Home Philosophy How it works FAQ Blog Pricing Try free

        Pavel Logačev/December 2025/9 min read
      

How to Pick the Right Discount (Instead of Guessing 20% Off)

Most merchants pick discount percentages by gut feel. Here's how to figure out what level actually maximizes your profit.

You run a promotion because you want to grow — more customers, more visibility, more word of mouth. A discount is the price you pay to make that happen. It's an investment.

But how much should you invest? Is 10% off enough to get the growth you want, or do you need 20%? Most merchants pick a number that feels right — or copy what they did last time — and hope for the best. There are hundreds of apps that help you create discounts. Almost none that help you figure out the right level.

That matters, because a 20% discount might buy you the exact same growth as 15% off. If so, that extra 5% is just margin you gave away for nothing. The goal isn't to minimize discounts — it's to not overpay for the growth you're after.

Know what you're actually spending

Before you can judge whether a discount is worth it, you need to know what it actually costs you. And here most merchants get it wrong: a discount comes out of your margin, not your revenue.

Say you sell a product for $50. Your cost is $30. That's a $20 margin per unit.

Discount
Sale price
Margin/unit
Margin lost

0% (regular)

$50.00

$20.00

—

10% off

$45.00

$15.00

25% of margin

15% off

$42.50

$12.50

37.5% of margin

20% off

$40.00

$10.00

50% of margin

30% off

$35.00

$5.00

75% of margin

A 20% discount on a product with 40% margins cuts your profit per unit in half. To break even on that promotion, you need to sell twice as many units as you would have without the discount.

Know the price of your investment

A 20% discount on a 40% margin product means you need to sell twice as many units just to break even during the promotion. That might still be worth it if the promotion grows your customer base and lifts your baseline afterwards. But you should know what you're spending — so you can judge whether a smaller discount would buy you the same growth for less.

Why the same discount doesn't work for every product

Generic advice says "run 15% off, it's the sweet spot." But 15% off can be brilliant for one product and a waste for another — depending on how your customers respond to price changes.

Product A: High price sensitivity

Customers are very responsive to price. Small discounts drive big volume increases.

Regular: 50 units/mo
10% off: 85 units (+70%)
20% off: 110 units (+120%)

10% off is enough. Going to 20% adds 25 more units but costs $5 more per unit in margin. The extra volume doesn't make up for it.

Product B: Low price sensitivity

Customers barely respond to price changes. Discounts don't drive much extra volume.

Regular: 50 units/mo
10% off: 54 units (+8%)
20% off: 59 units (+18%)

Neither discount is worth it. You're giving up margin on all 50+ units to gain just a handful of extra sales.

Same products. Same starting sales volume. Completely different answers. The "right" discount depends entirely on how sensitive each product's customers are to price — and you can't know that without looking at your data.

What your past promotions already tell you

You've probably run several promotions over the past year. Each one is a data point. Here's how to read them:

1

Compare the volume lift to the discount level

Look at each product that was on sale. How many units did you sell during the promotion versus a normal period of the same length? The ratio matters more than the absolute number. If you ran 20% off and sales went up 30%, that's a modest response. If sales went up 200%, that's a strong response.

2

Check if bigger discounts actually drove proportionally more volume

If you've run different discount levels at different times — say 10% in spring and 20% in fall — compare the volume lifts. Did doubling the discount double the extra sales? If not, the bigger discount was probably overkill. Often, you'll find that going from 10% to 20% off adds far less incremental volume than you'd expect.

3

Account for the event, not just the discount

This is the hardest part. Your Black Friday sales went through the roof at 25% off. But was that the discount — or was it Black Friday? If people were going to buy anyway because it's the biggest shopping weekend of the year, a large chunk of those sales would have happened at a smaller discount, or even at full price. Promotional events amplify demand on their own. Don't give the discount all the credit.

4

Look at which products respond and which don't

During your last sitewide sale, some products probably saw a big spike and others barely moved. That's useful. The products that spiked are price-sensitive — they're good candidates for targeted promotions. The products that didn't move much are poor discount candidates — you're just giving away margin to people who would have bought at full price.

How to run smarter promotions going forward

You probably can't undo past promotions, but you can design better ones starting now.

Don't discount everything the same amount

A sitewide "20% off everything" is simple, but it's wasteful. Your price-sensitive products don't need 20% to move. Your price-insensitive products won't move much even at 20%. Instead, consider tiered discounts: 10% on some products, 15% on others, 20% on a few high-sensitivity items where you'll actually see the volume.

