Pricing, Promotions and Revenue Management

Introduction and Context

Price is the only element of the marketing mix that directly generates revenue. In e-commerce, however, pricing is not simply the act of assigning a number to a product. It is a coordinated managerial process involving customer value, costs, competitor behaviour, demand uncertainty, inventory, technology, promotion design and legal constraints. Digital markets make prices highly visible and easy to change. A retailer can test thousands of price points, personalise offers and respond to competitors within minutes. These capabilities create opportunities for profitable growth, but they also create risks: price wars, margin erosion, customer distrust, discrimination and regulatory scrutiny.

Pricing decisions should therefore be evaluated at three levels. First, the firm must determine whether the price covers relevant costs and supports its strategic objectives. Second, it must assess how customers and competitors are likely to respond. Third, it must manage the timing and structure of prices across products, channels and customer segments. Revenue management integrates these decisions by allocating limited capacity or inventory to the most profitable demand over time.

Learning Objectives

By the end of this chapter, you should be able to:

1. Apply cost-, competition- and value-based pricing approaches.
2. Calculate and interpret contribution, break-even volume and price elasticity.
3. Evaluate dynamic pricing, markdowns, bundling, coupons and promotional calendars.
4. Distinguish revenue from profitable revenue using unit economics and customer metrics.
5. Design a basic revenue-management process for an online marketplace or retailer.
6. Recognise legal, ethical and governance limits on algorithmic pricing.

1. The Economics of an E-Commerce Price

A price communicates economic value, positioning and fairness. The customer's relevant comparison is not always the listed price. It may be the total delivered price, including shipping, taxes, subscriptions, payment fees and the value of waiting. A retailer should therefore monitor both listed price and landed price.

Important terminology includes:

• List price: The displayed or reference price before discounts.
• Net price: The amount retained after discounts, rebates and credits.
• Variable cost: A cost that changes with each additional unit, such as payment processing or fulfilment.
• Fixed cost: A cost that does not vary directly with short-run unit volume, such as platform development.
• Contribution margin: The amount available from each sale to cover fixed costs and profit.
• Price architecture: The structured relationship among prices across products, versions, quantities and customer segments.
• Reference price: The price a customer expects or uses for comparison.

For a unit sold at price P, with variable cost V, unit contribution is:

Here, C, P and V are measured in currency per unit. This formula assumes that variable cost is correctly identified and that the sale does not create additional variable costs omitted from the calculation. If a 50 currency-unit product has a variable cost of 32, its contribution is 18 currency units. A discount that reduces price by 5 reduces contribution by 5 unless it also reduces variable cost.

Contribution margin percentage is:

It indicates the proportion of sales revenue remaining after variable costs. It should not be confused with gross margin when the firm's cost classifications differ.

Cost-Based Pricing

Cost-plus pricing sets price by adding a markup to cost:

C₀ is the chosen cost base and m is the markup rate expressed as a decimal. If cost is 40 and markup is 25%, price is 50. The method is simple and useful where costs are stable, demand is predictable and the offer is not strongly differentiated. Its weakness is that it may ignore willingness to pay, competitor response and demand effects. A markup on cost also does not necessarily produce a desired margin on sales. A 25% markup gives a margin of 20% because 10 ÷ 50 = 20%.

If the target is a margin on selling price, the required price is:

where g is the target margin as a decimal. For a cost of 40 and a target margin of 25%, price is 53.33, not 50. This distinction is essential in e-commerce dashboards.

Competition-Based Pricing

Competition-based pricing uses observed rival prices as a reference. A firm may price below, at or above the market depending on its service, brand, delivery speed, assortment and customer experience. Competitive intelligence should compare equivalent products and total delivered cost rather than headline prices alone. Automated competitor matching can be useful, but blindly matching a competitor may amplify errors, respond to temporary stock-clearance prices or trigger a price war.

Value-Based Pricing

Value-based pricing begins with the benefits perceived by a defined customer segment. A customer may value time saved, reliability, status, lower risk, convenience or an outcome rather than the physical product alone. The firm estimates willingness to pay through interviews, conjoint analysis, experiments, historical data and usage research. Value-based pricing is particularly powerful for differentiated products, subscriptions and business services.

A practical value model asks:

1. What customer problem is solved?
2. What is the customer's best alternative?
3. What measurable economic or emotional benefit is created?
4. Which features are essential, attractive or irrelevant?
5. What evidence supports the estimated willingness to pay?

Value-based pricing does not mean charging the maximum imaginable price. It requires a defensible relationship between benefits, alternatives and price, supported by clear communication.

