Price Elasticity of Demand: How to Estimate It for Your Products

Price elasticity of demand measures how much the quantity customers buy changes when you change the price, and it is the number that tells you whether a price rise will make you money or cost you customers. The calculation is simple, a ratio of two percentage changes, but estimating it from real data is hard because many other things change at the same time. This guide gives the formula, works through inelastic and elastic examples, shows that a price move can raise revenue and lower profit, or the reverse, covers a regression estimate with checked numbers, and lists the pitfalls. It belongs to unit economics and budgeting and forecasting, builds on value-based pricing and pricing strategy, and leads to how discounts erode margin and how to raise prices.

The formula

Price elasticity of demand = % change in quantity demanded / % change in price

Because demand usually falls when price rises, the result is normally negative. Many people quote the absolute value.

Example 1: an inelastic product

A product priced at $50 sells 1,000 units a month. The price rises to $55 (+10%) and sales fall to 920 units (-8%).

• Elasticity = -8% / +10% = -0.8
• Revenue before: 50 × 1,000 = $50,000. After: 55 × 920 = $50,600
• With a variable cost of $30, contribution before: 20 × 1,000 = $20,000. After: 25 × 920 = $23,000, up 15%

Revenue rose slightly, and contribution rose by 15 percent. The product is inelastic in this range, so the price rise was profitable.

The midpoint method

A percentage change depends on the starting point: a rise from 50 to 55 is +10 percent, but a fall from 55 to 50 is -9.1 percent. The midpoint (arc) method divides by the average of the old and new values and gives the same answer in both directions.

• % change in price = 5 / 52.5 = 9.52%
• % change in quantity = -80 / 960 = -8.33%
• Arc elasticity = -8.33 / 9.52 = -0.875

Use the midpoint method for larger changes and the simple method for small ones.

Example 2: an elastic product, where revenue and profit disagree

Another product is priced at $40 and sells 1,000 units a month. The price is cut by 10 percent to $36, and sales rise by 30 percent to 1,300 units.

• Elasticity = +30% / -10% = -3.0 (elastic)
• Revenue before: 40 × 1,000 = $40,000. After: 36 × 1,300 = $46,800, up 17%
• With a variable cost of $25: contribution before 15 × 1,000 = $15,000. After: 11 × 1,300 = $14,300, down 4.7%

Revenue rose 17 percent and contribution fell. The price cut lifted sales, but the margin per unit fell more than the volume rose. Decisions about price should be judged on contribution, not revenue. At a lower margin, the volume has to rise by a larger proportion to compensate, as shown in pricing strategy.

What elasticity implies for the best price

If elasticity is constant at a value e (negative), the profit-maximizing price satisfies a simple rule: the margin on price should equal 1 divided by the absolute value of elasticity.

Optimal price = Variable cost / (1 - 1/|e|)    (only if |e| is greater than 1)

With e = -3 and a variable cost of $25: optimal price = 25 / (1 - 1/3) = 25 / 0.6667 = $37.50. At that price, the margin is 12.50 / 37.50 = 33.3 percent, which equals 1/3. The current price of $40 is above this level, and the cut to $36 went beyond it. If elasticity is less than 1 in absolute value (inelastic), the formula says to keep raising the price until demand becomes elastic. In practice, elasticity changes with the price level, and the rule works only within the range where you have data.

Estimating elasticity from data

Methods, in order of reliability:

1. Controlled experiments: price tests with a control group. See how to run a price test.
2. Natural experiments: a price change in one region or channel but not another.
3. Regression on historical data: controlling for seasonality, promotions, competitor prices and trend.
4. Survey methods: stated willingness to pay, less reliable.

A regression example

A log-log regression gives the elasticity directly: regress the natural log of quantity on the natural log of price, and the slope is the elasticity. In Excel, put prices in column A and quantities in column B, create ln columns, and use =SLOPE(lnQ range, lnP range).

Suppose six weeks of data with no other changes (illustrative):

The slope of the regression of ln(units) on ln(price) for these data is about -1.34. The simple two-point calculation from $20 to $25 gives a similar answer: price +25 percent, units -26 percent, elasticity about -1.04 on the simple method and higher with midpoints. The product is somewhat elastic: revenue falls as price rises, from $20,000 to $18,500. Yet contribution rises from $6,000 to $8,140, because the unit margin grows from $6 to $11 faster than volume falls. The margin on price at $20 is 30 percent, much lower than the 1 / 1.34 = 75 percent that the rule above would imply as optimal, so there is room to keep raising price in the tested range, provided the estimate holds beyond it. Six weeks of data and no controls make this a weak estimate, and any decision would need a test.

Pitfalls in estimation

• Confounding. Prices usually change together with promotions, seasons or competitor moves. Control for them.
• Short-run versus long-run. Customers may react slowly, so long-run elasticity is usually larger in absolute value.
• Reference prices. Customers compare with past prices, so a price rise after discounts feels larger than the number suggests.
• Cross-elasticity. A price change on one product shifts sales of others. Include substitutes and complements.
• Range. Elasticity estimated between $20 and $25 may not hold at $35.
• Segments. Aggregate elasticity hides large differences between customers and channels.
• Supply constraints. If you stock out, sales do not reflect demand.

Using elasticity in decisions

Elasticity and supply

Elasticity estimates depend on stock availability. When products are out of stock, observed sales understate demand and the estimate is biased toward inelastic. When a price cut triggers more demand than supply can meet, the promotion fails. Use planning data to correct for stockouts and to make sure inventory is ready for the volume a price change is expected to produce. See demand planning and inventory optimization.

Common mistakes

• Reading revenue instead of contribution.
• Estimating from a single price change.
• Ignoring other changes at the same time.
• Assuming a constant elasticity over a wide range.
• Applying a company-wide figure to every product.
• Forgetting competitors' reactions.

FAQs

What is price elasticity of demand?

The percentage change in quantity demanded divided by the percentage change in price. It shows how sensitive buyers are to price.

How do you calculate price elasticity?

Divide the percentage change in quantity by the percentage change in price. For large changes, use the midpoint method.

What does an elasticity of -0.8 mean?

A 10 percent price rise reduces quantity by about 8 percent. Demand is inelastic, and revenue will rise with price.

Can a price increase raise profit if demand is elastic?

Yes, if the contribution margin per unit rises enough to offset the volume lost. Judge price changes on contribution, not revenue.

How do I estimate elasticity for my products?

Use price tests with controls where possible. Otherwise regress log quantity on log price using historical data, controlling for promotions, seasonality and competitors.