Price Elasticity of Demand Formula: A Practical Guide

Price elasticity of demand is the single most useful number in pricing, and one of the most misunderstood. It tells you, in one figure, how much your sales volume will move when you change a price. Get it right and you know whether a price increase will quietly grow profit or torch your volume. The formula itself takes thirty seconds to learn. Using it well is where the money actually is.

This guide covers both. The formula and how to compute it cleanly, then the part most explainers skip: how to turn an elasticity number into an actual price, and what breaks when you try to do it with real data.

The formula

Price elasticity of demand, usually written PED or just E, is a ratio of two percentage changes:

PED = % change in quantity demanded / % change in price

That's it. If a 10% price increase causes a 5% drop in units sold, elasticity is −5% / 10% = −0.5. The result is almost always negative, because demand and price move in opposite directions: raise the price, sell fewer. By convention most people drop the minus sign and talk about the absolute value, so this product has an elasticity of 0.5.

The number answers a precise question: for a 1% change in price, what percentage change in volume should I expect? An elasticity of 0.5 means a 1% price rise loses you half a percent of volume. An elasticity of 2 means it costs you two percent.

A worked example, and why the simple version misleads

Take a lunch special. You raise the price from $5 to $6, and weekly orders fall from 2,000 to 1,400.

The naive calculation uses the starting values as the base:

% change in quantity = (1400 − 2000) / 2000 = −30%
% change in price     = (6 − 5) / 5         = +20%
PED = −30% / 20% = −1.5

So elasticity is 1.5, elastic. But there's a trap. Run the exact same change in reverse, dropping the price from $6 back to $5, and you get a different elasticity, because now you're dividing by different base numbers. The same price change shouldn't produce two different answers. That inconsistency is why economists use the midpoint method.

The midpoint method

The midpoint method fixes the base problem by dividing each change by the average of the start and end values rather than the starting value alone:

PED = [(Q₁ − Q₀) / ((Q₁ + Q₀) / 2)] / [(P₁ − P₀) / ((P₁ + P₀) / 2)]

Run our lunch special through it:

% change in quantity = (1400 − 2000) / ((1400 + 2000) / 2) = −600 / 1700 = −35.3%
% change in price     = (6 − 5) / ((6 + 5) / 2)            =    1 / 5.5  = +18.2%
PED = −35.3% / 18.2% ≈ −1.94

Now the elasticity is about 1.94 whether the price went up or down. Same number, both directions. That consistency is the whole reason the midpoint method exists, and it's the version worth defaulting to.

Key term — Midpoint method

A way of computing elasticity that uses the average of the old and new values as the base for each percentage change. It gives one consistent elasticity for a price move regardless of direction, unlike the simple method, which gives different answers depending on where you start.

Reading the number

The threshold is 1. Everything hinges on which side of it you land.

The revenue logic follows directly. When demand is elastic, customers are price-sensitive, so a price hike loses more volume than it gains in margin per unit, and total revenue drops. When demand is inelastic, buyers stick around despite the increase, so raising the price grows revenue. Gasoline is the textbook inelastic good: prices climb and people keep filling up because there's no easy substitute on the morning commute.

From elasticity to an actual price

This is the step that turns a classroom formula into money. Revenue isn't the goal, profit is, and elasticity ties straight to the profit-maximizing price through a clean relationship. For a product with marginal cost MC and elasticity E (greater than 1), the optimal price is:

price = MC × E / (E − 1)

Work an example. A product costs you $40 to make and has an elasticity of 2:

price = 40 × 2 / (2 − 1) = 40 × 2 = $80

Now suppose it's more elastic, say E = 3, because shoppers have more alternatives:

price = 40 × 3 / (3 − 1) = 40 × 1.5 = $60

Same cost, very different prices. The more elastic the product, the lower its profit-optimal markup, because every extra dollar drives away too much volume to be worth it. This is the empirical engine behind value-based pricing, and we contrast the two head-on in cost-plus pricing vs value-based pricing. It's also why an AI price optimization engine leans on elasticity rather than a flat markup: the formula already tells you a fixed markup is the wrong answer for most products.

Where the formula meets reality

Everything above assumes you have two clean data points and nothing else changed between them. Real catalogs are messier, and pretending otherwise is how elasticity gets a bad name.

In practice, you rarely run a tidy A/B price test per product. You estimate elasticity from your own sales history, and that history is full of confounders. A price changed, but so did the season, a competitor ran a promotion, you were out of stock for a week, or a holiday landed mid-period. Attribute all the volume change to price alone and your elasticity is garbage. Real estimation has to separate the price effect from everything else moving at the same time.

Elasticity also isn't one number per product. It differs by customer segment, by channel, and by how far you push the price. The elasticity between $10 and $11 may be nothing like the elasticity between $20 and $22. That's why serious pricing treats elasticity as a curve estimated per product and segment, not a single figure, and why it pairs naturally with customer-segment pricing.

How a pricing engine uses it

A real engine doesn't ask you to compute elasticity by hand. It learns price sensitivity per SKU from your sales history, then uses that to inform the target price rather than to set it outright. Elasticity is one input weighed alongside cost, the competitor median, and the product's role, and the result is clamped by the margin floors and change limits you set so a confident elasticity estimate can't push a price somewhere reckless.

That combination is the point. Elasticity tells you which products can carry a higher price and which punish every cent, cost sets the floor, competitor data keeps you on-market, and guardrails keep the whole thing safe. You can read how those pieces fit together in how AI price optimization works, and how each pricing strategy uses elasticity in the strategies documentation.

Common questions

Ready to turn elasticity into prices you can defend? See how Elastly handles price optimization, read how AI price optimization works, or explore Elastly's pricing.