2. Demand, Elasticity and Profitability

Demand usually changes when price changes, although the direction and size vary by category, customer segment, season and competitive context. Price elasticity of demand measures the percentage response in quantity to a percentage change in price:

Using a midpoint method for two observations:

P₁ and P₂ are the two prices; Q₁ and Q₂ are corresponding quantities. The elasticity is commonly negative because quantity falls as price rises. The midpoint method reduces dependence on which observation is treated as the starting point. It assumes that other important influences, such as advertising, stock availability and competitor prices, are sufficiently controlled or adjusted for.

If price rises from 20 to 22 and quantity falls from 1,000 to 900, the percentage price change is 2 ÷ 21 = 9.52%, and the percentage quantity change is −100 ÷ 950 = −10.53%. Elasticity is approximately −1.11. Demand is therefore elastic in this observed interval. Elasticity is not a permanent characteristic of a product: it can differ by customer, device, time and price range.

Revenue is:

In the example, revenue changes from 20,000 to 19,800. Revenue falls despite the higher price because quantity declines proportionately more. Profit requires costs as well as revenue:

π is operating profit, V is variable cost per unit and F is fixed cost over the relevant period. This model assumes one product or a constant weighted average contribution and excludes taxes and financing costs unless explicitly included.

Worked Break-Even Example

Suppose an online retailer sells a product for 60, has variable cost of 36 and fixed monthly costs of 24,000. Contribution is 24 per unit. Break-even volume is:

Thus, Q_BE = 24,000 ÷ 24 = 1,000 units. At 1,300 units, operating profit is (60 − 36) × 1,300 − 24,000 = 7,200 currency units. If a coupon lowers price to 54 and variable cost remains 36, contribution becomes 18; break-even volume rises to 1,333.33, meaning at least 1,334 whole units must be sold to exceed break-even. The promotion must generate sufficient incremental volume or other strategic value to justify this loss of contribution.

3. Promotional Pricing and Price Architecture

Promotions can acquire customers, accelerate trial, clear inventory, respond to competitors, reward loyalty or increase basket size. They can also train customers to wait, reduce reference prices and shift purchases forward without increasing long-run demand.

Coupons and Discounts

A coupon may be a percentage discount, fixed amount, free delivery offer, gift, cashback reward or conditional incentive. Effective design specifies eligibility, minimum basket value, expiry, redemption limits, funding responsibility and treatment of returns. The relevant measure is incremental contribution, not redemption volume alone.

A promotion should be assessed against a baseline. Incremental contribution can be represented as:

Q_p and C_p are promoted quantity and contribution per unit; Q₀ and C₀ are baseline quantity and contribution; K is incremental promotional cost. The calculation must account for cannibalisation, subsidised sales that would have occurred anyway, returns and post-promotion demand decline.

Markdowns

A markdown is a planned price reduction, often used for seasonal, fashion, perishable or obsolete inventory. The markdown decision compares the contribution from selling now with the expected contribution from waiting, including holding cost and the probability of later sale. It should not be based solely on the original cost or list price, which may be sunk or irrelevant.

A simple expected-value comparison is:

The current sale is preferable when its contribution exceeds the expected value of waiting. This is a simplified model: capacity constraints, customer experience and liquidation costs may also matter.

Bundling

Bundling combines products and charges one price. Pure bundling makes the items available only together; mixed bundling offers both the bundle and separate items. Bundles can increase average order value, simplify choice, use complementary products and reduce fulfilment cost. They can also obscure unit prices or force customers to buy unwanted items.

Bundle evaluation should compare bundle contribution with the contribution that would have been earned from separate purchases, while considering cannibalisation and inventory. A bundle is not automatically profitable merely because its price is higher than the cost of its components.

Promotional Calendars

A promotional calendar coordinates events across products, channels and periods. Good calendars avoid overlapping discounts that make attribution impossible and prevent excessive discount frequency. They should identify objectives, target segments, offer mechanics, inventory assumptions, communication dates, eligibility rules, measurement windows and stop conditions.

A useful planning sequence is:

1. Forecast baseline demand.
2. Identify strategic objective and eligible customers.
3. Estimate expected volume, margin, inventory and operational capacity.
4. Define price floors and approval limits.
5. Launch with tracking tags and control groups where feasible.
6. Monitor sales, margin, cancellations, returns and customer complaints.
7. Evaluate incremental rather than reported performance.
8. Record learning for the next cycle.

