Advanced marketing · Module 3

Pricing strategy

How should a firm set prices when customers value the product differently, rivals respond, and every price also sends a signal?

Beyond the IB course UndergraduateMBAProfessional About 18 minutes · 2,998 words · checked 16 November 2026

From the IB course to this module

The IB course covers cost-plus, penetration, loss leader, predatory and premium pricing and, at HL, dynamic, competitive and contribution pricing with price elasticity of demand (4.5), alongside costs and revenues (3.3) and break-even (5.5). This module develops value-based pricing, the arithmetic linking price, volume and profit, price discrimination, the psychology of price, and the limits set by fairness and regulation.

Section 01: Conceptual foundation

Price is the marketing decision that most directly captures value. Marn and Rosiello (1992), analysing the accounts of large companies, found that a 1% rise in price with no loss of volume would raise operating profit by about 11% on average. The catch is the condition about volume: customers respond to prices, and so do rivals.

Three considerations bound a price. Cost sets the floor below which a sale loses money: variable cost in the short run, full cost in the long run. Customer value sets the ceiling, since no customer pays more than a product is worth to them compared with the alternatives. Competition decides where between the two a price can settle. Cost-plus pricing starts from the floor and ignores the ceiling; value-based pricing starts from the ceiling.

Because customers value products differently, a single price leaves money on the table: some would have paid more, and others who would have paid a lower price, still above cost, are lost. Price discrimination, charging different customers different prices for the same or similar products, captures more of that value.

Prices also carry meaning: customers infer quality from price, judge prices against reference points and react strongly to prices they consider unfair (Kahneman, Knetsch and Thaler, 1986).

Section 02: Advanced explanation

Price, volume and profit: the arithmetic

The first question about any price change is how much volume it must win, or may lose, before profit suffers. If contribution is CM% of the price, a price change of ΔP% leaves total contribution unchanged when volume changes by:

volume change needed = −ΔP% ÷ (CM% + ΔP%)

A worked example. A product sells at £50 with a variable cost of £30, so contribution is £20, or 40% of price. A 10% price cut needs 10 ÷ (40 − 10) = 33.3% more volume just to stand still; a 10% rise can afford to lose 10 ÷ (40 + 10) = 20% of volume. The thinner the margin, the more a price cut must sell and the less volume a rise can afford to lose.

Compare elasticity. If the price elasticity of demand is −2, a 10% cut raises volume by about 20%, short of the 33.3% needed, so contribution falls even though revenue rises: revenue is the wrong test of a price change.

Economics adds a rule for the profit-maximizing price when elasticity is constant: the markup over marginal cost, as a share of price, equals the inverse of the elasticity (Lerner, 1934), so P = MC × |e| ÷ (|e| − 1). With marginal cost of £30 and elasticity of −3, P = 30 × 3 ÷ 2 = £45; with elasticity of −2, P = £60. Less price-sensitive customers support higher markups, which is why differentiation and brand matter for price.

Finally, the list price is rarely the price received. Marn and Rosiello's price waterfall traces the leakage from list price to pocket price through discounts, rebates, payment terms, free freight and promotional allowances, many granted customer by customer and never added up.

Value-based pricing

Value-based pricing starts from the customer's next best alternative. Economic value to the customer (EVC) is the price of that alternative, the reference value, plus the value of everything that makes the product better, minus anything that makes it worse.

A worked example. A new industrial pump competes with the market leader's, which sells for £9,000. Over a five-year life the new pump saves £1,500 a year in energy and £400 a year in maintenance. Discounted at 8%, those savings are worth £1,900 × 3.993 = £7,586, so the EVC is about £16,586. Any price from £9,000 to £16,586 leaves the customer better off than buying the incumbent. A price of about £12,800, sharing the extra value roughly equally, gives customers a clear reason to switch while capturing far more than a cost-plus price would.

The method demands that the firm quantify value in the customer's terms, communicate it, since customers will not credit savings they cannot see, and segment by value, because the same pump saves more for a customer who runs it day and night.

Willingness to pay can be estimated by conjoint analysis, which infers the value of features from choices between hypothetical products, by direct questions such as the price-sensitivity meter, and by tests in the market, the most reliable of the three.

Price discrimination: segments, versions, bundles and dynamics

Economists distinguish three degrees of price discrimination. First degree charges each customer their own willingness to pay, the logic of personalized pricing. Second degree offers a menu, such as product versions or quantity discounts, and lets customers sort themselves. Third degree charges different prices to identifiable groups, such as students, regions, or new and existing customers. All three need some market power, a way of telling customers apart or making them sort themselves, and a way to stop low-price buyers reselling to high-price ones.

