxury, technology or personal-care brand may be able to sustain higher prices if customers perceive superior design, status, performance or trust. But if the premium is unsupported, BPTO will reveal the switching point.
For value brands, BPTO can show how much low price matters relative to familiarity, availability and perceived quality. A cheap brand is not automatically chosen if consumers fear poor performance. In categories with risk, such as baby care, financial services, health, travel or electronics, the lowest price can even be a warning signal.
The danger of discount addiction
The marketing industry has become good at promotion and less good at pricing confidence.
Discounts are easy to measure. They create short-term volume. They help sales teams close gaps. They make performance dashboards move. But they can also train customers to wait, weaken reference prices and undermine the very brand equity that allows a business to protect margin.
Choice models can help quantify this risk. If conjoint analysis shows that brand and product benefits carry strong utility, the business may not need aggressive discounting. If BPTO shows that loyal customers remain resilient across moderate price differences, promotions should be more targeted rather than blanket.
The opposite can also be true. If a brand’s BPTO curve collapses quickly when price rises, the issue may not be price execution. It may be weak differentiation. In that case, discounting is not a strategy. It is a symptom.
A useful pricing model should therefore inform not only the price point, but also the brand strategy. If customers will not pay more, the question is not only “what is the right price?” It is “why is the brand not worth more?”
Conjoint analysis across the marketing value chain
A 2025 paper by Marco Vriens and colleagues, Using Conjoint Analysis Across the Marketing Value Chain, reflects the continued relevance of conjoint analysis beyond narrow pricing research. The method can inform innovation, product development, communications, pricing, segmentation and portfolio management.
That broader use matters because pricing decisions are connected to the entire value chain. A price is not set after the offer is built. The offer is built so that the price makes sense.
For example, conjoint analysis can help a brand decide whether to invest in sustainable packaging, faster delivery, premium materials, extra data, AI features, extended warranties or human support. Each of these has a cost. The model helps estimate whether customers value the attribute enough to justify that cost.
In product portfolios, conjoint can help avoid cannibalisation. A cheaper tier may attract new customers, or it may simply downgrade existing ones. A premium tier may raise average revenue, or it may appeal to too few people to matter. Market simulation based on conjoint utilities can estimate these scenarios before launch.
For subscription businesses, conjoint can inform packaging. Which features belong in the entry plan? Which ones should be reserved for premium? Which price gap is acceptable? Which bundles look generous, and which feel manipulative?
For retailers, conjoint can evaluate private-label positioning. If consumers value national brands mainly because of perceived quality, a retailer can design private-label products that reduce that gap. If the national brand premium remains strong even at higher prices, the retailer may need a different strategy.
BPTO in inflationary markets
BPTO becomes especially relevant when prices are moving.
Inflation creates a noisy environment. Consumers expect some price increases but punish others. Brands need to know whether a price rise is likely to be absorbed, whether it will trigger switching, and which competitors are most dangerous.
A BPTO study can simulate these movements. It can show whether a brand has headroom, whether it is already at a fragile threshold, and whether competitors have more or less pricing power.
In categories such as consumer electronics, grocery, fashion and household goods, price sensitivity has become a visible feature of market behaviour. Numerator’s 2025 article, “Price Sensitivity Redefines the Consumer Electronics Market”, reflects the point: shoppers under confidence pressure become more intentional and value-focused. But “value-focused” does not always mean “cheapest”. It means customers are scrutinising the relationship between price, utility and trust.
That is precisely the relationship BPTO and conjoint analysis are designed to interrogate.
How managers should use these tools
The best use of conjoint analysis and BPTO begins with a managerial decision, not with a research technique.
A company should first clarify the decision at hand.
Is it deciding a launch price?
Testing a price increase?
Designing a tiered offer?
Quantifying brand premium?
Assessing private-label risk?
Understanding feature value?
Planning a promotional strategy?
Repositioning against a competitor?
Once the decision is clear, the method follows.
Use conjoint analysis when the offer has multiple attributes and the business needs to understand the structure of preference. Use choice-based conjoint when realistic choice behaviour matters. Use adaptive or menu-based conjoint when offers are complex or configurable. Use hierarchical Bayes estimation when individual-level utilities and segmentation are important.
Use BPTO when the key question is brand switching under price changes. It is particularly useful when the competitive set is clear and brand-price trade-offs dominate the buying decision.
Use both when the business needs to design the offer and then stress-test its price in a competitive context.
The output should not be a research deck alone. It should feed into decisions: price corridors, launch recommendations, offer architecture, promotional guardrails, sales scripts, messaging priorities and financial forecasts.
What good research design requires
Choice models are only as good as their design.
