Price Elasticity
Date: 2026-08-16
How much demand moves when price does. It’s the number that decides whether a price rise makes or loses money — and because margin amplifies it, the break-even volume loss is usually far larger than people expect.
What it is
Price elasticity of demand is the percentage change in quantity sold for a percentage change in price.
price +10%, quantity −15% E = −1.5 elastic — revenue falls
price +10%, quantity −5% E = −0.5 inelastic — revenue rises
price +10%, quantity −10% E = −1.0 unit elastic — revenue flat
Elasticity is negative by convention; people usually quote the absolute value.
Margin changes where break-even sits
The part that matters commercially, and it’s counterintuitive in a useful direction: a price rise can lose volume and still make more money, because you keep more of each remaining sale.
On the running model — £50 price, £15 contribution:
raise price 10% → £55, contribution £20.00
(cost side unchanged)
break-even volume loss = 1 − (15.00 ÷ 20.00) = 25%
You can lose a quarter of your unit sales and be no worse off. Very few price rises lose 25% of volume, which is why price increases are so often under-used.
The same arithmetic in reverse is why discounting is so punishing — a 10% cut needs a 49% volume gain to break even. Price rises and price cuts are wildly asymmetric, and the asymmetry is entirely a function of margin — Discount Impact on Margin.
Estimating it without a formal test
Price testing is legally and practically awkward in retail, so most elasticity estimates come from observation:
- Historical price changes. Compare volume before and after, adjusting for seasonality. Confounded, but directional
- Promotional periods. A discount is a price change; the volume response is an elasticity estimate — with the caveat that a temporary discount produces a much larger response than a permanent price cut, because it triggers stockpiling and urgency
- Cross-sectional comparison of similar products at different price points. Confounded by everything that makes them different
- Geographic or channel variation, where it exists naturally
In plain terms: most elasticity numbers in retail are informed guesses. Treat them as ranges, and use break-even volume loss — which requires no elasticity estimate at all — as the primary decision tool.
What drives it
| More elastic | Less elastic |
|---|---|
| Easily comparable products | Distinctive or exclusive |
| Many substitutes | Few alternatives |
| Discretionary purchase | Necessity or habitual |
| Price-visible categories | Bundled or complex pricing |
| Price-sensitive acquisition channels | Brand and direct traffic |
That last row matters for growth work: elasticity varies by acquisition channel. Customers arriving from a deal site are far more price-sensitive than those arriving via branded search, which is an argument against a single blended elasticity — Channel Mix.
Where it’s used
- Price rises. Compute break-even volume loss first; it’s usually more permissive than intuition
- Discount decisions — the mirror image, and far less permissive — Discount Impact on Margin
- Shipping Thresholds, which are a price change to part of the basket
- Range architecture. Elastic lines drive traffic; inelastic lines carry margin — Merchandising
Cautions
- Short-run and long-run elasticity differ. A price rise may hold volume this month and erode share over a year as customers reassess
- Competitor response. Elasticity assumes everyone else stands still. They may not
- Reference price effects. Customers remember what you charged before, so a rise from a long-standing price feels worse than the same price newly set — Price Anchoring
- UK pricing-claim rules constrain how you present changes, particularly around was/now framing [CHECK: current CMA guidance on reference pricing before advertising any comparison]