Versioning (Shapiro and Varian, 1999) designs a product line so that the customers who value the product most choose the premium version: good, better and best tiers, or a professional edition with features that matter only to heavy users. The basic version is sometimes deliberately limited so that it does not cannibalize the premium.

Bundling extracts value when customers value products differently, as Adams and Yellen (1976) showed. Suppose one customer values product A at £60 and B at £30, and another values A at £30 and B at £60, with negligible costs. Priced separately, each product earns at most £120 in total. Sold as a bundle at £90, both customers buy and the firm earns £180, because the bundle averages out the differences in valuation.

Dynamic pricing changes prices with demand, time or capacity, as airlines and hotels do. It fills capacity efficiently, but it becomes a flashpoint when prices jump at moments of need, or when the customers paying most are simply those who did not shop around.

The psychology and ethics of price

Customers do not judge prices in isolation. Prospect theory (Kahneman and Tversky, 1979) implies that prices are judged against a reference price, and that paying more than the reference feels like a loss, weighted more heavily than an equal saving. Framing therefore matters: a discount and a surcharge with the same economics are perceived differently, and the end of an introductory offer can feel like a price rise. Thaler's (1985) mental accounting explains why combining small charges into one, or splitting a large saving into several, changes how a deal feels.

Endings matter too. In field experiments with a mail-order catalogue, Anderson and Simester (2003) found that prices ending in 9 raised demand compared with nearby prices, especially for new items. Prices also signal quality where customers cannot judge it directly, so a very low price can lose sales.

Fairness constrains pricing power. Kahneman, Knetsch and Thaler (1986) found that people accept price rises that pass on higher costs but regard rises that exploit a surge in demand as unfair, even when the economics are identical.

Regulation sets harder limits. Competition law restricts predatory pricing and collusion, and research shows that pricing algorithms can learn to sustain high prices without any explicit agreement (Calvano, Calzolari, Denicolò and Pastorello, 2020). Consumer law increasingly restricts hidden charges, misleading reference prices and, in some markets, charging loyal customers more than new ones.

Section 03: Models and frameworks

Where each comes from, how to use it, and where it stops being useful.

Nagle and Holden (1995)

Break-even volume for a price change

A price change of ΔP% leaves contribution unchanged if volume changes by −ΔP% ÷ (CM% + ΔP%), where CM% is the contribution margin as a share of price. Compare the result with the volume change the market will plausibly deliver.

How to use it

Before any price change, calculate the break-even volume change and set it beside the best available estimate of elasticity. If the needed volume is implausible, the change will lose contribution whatever it does to revenue.

Where it fails

It is static: it ignores competitors' responses, effects on other products, capacity limits and the longer-run effect of price on brand and reference prices.

Thomas T. Nagle and Reed K. Holden (1995). The Strategy and Tactics of Pricing: A Guide to Profitable Decision Making (second edition). Prentice Hall.

Forbis and Mehta (1981)

Economic value to the customer

The most a rational customer would pay is the price of the next best alternative plus the net value of the product's differences. The gap between that value and cost is the space within which price divides the gains.

How to use it

Identify the reference product, quantify each difference in money over the product's life, and decide what share of the differentiation value to capture. Use the calculation in sales conversations, not only in the pricing department.

Where it fails

It suits business markets where value can be counted. For consumer goods with emotional or social value it is much harder to apply, and customers may not trust the supplier's estimate of their savings.

John L. Forbis and Nitin T. Mehta (1981). Value-based strategies for industrial products. Business Horizons, 24(3).

Marn and Rosiello (1992)

The price waterfall

Between list price and the pocket price actually received lie layers of on-invoice discounts and off-invoice leakage: rebates, payment terms, freight, promotional allowances. Pocket prices for the same product often vary widely between customers.

How to use it

Build the waterfall for each major customer and plot pocket price against volume. Customers who receive large concessions without buying much reveal where discipline has slipped.

Where it fails

It needs transaction data that many firms do not collect in one place, and it is a tool for capturing value already created, not for deciding what value to create.

Michael V. Marn and Robert L. Rosiello (1992). Managing price, gaining profit. Harvard Business Review, 70(5).

Kahneman and Tversky (1979); Kahneman, Knetsch and Thaler (1986)

Reference prices and fairness

Prices are judged against reference points; losses relative to the reference weigh more than gains. Customers accept increases justified by costs but resent increases that exploit demand or circumstances.

How to use it

Before a change, ask what reference customers will use and how the change will be framed against it. Explain cost-driven rises; avoid rises that look like exploitation of urgency or loyalty.

Where it fails

Fairness norms vary between markets and cultures and shift as practices such as dynamic pricing become familiar. Laboratory effects do not always translate into lost sales.

