Demand Analysis and Price Sensitivity

Understanding Your Demand Curve

Every pricing decision involves an assumption—explicit or implicit—about how quantity demanded will respond to price. Get this assumption wrong, and even carefully calculated prices will miss their targets. Demand analysis makes these assumptions explicit and tests them against evidence.

The fundamental concept is the demand curve: the relationship between price and quantity demanded, holding other factors constant. Generally, higher prices mean lower quantities demanded—the law of demand. But the shape and steepness of this curve varies enormously across products, customers, and contexts. Some products can sustain significant price increases with minimal volume loss; others see demand collapse at the slightest price increase.

Your task as a pricing strategist is to understand where your product falls on this spectrum—and more precisely, to estimate how much volume will change for any given price change you're considering.

Price Elasticity: The Key Metric

Price elasticity of demand measures the percentage change in quantity demanded for a given percentage change in price. The formula is simple:

Elasticity = % Change in Quantity ÷ % Change in Price

If a 10% price increase causes a 15% volume decrease, elasticity is -1.5. If the same increase causes only a 5% decrease, elasticity is -0.5. The negative sign reflects the inverse relationship between price and demand—higher prices mean lower quantities.

Elasticity Value

Classification

Revenue Impact of Price Increase

Strategic Implication

< -1.0

Elastic

Revenue decreases

Be cautious with increases; consider decreases

= -1.0

Unit Elastic

Revenue unchanged

Price changes are revenue-neutral

-1.0 to 0

Inelastic

Revenue increases

Price increases boost revenue

Near 0

Perfectly Inelastic

Revenue increases proportionally

Maximize price subject to other constraints

The elasticity threshold of -1.0 is crucial. When demand is elastic (elasticity below -1.0), price increases reduce total revenue because the volume lost exceeds the per-unit gain. When demand is inelastic (elasticity between -1.0 and 0), price increases boost revenue because volume loss is proportionally smaller than the price gain.

Note that revenue isn't the same as profit. Even with elastic demand, a price increase might increase profit if the margin gain on remaining volume exceeds the contribution lost from departed customers. But elasticity gives you a starting point for analysis.

Methods for Measuring Elasticity

Several approaches exist for estimating price elasticity, each with strengths and limitations. Sophisticated pricing organizations use multiple methods and triangulate across them.

Historical Analysis

Examine past price changes and their volume effects. This uses real data—not hypotheticals—but suffers from confounding factors. Other things changed simultaneously with price: marketing spend, competitive actions, economic conditions, seasonality. Isolating the price effect requires statistical controls.

Regression analysis can help disentangle price effects from other factors if you have enough data points. You might model volume as a function of price, marketing spend, competitor prices, seasonal factors, and economic indicators. The coefficient on price, properly specified, estimates elasticity.

Even without sophisticated statistics, historical analysis provides useful intelligence. When we raised prices 8% in Q3 2022, what happened to volume? Did customers push back? Did competitors respond? Historical patterns, even imprecisely measured, inform judgment.

Customer Research: Surveys and Stated Preference

Ask customers directly how they would respond to different prices. This is easy to conduct but often unreliable—what people say and what they do are different. Customers typically overstate price sensitivity in surveys because stating high willingness to pay feels like giving up negotiating leverage.

Survey methods work better for directional insight than precise estimates. If 80% of surveyed customers say they would 'definitely' or 'probably' stop buying at a 20% price increase, you know you have elastic demand—even if the actual defection rate would be lower than stated.

Conjoint Analysis

Conjoint analysis presents customers with product configurations varying in features and price, then analyzes their choices to infer willingness to pay. Because it mirrors real purchase decisions (choosing among alternatives rather than stating preferences directly), it tends to be more reliable than simple surveys.

In a conjoint study, customers might see several product options with different combinations of features and prices. Their choices reveal the implicit trade-offs they make—how much they value each feature, and therefore how much they'd pay for it. Properly designed conjoint studies can estimate price sensitivity with reasonable precision.

Conjoint requires careful design and significant sample sizes. It works best for products with distinct features that can be varied independently. It's less useful for commodity products or services where differentiation is subtle.

Price Experiments

Test different prices in controlled settings and measure actual behavior. This provides the most reliable estimates because it uses real purchase decisions, not stated preferences.

Experimental approaches include A/B testing (showing different prices to randomly selected customer segments), geographic testing (testing prices in different regions), temporal testing (testing prices in different time periods), and controlled promotions (measuring volume response to promotional price reductions).

Experiments require careful design to ensure valid comparisons. You need sufficient sample sizes, proper randomization, and controls for confounding factors. There are also risks: customers who discover price differences may feel treated unfairly, and competitors may respond to test prices.

Case Study: Price Experimentation at Scale: E-Commerce A/B Testing A major e-commerce retailer wanted to optimize pricing on its top 1,000 SKUs. They implemented an automated A/B testing program that randomly showed different prices to different visitors, measured conversion rates at each price point, and calculated elasticity estimates for each product. The program ran continuously, refining estimates as more data accumulated. Results varied dramatically: some products had elasticity near zero (customers barely noticed price changes), while others had elasticity below -3 (small price increases caused major volume drops). Armed with product-level elasticity estimates, the retailer optimized prices individually—raising prices on inelastic products, holding or reducing prices on elastic ones. The program generated an estimated $23 million in additional annual profit.

Expert Judgment

Gather estimates from people who know your customers: sales teams, product managers, customer service staff, industry experts. This is quick, inexpensive, and captures institutional knowledge. But it's subject to biases—sales teams typically overestimate price sensitivity because they hear complaints, while product teams may underestimate because they're enthusiastic about value delivered.

Structured approaches reduce bias. The Delphi method gathers independent estimates from multiple experts, shares aggregated results, and iterates until estimates converge. Prediction markets let experts bet on outcomes, creating incentives for accuracy. Even informal calibration—asking experts to estimate confidence intervals, not just point estimates—improves quality.

Factors Affecting Price Sensitivity

Understanding what drives elasticity helps you predict sensitivity even without precise measurement. Consider these factors:

• Availability of substitutes: More available, attractive substitutes increase price sensitivity. Unique products face less elastic demand.
• Necessity vs. discretionary: Essential products have more inelastic demand. Discretionary purchases are more price-sensitive.
• Budget share: Products that represent a larger share of customer budget face more scrutiny and higher sensitivity.
• Comparison difficulty: When quality comparison is difficult, customers rely less on price and sensitivity decreases.
• Switching costs: Higher switching costs reduce sensitivity because customers are locked in.
• End-benefit value: If your product enables valuable outcomes (productivity tools, efficiency equipment), sensitivity decreases.
• Quality signal: When price signals quality, raising prices may actually increase demand in some segments.
• Shared cost: When someone else pays (insurance, employers), customers are less price-sensitive.
• Urgency: Urgent needs reduce price sensitivity—customers pay more when they need it now.
• Inventory effects: If customers can stockpile (buying ahead of price increases), apparent sensitivity spikes around changes.

Key Takeaways

• Every pricing decision assumes something about demand response—make these assumptions explicit
• Price elasticity quantifies demand sensitivity and has direct revenue implications
• Multiple methods exist for measuring elasticity—triangulate across approaches for reliability
• Understanding drivers of sensitivity helps predict responses even without precise measurement