Price Elasticity in Plain English
Price elasticity. Two words that make most business owners glaze over, because they sound like the start of a lecture with a supply and demand curve on a whiteboard and a lecturer who has never sold anything in his life. So let me strip all of that away, because underneath the jargon price elasticity is one plain question, and it is a question you can answer about your own shop this week. Here is the whole thing. If I move my price by ten percent, how much does my volume move? That is it. Everything else is decoration. And the reason it matters is that the answer tells you something most owners are desperate to know and too scared to test: are you leaving money on the table, or are you propping up your sales with advertising you should not need?
The phrase that sounds like an economics lecture
Price elasticity. Two words that make most business owners glaze over, because they sound like the start of a lecture with a supply and demand curve on a whiteboard and a lecturer who has never sold anything in his life. So let me strip all of that away, because underneath the jargon price elasticity is one plain question, and it is a question you can answer about your own shop this week.
Here is the whole thing. If I move my price by ten percent, how much does my volume move? That is it. That is price elasticity. Everything else is decoration. And the reason it matters is that the answer to that one question tells you something most owners are desperate to know and too scared to test: are you leaving money on the table, or are you propping up your sales with advertising you should not need?
I want to walk you through it with numbers you can copy into a spreadsheet, in plain words, no lecture. By the end you will know whether you are underpriced, you will have a worked mini case to copy, a list of the mistakes that quietly ruin a price test, and a simple ninety day plan to find the real answer for your own business rather than guessing at it.
One question, written down properly
Let us put that plain question into a formula, because the formula is genuinely friendlier than the sentence once you see it.
That funny symbol on the left, epsilon, is just the name we give the answer. On the right, the top is the percentage change in quantity sold, and the bottom is the percentage change in price. Percentage change in volume, divided by percentage change in price. That is the entire machine.
So work one example. You raise your price by ten percent. You watch, and your volume falls by six percent. Drop those into the formula: six on top, ten on the bottom, which is 0.6. Because volume went down while price went up, the sign is negative, so by convention we say the elasticity is about minus 0.6, but the number you actually care about is its size, 0.6. Hold on to that number, because everything hinges on whether it is bigger or smaller than one.
If the size is less than one, like our 0.6, your demand is inelastic. That is a fancy word for stubborn. You moved price a lot and volume barely flinched. If the size is more than one, your demand is elastic, which means springy. You nudged price and volume moved even more. Less than one, stubborn. More than one, springy. That is the whole vocabulary and you now have all of it.
A quick word on that minus sign
One tidy up before we go further, because it trips people. Elasticity is nearly always a negative number, because price and volume pull in opposite directions: push price up, volume comes down. So when I say an elasticity of 0.6, or of 1.4, I am talking about the size, the bit after the minus, and I will keep doing that all the way through. Economists call this the absolute value, and all it means is we agree to ignore the sign and read the size, because the size is the thing that tells you stubborn from springy. If you ever see someone quote a positive price elasticity, they have either dropped the sign like I am doing, or something very strange is going on with their product. For everything below, read every elasticity as its size.
There is one more line worth learning, because it turns elasticity into a decision instead of a description.
Read it slowly. If your elasticity size is below one, then one minus that size is positive, so a price rise lifts revenue. If your elasticity size is above one, one minus it goes negative, so a price rise cuts revenue. And right at one, the two cancel and revenue does not move at all. That single line is the whole game written in shorthand, and we are about to watch it play out in money.
The first thing that should stop you in your tracks
Here is the moment that gets people, and it got me. If a ten percent price rise loses you less than ten percent of your volume, you are underpriced, full stop, and you are handing money away every single day you do not act on it.
Look at why. You raised price by ten percent. You lost six percent of your buyers. But every buyer who stayed is now paying you ten percent more. You gave up a small slice of volume and got a bigger slice of margin on everyone who remained. You are richer. We will prove it with the money in a moment, but sit with the shape of it first, because it is the opposite of what fear tells you. Fear says a price rise costs you customers. The maths says that when demand is stubborn, a price rise pays you, because the customers you keep more than cover the few you lose.
And here is the uncomfortable part. Most owners never test this. Not because they are lazy, but because they are frightened of losing customers they would in fact keep. They imagine the six percent walking out and they never picture the ten percent more coming in from everyone else. So they sit at a price they picked years ago by copying a competitor or by adding a round number to their cost, and they never once move it ten percent to see what actually happens. The test is nearly free. The fear is expensive.