4. Dynamic Pricing and Revenue Management

Dynamic pricing changes prices in response to time, demand, inventory, capacity, competitor conditions or customer context. Common applications include travel, ride services, event tickets, food delivery, advertising and limited-stock retail. Dynamic pricing is not synonymous with individualised pricing. A firm may change one public price for everyone based on aggregate demand, whereas individualised pricing varies prices by customer or inferred willingness to pay.

Revenue management is especially relevant when inventory is perishable or capacity is fixed. An unsold hotel room tonight cannot be sold tomorrow for the same night; an unused delivery slot may also be lost. The firm attempts to allocate capacity among demand classes while protecting availability for potentially higher-value demand.

A basic revenue-management process includes:

• demand forecasting by product, time and segment;
• inventory and capacity monitoring;
• price or availability rules;
• constraint management, such as minimum order quantities or booking limits;
• real-time performance monitoring; and
• post-period forecast and model evaluation.

A simple expected contribution for a price option is:

where E[Q(P)] is expected quantity at price P. The firm should select the option with the highest expected contribution subject to capacity, legal and strategic constraints. Forecast uncertainty means that confidence intervals, scenario analysis and human review are preferable to treating a single model prediction as fact.

Algorithms and AI

Pricing systems may use rules, regression, time-series forecasting, machine learning or reinforcement learning. Inputs can include historical sales, inventory, seasonality, traffic, conversion, competitor prices and delivery capacity. AI can improve forecasts and identify segments, but it can also learn historical bias, infer sensitive characteristics, or create unstable price movements.

Good governance requires data minimisation, documented objectives, model validation, monitoring for disparate outcomes, access controls, audit logs, override procedures and clear accountability. Employees should understand when the model is outside its training conditions. A technically accurate forecast does not justify an unlawful or unfair pricing policy.

5. Legal, Ethical and Consumer-Trust Boundaries

Pricing is constrained by competition law, consumer-protection law, privacy and data-protection rules, sector-specific regulation and contractual obligations. Exact requirements differ by jurisdiction, so managers must obtain local legal advice. Common risk areas include:

• Misleading reference prices: presenting an inflated “original” price that was not genuine or commercially available.
• Hidden charges: adding compulsory fees late in checkout rather than displaying the total price clearly.
• False scarcity: claiming limited stock or time when the condition is not real.
• Deceptive countdowns: restarting timers to create artificial urgency.
• Unfair discrimination: using protected or sensitive attributes, or proxies for them, to produce harmful price differences.
• Collusion risks: using algorithms that coordinate with competitors or implement an agreement to maintain prices.
• Resale-price restrictions: pressuring marketplace sellers to maintain a particular retail price where prohibited.
• Dark patterns: designing interfaces that make a discount or subscription difficult to understand or refuse.

Ethical pricing asks not only whether an action is legal but whether it is transparent, proportionate and consistent with reasonable customer expectations. A useful fairness test is whether the firm could explain the pricing rule publicly to customers, employees, regulators and investors without misleading them. Firms should distinguish legitimate segmentation—such as student discounts with clear eligibility—from covert exploitation of vulnerability.

6. Performance Measurement and Control

A pricing dashboard should connect commercial outcomes with customer and operational consequences. Useful measures include:

Average order value is useful but incomplete. A higher basket can accompany lower contribution if customers use deep discounts or buy low-margin items. Similarly, revenue growth may conceal rising acquisition cost, returns or fulfilment expense.

A/B testing can estimate the effect of a price or promotion when random assignment is feasible and ethical. The test should define the unit of randomisation, treatment, control, duration, primary metric and guardrail metrics. Interference between customers, inventory constraints, seasonality and repeated exposure can weaken validity. A statistically detectable effect is not necessarily commercially material; managers should consider confidence, effect size and implementation cost.

7. Applied Case: MarketNest’s Seasonal Electronics Promotion

The following is an illustrative case.

Background

MarketNest, an online marketplace, sells wireless headphones from several merchants. The typical price is 80 currency units. The marketplace earns a 12% commission on the item price, while a payment and service cost of 4 per order is borne by MarketNest. A merchant proposes a 10% weekend coupon funded jointly by the merchant and marketplace. Management expects the coupon to increase orders but is concerned about margin and future reference prices.

Problem

Historical comparable weekends show 2,000 orders at 80. The forecast under the coupon is 2,500 orders at 72. The marketplace's contribution before advertising is commission revenue minus the 4-unit service cost. The coupon funding is 2 units per order for the marketplace, and a campaign costs 3,000 units.