Calculate the break-even volume before you commit

For each product, before running a discount, do this quick calculation:

Regular margin per unit = price − cost
Discounted margin = discounted price − cost
Break-even units = regular units × (regular margin / discounted margin)

If you normally sell 50 units at $20 margin, and the discount brings margin down to $12.50, you need to sell 80 units just to break even. Is that realistic for this product? If not, shrink the discount.

Test different levels on different products

Next time you run a sale, don't put every product at the same discount. Try 10% on one group, 15% on another, and 20% on a third. Keep the groups roughly similar in terms of product type and sales volume. After the promotion, compare the results. This gives you real data about which discount levels actually drive profitable volume for different parts of your catalog.

Track incremental profit, not just revenue

A promotion that generated $10,000 in revenue sounds great. But if you would have sold $7,000 anyway, the promotion only generated $3,000 in incremental revenue — at a lower margin. Subtract the margin you gave up on the sales that would have happened at full price, and the actual profit impact might be negative. Always compare against what you would have sold without the discount.

Mistakes that cost merchants money

"We always do 20% off."

If you've never tested whether 12% or 15% would drive the same volume at better margins, you're probably giving away money. There's a good chance a smaller discount works just as well for many of your products.

"The sale was a success — we sold a lot."

High volume during a sale isn't automatically a win. If the extra volume doesn't exceed the margin you gave up, the sale lost you money. Revenue is vanity, profit is sanity.

"Our BFCM discount worked great, so we'll use it year-round."

Black Friday shoppers behave differently from normal shoppers. The lift you saw during BFCM includes the effect of the event itself, not just the discount. A 25% discount in January will not produce the same results.

"Discounts train customers to wait for sales."

This one's actually true — and often overlooked. Frequent discounting can condition your regular customers to delay purchases until the next sale. If your repeat customers start buying mostly during promotions, your "extra" volume is partly cannibalized from future full-price sales.

The real payoff of a promotion isn't during the sale

Everything above focuses on the immediate math: did the extra volume during the promotion cover the margin you gave up? That matters. But it's not the whole picture.

The more important question is: did your baseline go up afterwards?

If you were selling 50 units a month before the promotion, and after the promotion you're selling 55 units a month at full price — that lasting lift is the real return. It doesn't matter whether it's because discounted buyers came back, or they told friends, or the promotion got you more visibility in search results. The mechanism doesn't matter. The number went up.

This reframes the entire question. The cost of a discount is the margin you gave up. The benefit is any lasting increase in demand. Was the trade worth it?

A simple way to check

After every promotion, compare your average weekly sales for the month before the promotion to the month after (not during — that's the promotion itself). If the "after" number is higher than the "before" number, the promotion did something lasting. If it's the same or lower, the promotion only pulled demand forward — people who would have bought anyway just bought sooner at a lower price.

This also means that not all discounts need to be immediately profitable to be worth running. A promotion that loses a bit of margin during the sale but permanently lifts your baseline by 10% is a great deal. And a promotion that looks profitable on paper but just cannibalized next month's sales is actually a loss.

When discounts genuinely make sense

None of this means discounts are bad. They're a tool — and like any tool, they work when used deliberately. Discounts make sense when:

→They grow your baseline — the promotion brings in new customers or generates word of mouth that lifts your ongoing sales. This is the best outcome: you pay a short-term cost for a long-term gain.

→Clearing slow-moving inventory — you need the shelf space or the cash more than the margin. Even a money-losing sale can be rational if the alternative is dead stock.

→Price-sensitive products where the immediate math works — if a product has high sensitivity and decent margins, a targeted discount can increase profit even during the sale itself.

→Seasonal products approaching end of season — a progressive markdown strategy makes sense. But start smaller than you think — you can always go deeper, but you can't take a discount back.

About Sell Smart Price Optimization

Sell Smart Price Optimization analyzes your Shopify order history to estimate how price-sensitive each of your products is. That same analysis tells you which products are good discount candidates, how much volume a discount will actually drive, and what discount level maximizes your incremental profit — so you can stop guessing and start running promotions that actually make you money.

Try free How it works

Related

How to Find Your Optimal Price Without A/B Testing → Your sales data already contains the information you need to price better. When A/B Testing Prices Makes Sense — And When It Doesn't → When price testing works for Shopify stores, and what to do instead.

In this article

What you're actually spending Not every product What past promotions tell you Smarter promotions Costly mistakes The real payoff When discounts make sense

Which products to discount

Sensitivity per product tells you where a discount pays for itself — and where it doesn't.

Read the philosophy →

Sell Smart Price Optimization · apps.shopify.com/smart-price-optimization

Live demo Contact Privacy policy Impressum