Daniel Kahneman and Amos Tversky (1979). Prospect theory: an analysis of decision under risk. Econometrica, 47(2). Daniel Kahneman, Jack L. Knetsch and Richard Thaler (1986). Fairness as a constraint on profit seeking: entitlements in the market. American Economic Review, 76(4).

Section 04: Real business applications

Real organizations appear only with dated, cited facts; illustrative examples are labelled.

Real organization, 2025-07

Insurance renewals: a regulator bans price walking and checks the result (2021 to 2025)

In a market study of home and motor insurance, the FCA found that firms raised prices year after year for customers who renewed, a practice known as price walking, while offering cheaper deals to new customers. An online experiment it published in 2021 found that promotions resembling cash, such as vouchers and cashback, made it harder for consumers to pick the best deal. From 1 January 2022 its rules required insurers to offer renewing customers a price no higher than an equivalent new customer would pay, with such promotions counted in the comparison. It expected consumers to save £4.2 billion over ten years.

Its July 2025 evaluation found that in home insurance the average extra paid by existing customers over new ones had almost halved, from £95.38 to £49.17. In motor insurance, where renewing customers are on average lower risks, new customers' prices rose by £111.14 against £22.71 for existing customers. The FCA estimated that the remedy cut motor prices by £6.63 per policy on average, a central £1.6 billion over ten years, but found no statistically significant effect on home prices, and its evidence on product quality was inconclusive.

Interpretation. Price walking was third-degree discrimination based on inattention. Banning it did not end cross-subsidy so much as move prices towards cost and risk, with new customers now paying more.

Real organization, 2026-03

Autotrader: an annual price event, its levers and a backlash (2025 to 2026)

Autotrader, the UK's largest car marketplace, earns most of its revenue from retailers who pay monthly to advertise. In the year to 31 March 2026 its average revenue per retailer rose 5%, or £141, to £2,995 a month. The company splits the change into three levers. Price added £117, from its annual pricing and product event in April 2025, which combined a like-for-like price rise with additional products. Product added £72, mainly from including a generative AI tool, Co-Driver, in retailers' packages. Stock subtracted £48, as some retailers cut the number of cars they advertised, a response the company linked to the speed and nature of its roll-out of a new product, Deal Builder, and to harder trading conditions. Average retailer forecourts fell 0.5% to 13,942.

Interpretation. A platform with a strong market position can raise prices every year, especially when new features justify the rise. But customers can push back through quantity rather than by leaving: here the stock lever offset about 40% of the gain from price.

Real organization, 2026-03

Wegovy in the United States: one medicine, many prices (2025 to 2026)

Most US patients prescribed Wegovy, Novo Nordisk's obesity medicine, pay through insurance; in March 2025 the company said 90% of covered patients paid $0 to $25 a month. For patients paying cash it launched a direct-to-patient pharmacy that month, offering all doses at $499 a month and contrasting its authentic, approved medicine with compounded copies. In November 2025, after an agreement with the US Administration, it cut the standard self-pay price to $349 and offered new patients $199 a month for their first two months on the two lowest doses. In March 2026 it added subscriptions through telehealth providers: $329 a month for three months, $299 for six and $249 for twelve, a saving of up to $1,200 a year against the standard price.

Interpretation. One medicine carries many prices: by channel (insured or self-pay), by stage (introductory or standard) and by commitment (one month or twelve). The subscription discount rewards staying on treatment, which for the seller is also retention.

Section 05: Case studies

Fictional case, simulated figures

Northgate Tours: fill the coach, or protect the price?

Northgate Tours, a fictional operator, runs weekend coach trips from Manchester to the Lake District. Each coach has 49 seats, and seats unsold at departure are worth nothing. The sales manager believes a lower price would fill more seats and grow the business; the owner worries about margins. Last season the company tested two prices against its current one on matched departures.

Price tests on weekend departures (averages per departure)
Price test Price per seat (£) Seats sold Variable cost per seat (£) Fixed cost per departure (£)
Current price12038452,100
Test A: price cut9947452,100
Test B: price rise13533452,100

Simulated figures. Each price ran on ten departures; four of the ten Test A departures sold out.

Questions

  1. Calculate contribution and profit per departure at each price. Which earns most?
  2. Calculate the price elasticity implied by each test and the rise in seats the price cut would need to keep contribution unchanged. Why can the cut never pay on a 49-seat coach?
  3. Four Test A departures sold out. What does that imply about the measured elasticity, and about using a dynamic price rather than a single one?
  4. Design a dynamic pricing rule for the coaches and explain how you would keep it within customers' sense of fairness.

Section 06: Critical evaluation

What the evidence supports, where the debate is, and what the models leave out.