Attributes must be realistic. If price levels are implausible, results will mislead. If product features are poorly explained, respondents will ignore them. If the competitive set is incomplete, switching patterns will be distorted. If too many attributes are included, the exercise becomes cognitively exhausting.
Price levels need particular care. They should cover the realistic range of managerial action, including likely competitor positions. Too narrow a range makes elasticity hard to estimate. Too wide a range can produce artificial responses.
The sample also matters. A conjoint study among general consumers may be useless for a premium niche brand. A BPTO study among category buyers may miss future entrants. For B2B, respondents must reflect real decision-makers and buying committees, not just convenient contacts.
Finally, the model must be interpreted with humility. A choice model estimates structured preference under research conditions. It should be combined with behavioural data, sales history, experiments and managerial context.
The AI temptation
AI is entering pricing research quickly. It can improve survey design, simulate scenarios, process open-ended responses and combine structured research with behavioural data. It may also allow faster testing of product concepts and price architectures.
But there is a risk. AI can make weak assumptions look sophisticated. Synthetic respondents, automated pricing recommendations and black-box elasticity estimates are not substitutes for understanding real customer trade-offs.
Conjoint analysis and BPTO remain valuable precisely because they force structure. They make the business define the competitive set, the attributes, the price levels and the decision context. AI may accelerate the work, but it should not remove that discipline.
Pricing is too important to automate without judgment.
The comparative verdict
Individual choice models help managers understand pricing at the level where decisions actually happen: the customer’s trade-off.
Conjoint analysis is the broader strategic tool. It reveals how customers value attributes, brands and prices in combination. It is useful for innovation, product design, segmentation, bundling and willingness-to-pay analysis.
BPTO is the sharper pricing weapon. It reveals how customers switch between brands as prices move. It is useful for price increases, competitive response, brand premium assessment and identifying switching thresholds.
Traditional elasticity remains useful, but it is often too aggregated. It tells managers what happened to demand. Conjoint and BPTO help explain why demand may move and what to do about it.
The most important lesson is that price sensitivity is not simply a reaction to price. It is a reaction to perceived value. And perceived value is built by brand, product, context, trust and alternatives.
The final point: pricing is brand strategy in numerical form
Every price tells a story.
A premium price says the brand believes it has earned a premium. A discount says the brand is willing to trade margin for immediacy. A tiered architecture says the brand understands different needs. A sudden price rise says the brand believes customers will stay. A poorly justified increase says the brand may be taking trust for granted.
Conjoint analysis and BPTO give managers a more disciplined way to read that story before the market writes the ending.
They do not remove risk. They make risk more visible.
In a market where consumers are more informed, more pressured and more willing to switch, that visibility is valuable. Pricing can no longer be left as a back-office calculation. It has to be treated as a customer decision, a brand signal and a strategic growth lever.
The companies that understand this will not simply ask what the market will bear. They will ask what customers value, what competitors can credibly offer, and what their own brand has truly earned the right to charge.
That is the real promise of conjoint analysis and BPTO: not better research for its own sake, but better commercial judgment.
References
Green, P. E., & Rao, V. R. (1971). “Conjoint Measurement for Quantifying Judgmental Data.” Journal of Marketing Research.
Green, P. E., & Srinivasan, V. (1978). “Conjoint Analysis in Consumer Research: Issues and Outlook.” Journal of Consumer Research.
Green, P. E., & Srinivasan, V. (1990). “Conjoint Analysis in Marketing: New Developments with Implications for Research and Practice.” Journal of Marketing.
Louviere, J. J., & Woodworth, G. (1983). “Design and Analysis of Simulated Consumer Choice or Allocation Experiments: An Approach Based on Aggregate Data.” Journal of Marketing Research.
McFadden, D. (1974). “Conditional Logit Analysis of Qualitative Choice Behavior.” In Frontiers in Econometrics.
Orme, B. K. (2010). Getting Started with Conjoint Analysis: Strategies for Product Design and Pricing Research. Research Publishers.
Rao, V. R. (2014). Applied Conjoint Analysis. Springer.
Vriens, M., et al. (2025). “Using Conjoint Analysis Across the Marketing Value Chain.” Applied Marketing Analytics.
Wittink, D. R., & Cattin, P. (1989). “Commercial Use of Conjoint Analysis: An Update.” Journal of Marketing.
The Business of Fashion. (2025). “Why CMOs Should Have a Seat at the Pricing Table.”
Numerator. (2025). “Price Sensitivity Redefines the Consumer Electronics Market.”
Conjointly. (2025). “Comparing Four Methods of Conjoint for Pricing Research.”
DJS Research. “Brand Price Trade Off.”