Analysis

Baseline contribution per order is 0.12 × 80 − 4 = 5.60. Promoted contribution per order before coupon funding is 0.12 × 72 − 4 = 4.64. After the marketplace's 2-unit funding, promoted contribution is 2.64. Baseline contribution is 2,000 × 5.60 = 11,200. Promoted contribution before campaign cost is 2,500 × 2.64 = 6,600. After the campaign cost, contribution is 3,600. The promotion therefore reduces contribution by 7,600 relative to the baseline.

The result could change if the promotion generates valuable new customers, increases later purchases, improves merchant retention or prevents competitor switching. Those benefits must be measured rather than assumed. A holdout group, where operationally feasible, could estimate incremental orders and repeat behaviour. The firm should also check whether the forecast includes orders that would have occurred without the coupon.

Approach and Solution

MarketNest should not launch the offer in its original form. It could test a narrower offer, such as a minimum basket threshold, a smaller coupon, a merchant-funded discount or a targeted incentive for genuinely new customers. It should set a contribution floor, disclose the conditions clearly and monitor returns, complaints and repeat purchase. A control group would allow estimation of incremental contribution and customer lifetime effects.

Outcome and Learning Points

The initial analysis demonstrates that more orders and a higher conversion rate can coexist with lower contribution. Pricing decisions must use net economics, not gross sales. The case also illustrates the importance of funding responsibility, baseline estimation, customer lifetime value and transparent promotional design. A promotion is successful only when its incremental benefits exceed its incremental costs and risks.

8. Best-Practice Decision Framework

Before approving a price or promotion, managers should ask:

1. What objective is being pursued: profit, volume, acquisition, inventory clearance or positioning?
2. What is the relevant cost and contribution after all discounts and fulfilment effects?
3. Which customers are eligible, and why is that segmentation legitimate?
4. How will competitors and customers probably respond?
5. What is the expected elasticity and what assumptions support it?
6. Is there a control or baseline for measuring incrementality?
7. What operational constraints could make the offer unprofitable?
8. Are the price, conditions and reference claims clear and lawful?
9. What model risks, fairness concerns and human approvals apply?
10. What stop rule will end the offer if performance or compliance deteriorates?

Challenges and Limitations

Pricing analysis is limited by incomplete data, changing competitors, selection bias and uncertainty about causation. Historical prices may reflect earlier promotions, stock-outs or customer mix and therefore cannot be interpreted as pure demand evidence. Elasticity estimates can be unstable when price variation is small. Dynamic algorithms may optimise short-run contribution while damaging trust or long-run retention. Marketplace data may also create conflicts between platform objectives and seller interests.

Managers should use sensitivity analysis, scenario ranges and periodic review rather than presenting forecasts as precise facts. Human judgement remains important when markets experience shocks, when legal standards change or when customer welfare is difficult to quantify.

Chapter Summary

E-commerce pricing combines economic analysis, customer value, competitive intelligence, promotion design and operational control. Cost-based pricing provides a floor and supports financial discipline, but value and competition determine whether a price is viable in the market. Contribution, break-even analysis and elasticity help managers evaluate the consequences of price changes. Promotions should be judged by incremental contribution after discounts, cannibalisation, fulfilment costs and campaign expense. Markdowns address perishable inventory, while bundles can increase value and basket size when their economics are sound.

Dynamic pricing and revenue management use forecasts, inventory and capacity information to improve decisions over time. Algorithms can assist, but they require reliable data, monitoring, explainability and human accountability. Legal and ethical limits include truthful reference prices, transparent total costs, non-deceptive scarcity, privacy protection, avoidance of unlawful discrimination and competition-law compliance. The central managerial principle is to pursue profitable, measurable and trustworthy value exchange rather than sales volume alone.

Key Terms

Price: The amount charged for an offering.

Net price: Amount retained after discounts, rebates and credits.

Variable cost: Cost that changes with units sold.

Fixed cost: Short-run cost not directly changed by unit volume.

Contribution margin: Price less variable cost per unit.

Cost-plus pricing: Pricing based on cost plus a markup.

Value-based pricing: Pricing based on customer-perceived benefits and alternatives.

Competition-based pricing: Pricing informed by rival offers.

Price elasticity: Percentage change in quantity divided by percentage change in price.

Reference price: Expected or comparison price used by a customer.

Markdown: Planned price reduction, often for ageing or seasonal inventory.

Coupon: Conditional incentive reducing price or providing a benefit.

Bundling: Selling two or more products as a combined offer.

