Price Elasticity of Demand: Formula, CPG Benchmarks & Worked Examples

Price Elasticity of Demand (PED) measures how much the volume you sell changes, in percent, when your price changes by one percent. It answers the most fundamental question in pricing: "If I change my price, what happens to my volume?"

The sign convention

Elasticity is almost always negative. When price goes up, demand goes down. By convention in FMCG, practitioners often express it as a positive number (e.g., "elasticity of 2.0"), but the direction is understood: a 1 percent price increase leads to a roughly 2 percent volume decline.

An elasticity of exactly -1.0 is the break‑even point for revenue: the volume you lose exactly offsets the higher price, so the money coming in does not move. Anything shallower than that, closer to zero, and a price rise adds revenue. Anything steeper, further from zero, and a price rise costs you revenue. The Math tab works through the three ways of arriving at the number itself.

There are three ways to arrive at an elasticity. All three answer the same question. They differ in what data you need to have in front of you, and in how far you can trust the answer once the price move gets large.

If you have a before and an after: the two‑point method

You know last quarter's price and volume, and you know this quarter's. Divide the percentage change in volume by the percentage change in price. This is the version almost everyone computes in a spreadsheet, and it holds up well for moves of less than about 10 percent.

It has one weakness worth understanding, because it catches people out in review meetings. The formula divides by where you started, so the answer depends on which end you start from. Take a pack from $4.00 to $4.40 and the price rose 10.0 percent. Walk the same pack back from $4.40 to $4.00 and the price fell 9.1 percent. Two identical price points, two different percentage changes, and the elasticity you compute inherits the difference. The bigger the move, the wider those two answers sit apart.

If you want one answer whichever way the price moved: the midpoint method

The same two observations, divided by the average of the two prices and the average of the two volumes rather than by the starting ones. That removes the direction problem above, so a rise and a cut between the same two prices now report the same number. Nearly every published price test reports its result this way, which is also why you should ask which method was used before you compare two results side by side.

If somebody has fitted a demand curve: read it off the curve

Once an analyst has fitted an equation to your past sales, so that the equation describes how your volume has moved with price, you no longer need two measured points. You can ask the curve what the elasticity is at any price you like, including prices you have never charged. Every pricing model and every scenario tool works this way, including the Sandbox further down this lesson.

Which one you reach for. Use the two‑point method to sanity‑check a number quickly. Use the midpoint method when you are reporting a result somebody else will compare against theirs. Use the fitted curve when you need to test a price you have never charged.

The three regimes

• Inelastic (|E| < 1): volume drops by a smaller percentage than the price rises, so a price rise brings in more money.
• Unit elastic (|E| = 1): the turning point for revenue. A small price rise brings in the same money as before.
• Elastic (|E| > 1): volume drops by a bigger percentage than the price rises, so a price rise brings in less money.

There is no normal elasticity, and knowing that is the first thing worth learning

Nobody can tell you what elasticity a consumer‑goods brand has, and anyone who hands you one number for the whole industry is averaging brands, categories and markets that are not alike. It differs hugely by market, by category, by shopper, by product and by occasion. General averages exist, and the real range around them is much wider and entirely context‑specific. The same brand can read very differently in two markets in the same year.

So what follows is not a target and not a benchmark. It is the published work, with what each figure actually measured, so you can see the spread for yourself and judge where your own brand is likely to sit. You will meet three numbers described as typical, they look like they contradict each other, and they do not, because each is measuring something different.

The published average is -2.62. That is the mean of 1,851 elasticity estimates gathered from 81 academic studies, and it is one of the most quoted figures in pricing research. It is the number to cite when somebody asks what the literature says.

It reports a bigger volume response than your own brand is likely to show, and the study itself explains why. The 1,851 estimates do not all measure the same thing. Some measure how much volume moves, some how much market share moves, and some how likely a shopper is to pick the brand at all. None of those three answers the question you are actually asking, which is what your own volume does when the everyday shelf price changes. The pool also spans every category, market and method anyone chose to publish, while your number is one brand, in one category, in one country. And the studies that correct for the fact that companies choose their own prices, rather than being handed prices at random, report a bigger volume response than the studies that do not.

Many teams plan at around -1.7 to -1.8 when they have measured nothing. That is a habit rather than a finding, and it is the middle case on the chart in the Theory tab. It exists so a business case is not built on a blank, and it stops being useful the moment a real measurement arrives. Do not carry it into a meeting as what your brand does.