Now show me the money
Talk is easy, so let us put revenue on it. Imagine a product you sell at fifty euros, and at that price you shift a thousand units a period. That is fifty thousand euros of revenue. Now walk the price up and down and watch what your stubborn, inelastic demand does to the total.
| Price move | New price | Volume | Size of elasticity | Total revenue |
|---|---|---|---|---|
| Cut 10 percent | 45 | 1,060 | 0.6 | 47,700 |
| No change | 50 | 1,000 | reference | 50,000 |
| Raise 10 percent | 55 | 940 | 0.6 | 51,700 |
| Raise 20 percent | 60 | 870 | 0.65 | 52,200 |
| Raise 30 percent | 65 | 780 | 0.73 | 50,700 |
| Raise 40 percent | 70 | 670 | 0.83 | 46,900 |
Read the revenue column top to bottom. Cutting your price, the thing owners reach for when they want more sales, actually loses you money here: you drop to 47,700 because you gave a discount to the very people who would have paid full price. Standing still leaves you at fifty thousand. Raising price lifts revenue, all the way up to a peak around 52,200 when you have raised by twenty percent, and only then does it start to fall away.
That peak is the whole point. There is a price that maximises your revenue, and for stubborn demand it is almost always higher than the price you are charging now. The revenue maximising point sits where a further small rise would finally cost you as much volume, in percentage terms, as it gains you in price. That is exactly the price where the size of your elasticity passes one, where stubborn tips over into springy. Below that price you are underpriced and climbing costs you nothing. Above it you have pushed too far. Your job is to find that top of the hill, and almost nobody is standing on it. Most are sitting well down the near side, too scared to climb.
Now read the same table as profit, not revenue
Here is the next jolt, and it is the one that turns a nice chart into a decision you can actually feel in the bank. Revenue is not what you keep. Profit is. And once you bring cost in, the case for raising price gets stronger, not weaker, because the units you shed were the ones costing you money to make and ship.
Say each unit costs you thirty euros to produce and deliver. At fifty euros that is twenty euros of margin a unit. Watch what happens to the money you actually keep as you walk the same prices.
| Price move | New price | Volume | Unit cost | Unit margin | Total gross profit |
|---|---|---|---|---|---|
| Cut 10 percent | 45 | 1,060 | 30 | 15 | 15,900 |
| No change | 50 | 1,000 | 30 | 20 | 20,000 |
| Raise 10 percent | 55 | 940 | 30 | 25 | 23,500 |
| Raise 20 percent | 60 | 870 | 30 | 30 | 26,100 |
| Raise 30 percent | 65 | 780 | 30 | 35 | 27,300 |
| Raise 40 percent | 70 | 670 | 30 | 40 | 26,800 |
Now look where the peak has moved. On revenue alone, the top of the hill was a twenty percent rise. On profit, the top of the hill has slid further out, to a thirty percent rise, where you keep 27,300 against the twenty thousand you started with. That is the second A ha of the whole piece: the moment you count cost, the profit maximising price sits even higher than the revenue maximising one, because every buyer you lose takes their cost of goods out the door with them. You are not just keeping more per sale. You are quietly firing your least profitable orders.
This is why the fear of a price rise is almost always backwards. The customers who leave first are usually the most price sensitive and the least loyal, the ones who cost you the most to serve and complain the loudest. Raising price does not only lift margin, it improves the mix of who you sell to. Fewer, better orders, more money kept. If you take one table from this article into your next planning meeting, take this one, not the revenue one.
The straight line hiding in a curve
One more tool, because you are going to want more than a single before and after reading, and this is the one economists reach for. It sounds technical and is not.
All that says is this. Take the logarithm of your quantity and the logarithm of your price, which is just a way of measuring things in percentage steps instead of absolute ones, and plot one against the other. A demand curve that looks bent and awkward on a normal chart becomes a straight line. And the slope of that straight line, how steeply it tilts, is your elasticity directly. That is why this is called the constant elasticity model, or the log log model because both sides are in logs: it assumes one elasticity that holds across your whole price range, and it hands you that single number as the slope.
In practice you do not do this by hand. You take your own history, every price you have charged and the volume you sold at each, or better still a handful of deliberate price tests, and you fit that line. The slope drops out, and now you have your elasticity as one honest number pulled from your real sales rather than a guess. This is straight out of the standard marketing analytics playbook, the price and advertising elasticity chapter in Cutting Edge Marketing Analytics by Venkatesan, Farris and Wilcox, and it is the same move whether you are a corner shop or a listed brand. The only thing that changes is how much data you feed it.