Dynamic pricing: Changing prices in response to market or operational conditions.

Revenue management: Allocating constrained capacity to maximise expected revenue or contribution.

Cannibalisation: A promotion shifting customers from a full-price or alternative product.

Incrementality: Outcome caused by an intervention beyond the baseline.

Price discrimination: Charging different prices for equivalent offers under specified conditions.

Algorithmic pricing: Use of automated rules or models to set or recommend prices.

Price floor: Lowest permitted price under a policy or economic constraint.

Review Questions

Short Answer Questions

1. Distinguish markup percentage from margin percentage.
2. Why can revenue increase while profit decreases?
3. What is the purpose of a control group in promotion evaluation?
4. Distinguish dynamic pricing from individualised pricing.
5. Identify two legal or ethical risks associated with algorithmic pricing.

Long Answer Questions

1. Compare cost-, competition- and value-based pricing, including the situations in which each is most useful.
2. Explain how elasticity, contribution and break-even analysis should be combined in a pricing decision.
3. Discuss the benefits and risks of coupons, markdowns and bundles in an e-commerce business.
4. Describe a revenue-management process for a marketplace with limited fulfilment capacity.
5. Evaluate the governance requirements for using AI in price recommendations.

Analytical / Application Questions

1. A retailer sells at 45 with variable cost 27 and fixed monthly costs of 18,000. Calculate contribution per unit and break-even volume. If a promotion reduces price to 42, calculate the new break-even volume and explain the required managerial decision.
2. A website raises price from 30 to 33 while quantity falls from 4,000 to 3,400. Use the midpoint method to estimate price elasticity and determine whether revenue rises or falls.
3. Design a promotional-calendar experiment for a marketplace. Specify the objective, treatment, control, primary metric, guardrail metrics, legal checks and stop rule.

Multiple Choice Questions with Answers

Question 1

A product sells for 50 currency units and has variable cost of 32. What is its unit contribution?

A. 18 currency units

B. 22 currency units

C. 32 currency units

D. 82 currency units

Correct Answer: A. 18 currency units

Question 2

Which approach begins with customer-perceived benefits and alternatives?

A. Cost-plus pricing

B. Value-based pricing

C. Historical pricing

D. Inventory pricing

Correct Answer: B. Value-based pricing

Question 3

If price rises by 10% and quantity demanded falls by 5%, demand is approximately:

A. Perfectly elastic

B. Elastic

C. Inelastic

D. Unit elastic

Correct Answer: C. Inelastic

Question 4

A retailer has fixed costs of 12,000, price of 40 and variable cost of 28. What is break-even volume?

A. 300 units

B. 428 units

C. 1,000 units

D. 1,500 units

Correct Answer: C. 1,000 units

Question 5

Which measure best evaluates whether a promotion created additional performance beyond normal sales?

A. List-price percentage

B. Redemption volume alone

C. Incrementality against a baseline

D. Number of promotional emails sent

Correct Answer: C. Incrementality against a baseline

Question 6

What is a principal purpose of markdowns?

A. To increase fixed costs

B. To clear ageing or seasonal inventory

C. To eliminate all competitor activity

D. To guarantee higher unit margins

Correct Answer: B. To clear ageing or seasonal inventory

Question 7

Which statement correctly distinguishes dynamic pricing from individualised pricing?

A. Dynamic pricing always uses personal data

B. Individualised pricing changes one public price for everyone

C. Dynamic pricing responds to conditions, while individualised pricing may vary by customer

D. They are legally identical in every jurisdiction

Correct Answer: C. Dynamic pricing responds to conditions, while individualised pricing may vary by customer

Question 8

Which practice creates a consumer-protection risk?

A. Clearly stating total delivered cost

B. Using a genuine student eligibility rule

C. Displaying a false original price as a reference

D. Testing a price with a documented control group

Correct Answer: C. Displaying a false original price as a reference

Question 9

A bundle is economically attractive only when:

A. Its displayed price exceeds the sum of all list prices

B. Its incremental contribution and strategic benefits justify cannibalisation and costs

C. It contains at least three products

D. Customers cannot see component prices

Correct Answer: B. Its incremental contribution and strategic benefits justify cannibalisation and costs

Question 10

Which governance practice is most appropriate for an AI pricing system?

A. Allowing the model to change objectives without approval

B. Removing audit logs to protect the algorithm

C. Monitoring outcomes, documenting rules and providing human override

D. Using sensitive attributes whenever prediction improves

Correct Answer: C. Monitoring outcomes, documenting rules and providing human override