And the average can hide a split inside your own category. Retail measurement across several large markets now shows categories splitting at both ends, with premium and value both gaining ground while the mainstream middle, the familiar names priced between the two, loses it. Where that is happening, one brand is facing two groups of shoppers who answer a price change very differently, and a single brand‑level number is the average of two behaviors that are not alike. It describes neither of them, and it is least reliable at exactly the moment you need it most, which is a move large enough to push one group and not the other. PPA Lesson 1 has the evidence on how the mix between price tiers has been shifting, along with the case that ran the other way, where the middle grew.

The only number that should decide a price move is your own. Everything above is there to help you judge whether your own number looks believable. Your own measured elasticity might come back at -0.9, or at -3.4, or further from zero still in a crowded category where shoppers have somewhere cheap to go. Any of those is a reason to work out why your brand sits where it does, never a reason to distrust your own data.

And the published ranges are wide on purpose. Mainstream FMCG brands, the familiar names in the middle of the shelf, mostly land between -1.5 and -2.5. Premium tiers mostly land between -0.8 and -1.5, and value tiers and private label between -2.5 and -3.5. Those bands overlap, they are drawn from studies with different methods, and real brands sit outside them often enough that a number outside the band is a question rather than an error.

The category team is reading scanner data for CrunchField Original 300g through one major grocery account. Original is a different pack from the Heritage 300g the Sandbox prices, and its $4.29 here is a shelf price, not Heritage's $4.29 list price. In that account the pack sits on the shelf at $4.29 and sells 50,000 units per month, and the account‑level econometric read puts elasticity at -1.8. (Pulling several sources together, the brand's central estimate is -1.7. Readings taken inside a single account come out a little stronger where private label sits right next to you on the shelf. The Elasticity Estimation Methods card covers why the two differ.)

Scenario: a 5 percent list price rise, passed on to a $4.50 shelf

CrunchField raises its list price 5 percent, and the retailer, which decides the shelf price, passes the rise on in full.

The volume math

• Expected volume change: -1.8 x 5% = -9.0%
• New volume: 50,000 x (1 - 0.09) = 45,500 units

The sales math, at the shelf price

• Shoppers' spend before: $4.29 x 50,000 = $214,500
• Shoppers' spend after: $4.50 x 45,500 = $204,750

Because |E| > 1.0, the volume loss more than offsets the higher price. This price increase destroys revenue. If the retailer keeps its percentage margin, CrunchField's own Net Sales fall by the same 4.5 percent. The straightforward read on the elasticity coefficient says the move fails.

In FMCG, elasticity is the pricing input everything else builds on. Four working rules separate the people who use it well from the people who quote one number and walk away.

Calibrate to your own segment

The ranges in the previous tab describe whole tiers, and your brand sits inside one of them rather than at the middle of all of them. A premium brand that plans on the mainstream average expects to lose more volume than it really will, so it walks away from a price rise it could have taken. A value brand doing the same expects to lose less than it really will, so it approves a rise that costs more volume than the business case behind it allowed for.

Elasticity is not constant

It changes with the competitive environment, the economic cycle, the season, and the magnitude of the price change. A 3 percent increase might face an elasticity of -1.8, while a 15 percent increase in the same category could show -3.0 because you have pushed beyond psychological thresholds.

Triangulate, never trust one number

Three sources should agree before you commit a price decision:

1. Econometric model, a statistical model fitted to your own history of sales and prices
2. Price test run over a set window in two groups of stores picked to match each other
3. Category manager intuition informed by competitive context

When all three disagree, dig deeper. When two agree and one is an outlier, trust the two and investigate the outlier.

Base elasticity and promotional elasticity are two different numbers

Every brand has two separate elasticities, and treating them as one number is one of the more expensive mistakes in pricing work. Base elasticity is how volume responds to a change in the everyday shelf price. Promotional elasticity is how it responds to a temporary cut with a display or a feature behind it.

They routinely point in opposite directions. A brand can be inelastic on base price, meaning a 5 percent shelf rise barely moves its volume, and strongly elastic on promotion, where a 20 percent cut on a display at the end of the aisle triples that volume for a week. The reverse happens too, on brands whose shoppers are loyal to a deal rather than to the label.

Two things drive them apart. A shelf price change is seen by everyone who walks the aisle, week after week, and it resets what shoppers believe the pack is worth. A promotion is seen only by whoever shops during it, and part of what it sells is pulled forward from people who would have bought anyway. So promotional volume is partly borrowed against next month, while volume lost to a shelf price rise is gone.

What the two numbers decide. When the business needs another point of margin, the choice is between taking the shelf price up and spending less on promotion. A brand with mild base elasticity and strong promotional elasticity should take the shelf price up, because it loses little volume on the everyday price and keeps running the promotions its shoppers respond to strongly. A brand the other way round should cut promotional depth, because a shelf rise would cost it more.