Turning two price points into a slope by hand
You do not need software to feel how this works, so let us do it once with a pencil. Take two rows from our own table, the no change row and the twenty percent rise row. Price went from fifty to sixty, volume went from a thousand to eight hundred and seventy. To read a proper elasticity across a big jump like that, you compare the percentage move in log terms, and you can do it in four small steps.
Step one, the top of the fraction. The natural log of 870 is about 6.768, the natural log of 1,000 is about 6.908, so the top is 6.768 minus 6.908, which is minus 0.140. Step two, the bottom. The natural log of 60 is about 4.094, the natural log of 50 is about 3.912, so the bottom is 4.094 minus 3.912, which is 0.182. Step three, divide: minus 0.140 over 0.182 is about minus 0.77, so a size of 0.77. Step four, read it: still under one, still stubborn, still room to climb, which is exactly what the profit table told us when it kept rising past that point. Two data points, four lines of arithmetic, one honest number. That is the whole method in miniature, and every serious version of it is just this done properly across many points at once instead of two by hand.
The second thing that should stop you
Now the part almost nobody connects, and it is the third jolt. That same book measures two elasticities side by side: how much your volume responds to price, and how much it responds to advertising. And here is the pattern that shows up again and again across real studies. Price elasticity is usually big. Advertising elasticity is usually small.
Think about what that means for where you put your effort. If your demand is springy, elastic, that often means you are propping up your volume with promotions, discounts and ad spend, working hard and paying for every extra sale. If your demand is stubborn, inelastic, you have pricing power you are not using, and a small price rise drops almost entirely into profit while a euro of advertising barely moves the needle. A point of margin from pricing and a point of margin from advertising are not the same price to buy. Pricing is very often the far less expensive lever, and it is the one gathering dust, because moving a price feels scary and buying an ad feels like doing something.
So the two questions are really one question asked twice. How springy is my demand to price, and how springy is it to advertising? If price barely moves volume but ads do, you are living on advertising you might not need. If ads barely move volume but price does, you are underpriced and your ad budget is papering over it. Either way the elasticities tell you which lever is actually earning its keep, and most businesses are leaning on the expensive one.
Two elasticities on one grid
The cleanest way I know to place your own business is to hold both elasticities up at once. Put price sensitivity on one axis and advertising sensitivity on the other, and four rooms appear. Find the room you live in and the next move almost picks itself.
| Your situation | Price elasticity | Ad elasticity | What it means | The move |
|---|---|---|---|---|
| Underpriced and quiet | low, stubborn | low | You have pricing power and ads do little | Raise price, it drops to profit |
| Underpriced but buying reach | low, stubborn | high | Price could climb, ads still find new buyers | Raise price first, keep sensible spend |
| Fairly priced, ad led | high, springy | high | Volume leans on promotion | Hold price, protect the ad engine, build value |
| Squeezed on both sides | high, springy | low | Price bites and ads barely help | Fix the product and the offer before either lever |
Most owners I meet are sitting in the top left room, underpriced and quiet, without ever having checked. They assume they are in the bottom row, squeezed and springy, because that is what the fear whispers, and they never run the ten percent test that would show them the truth. The grid is not there to be admired. It is there to send you to the one lever that is actually loose.
So which are you
Here is the decision on one screen, the way I would sketch it on the back of a napkin for an owner.
It loops on purpose. You do not run this once and retire. You move price, you read the volume, you act, and then you test again, because your elasticity is not carved in stone. It shifts with the season, with a new competitor, with a better product. The owners who win treat price as a dial they are always gently turning and measuring, not a sticker they applied once and never touched.
A worked mini case: the coffee roaster
Let me make it concrete with a small business the way I actually meet them. A friend roasts coffee and sells a house blend online. He charges eighteen euros a bag and moves five hundred bags a month, so nine thousand euros of revenue. Each bag costs him about ten euros in beans, packaging and postage, so eight euros of margin, four thousand euros of gross profit a month. He is convinced that if he touches the price his regulars will bolt to the supermarket, so he has held eighteen euros for three years and instead pours money into ads to keep volume up.
We ran one clean test. He lifted the price to twenty euros, a rise of about eleven percent, on the website only, and left it for six weeks while nothing else changed. Volume settled at about four hundred and sixty bags. Let us read it.