What is established. Demand responds to price, and more than many managers assume. Meta-analyses of econometric studies put the average price elasticity of brands at about −1.8 (Tellis, 1988) and, in a later and larger review, about −2.6 (Bijmolt, van Heerde and Pieters, 2005), with wide variation by category and over time. The arithmetic of contribution and break-even volume is not in dispute; it is simply ignored more often than it is contested.

Measurement. Stated willingness to pay overstates real willingness because nothing is at stake. Market tests are better but can be distorted by capacity limits, as when sold-out departures hide demand, and by rivals' reactions.

Behavioural pricing. The core findings on reference prices and fairness are robust, but the size of many behavioural effects depends on context. Some popular claims have weakened under scrutiny: a meta-analysis of choice overload, the idea that too many options deter purchase, found an average effect close to zero (Scheibehenne, Greifeneder and Todd, 2010), which matters for decisions about product lines and tiers.

Discrimination and welfare. Price discrimination can serve customers who would otherwise be priced out, or extract surplus from the inattentive and vulnerable, depending on its basis. The FCA's evaluation is unusual in measuring a pricing rule's effect after the event, and its mixed findings show that interventions rarely work exactly as predicted.

What the models leave out. Most pricing models treat the product as fixed; in practice price, product design and channel are set together, and the most profitable changes often repackage the offer rather than change the number.

Section 07: Decision exercise

A fictional situation. Decide before you read the guidance.

Raising the price of a business software product

Ledgerline, a fictional maker of accounting software for small firms, has not raised its single price of £40 per user a month in four years, while costs and the product's features have grown. About a fifth of its customers are accountancy practices that use it heavily for many clients; the rest are small firms with light needs. A competitor charges £35. The board wants more revenue next year without a surge in cancellations.

  1. A. Raise the price for everyone

    Increase the single price by 8% at each customer's renewal, with a letter explaining the new features added since the last rise.

  2. B. Good, better, best

    Keep £40 for a standard tier, add a £30 basic tier with fewer features, and move multi-client and advanced reporting features to a £60 professional tier.

  3. C. Price by usage

    Replace the per-user price with a charge per transaction processed, so that heavy users pay more and light users less.

Judge the options on

  • Effect on revenue and contribution
  • Cancellations and customers' sense of fairness
  • Ease of explaining and selling the change
  • Long-run capture of value as customers grow
  • Competitive response
Read the guidance

Customers differ sharply in the value they get, the signal for price discrimination rather than a single higher price. A is simplest and its explanation will be accepted by many, but it bears hardest on the light users closest to switching to the £35 competitor, and leaves the practices, who get most value, paying the same as everyone else.

B is versioning: practices self-select into the professional tier because they need its features, light users can trade down rather than leave, and the standard tier anchors the middle. It needs careful design so that the basic tier does not tempt customers who would otherwise pay £40, and the move of features into the top tier must not feel like taking something away; grandfathering existing practices for a year softens that.

C aligns price with value most closely and grows revenue as customers grow, but it makes bills unpredictable, which small firms dislike, and it is the hardest change to explain. A strong answer is B, tested first on new customers, with C kept under review for the largest practices.

Section 08: Further reading

  • Thomas T. Nagle and Reed K. Holden (1995). The Strategy and Tactics of Pricing: A Guide to Profitable Decision Making (second edition). Prentice Hall.

    The standard practitioner text on value-based pricing and the economics of price changes.

  • Michael V. Marn and Robert L. Rosiello (1992). Managing price, gaining profit. Harvard Business Review, 70(5).

    The price waterfall and the case for disciplined pricing; short and still practical.

  • Carl Shapiro and Hal R. Varian (1999). Information Rules: A Strategic Guide to the Network Economy. Harvard Business School Press.

    Versioning, bundling and personalized pricing for information goods, clearly explained.

  • Daniel Kahneman, Jack L. Knetsch and Richard Thaler (1986). Fairness as a constraint on profit seeking: entitlements in the market. American Economic Review, 76(4).

    The experimental evidence on what customers regard as fair pricing.

  • William James Adams and Janet L. Yellen (1976). Commodity bundling and the burden of monopoly. Quarterly Journal of Economics, 90(3).

    The classic analysis of why bundling can capture value.

  • Tammo H. A. Bijmolt, Harald J. van Heerde and Rik G. M. Pieters (2005). New empirical generalizations on the determinants of price elasticity. Journal of Marketing Research, 42(2).

    A large meta-analysis of price elasticities and what makes them larger or smaller.

Every work cited here is listed in the platform's sources register. Real organizations appear only with facts from the sources shown; everything else is labelled illustrative or fictional.