Volume fell from five hundred to four hundred and sixty, a drop of eight percent, against a price rise of eleven percent. Size of elasticity, roughly eight over eleven, about 0.7. Stubborn. Under one. Underpriced, exactly as the fear had hidden from him. Now the money.
| Measure | Before | After | Change |
|---|---|---|---|
| Price per bag | 18 | 20 | up 11 percent |
| Bags per month | 500 | 460 | down 8 percent |
| Revenue | 9,000 | 9,200 | up 200 |
| Unit margin | 8 | 10 | up 2 |
| Gross profit | 4,000 | 4,600 | up 600 |
Look at the last row. Revenue barely moved, up two hundred, which is the part that scares people into thinking a price rise is not worth it. But gross profit jumped six hundred euros a month, from four thousand to four thousand six hundred, a fifteen percent lift, because he shed forty of his least profitable bags and made two euros more on every one of the four hundred and sixty he kept. Six hundred euros a month is seventy two hundred euros a year, from one afternoon changing a number on a page. And he had not even climbed to the top of the hill yet, because at a size of 0.7 there was still room to push. That is a real business, and the fear had cost him thousands a year for three years running.
The mistakes that quietly ruin a price test
Before you run off and yank a price, learn the ways this goes wrong, because a badly run test gives you a confident number that happens to be false, which is worse than no number at all. These are the ones I see again and again.
The first mistake is moving the price too little. A one or two percent nudge is swallowed by ordinary week to week noise, and you cannot tell the signal from the wobble. Move it by a real amount, ten percent, so the effect is bigger than the noise around it.
The second mistake is not waiting long enough. You change the price on Monday, sales dip on Tuesday, you panic and change it back on Wednesday. That tells you nothing. Give it weeks, not days, so a slow patch or a payday or a competitor's promotion does not masquerade as your result.
The third mistake is changing two things at once. If you raise the price and launch a new advert and redo the packaging in the same fortnight, you will never know which one moved volume. Change one thing, hold everything else still, read the result, then change the next thing. One lever at a time is the whole discipline.
The fourth mistake is ignoring seasonality. Comparing December to January and calling the difference elasticity is how you convince yourself a price rise destroyed demand when really the season did. Compare like with like, ideally the same weeks against a fair baseline, or the same period a year earlier.
The fifth mistake is testing on your whole book at once. You do not need to bet the business. Move the price on one product, or one region, or the website only, keep the rest as a control, and compare. A contained test with a control beats a brave test with none.
The sixth mistake is not being able to read your own numbers cleanly. If your sales data lives in three disconnected places and you cannot pull volume by product by week without a fight, you will give up before the reading is clear, and you will fall back on the gut feel that got you underpriced in the first place. Owning tidy, queryable sales data is not a luxury here. It is the thing that makes the whole test possible.
How to apply this in the next ninety days
Enough theory. Here is what I would actually do if this were your business and you gave me a quarter to prove it. You do not need a data science team and you do not need to bet anything you cannot afford to.
Month one, choose and measure. Pick one product or one segment that matters but will not sink you if the test wobbles, and nail down a fair baseline: how many units you sell at today's price, week by week, cleanly attributed. If you cannot see that easily, fixing your data comes first, because everything downstream depends on a clean before.
Month two, move and hold your nerve. Raise the price by a real ten percent and change nothing else. No new advert, no repackage, no discount code. Just the price, held steady long enough to read a clean signal, six weeks if you can manage it, while you watch volume against the baseline.
Month three, read and decide. Work out the size of your elasticity from the volume move, the same four lines of arithmetic we did by hand. If the size is under one you are underpriced, so raise again and repeat, climbing toward the top of the hill one careful step at a time. If the size is over one you have found your ceiling, so hold, and turn your attention to your advertising elasticity and to the product itself. Either way you now know something true about your business that you were only guessing at before.
Here is the practical spine under all three months. First, pick one product and actually move the price by a real amount, and leave it long enough to read a clean signal. Second, measure volume properly against a fair baseline, which means owning your own sales data in a system that remembers, rather than a rented box that hands you a pretty dashboard and hides the raw numbers. If you want a shop that actually lets you run and read a price test, that is the argument for owning your ecommerce platform rather than renting one. Third, once you have a few readings, fit the line. Two or three honest price points already give you a slope, a real elasticity, and that beats every gut feel in the building.
If you want a hand turning your own price history and sales into an elasticity you can actually bank decisions on, that is squarely the kind of work I do, and it is what my performance and data work is built around. Bring your numbers, we will find the top of your hill.
The one thing to leave with
If you take a single idea from this, make it this one. Price elasticity is not an economics lecture, it is one question you can answer about your own business: if I move my price ten percent, how much does volume move? If the answer is less than ten percent, you are underpriced and every day you wait is money left on the table, out of fear of losing customers you would have kept anyway. And remember the coffee roaster: the fear cost him thousands a year while the fix took one afternoon. Stop guessing your price. Move it, measure it, and let the number decide. The top of the hill is almost always higher than where you are standing.