NavonaAI logoNavonaAI

prevention

How Much Abandoned Cart Discount to Offer on Shopify

Work out how much abandoned cart discount to offer with plain arithmetic: what each discount costs against your order margin, and how many sales it must win.

NavonaAI Team9 min read

The right abandoned cart discount is the smallest one that still gets a shopper to buy, and it always stays below your gross margin on the order. There is no universal percentage. What you can do is check any number you are considering against two sums: what the discount costs per order, and how many of the orders that use it must be genuinely new sales for it to pay for itself.

Apart from one sourced survey figure near the end, every number below is plain arithmetic you can redo on a calculator. The dollar figures are illustrative, not a benchmark: an order worth $75 with a 50% gross margin, which are also the starting values in our Discount Margin Calculator.

What an abandoned cart discount costs you

A discount comes straight out of your margin. Gross margin here is what is left of the order value after the cost of goods. If you also pay shipping or payment fees on every order, take them off first, so your margin figure is smaller and the check below is stricter.

  • Margin on the order before the discount: order value × gross margin
  • Margin on the order after it: order value × (gross margin − discount)
  • Cost of the discount: order value × discount

On a $75 order at a 50% margin, you earn $37.50. Give 15% off and the discount costs $11.25, which leaves $26.25. Give 50% off and nothing is left: $37.50 of discount against $37.50 of margin.

That sets the first rule. Never pick a discount at or above your gross margin, because every order that uses it then earns nothing or loses money.

The break-even check for an abandoned cart discount

Every shopper who uses your code is one of two kinds:

  • A new sale. They would have left for good, so you earn the margin that remains after the discount, order value × (gross margin − discount).
  • A sale you would have had anyway. They would have bought without the offer, so the discount is pure cost, order value × discount.

You cannot see which kind each shopper is. You can work out how many of each you need. The discount pays for itself when the share of code orders that are new sales is at least discount ÷ gross margin.

Discount on a $75 orderCost of the discountMargin leftShare of code orders that must be new
5%$3.75$33.7510% (1 in 10)
10%$7.50$30.0020% (1 in 5)
15%$11.25$26.2530%
20%$15.00$22.5040%
30%$22.50$15.0060%
40%$30.00$7.5080%
50%$37.50$0.00Never pays for itself

Illustrative numbers: $75 order, 50% gross margin.

Read one row. At 10% off, the discount pays for itself only if at least 1 in 5 of the orders using the code would not have happened without it. At 30%, it needs 3 in 5. The deeper the discount, the more of your code orders have to be real wins.

Here is the same break-even counted another way. The extra orders needed for every order you would have had anyway is discount ÷ (gross margin − discount). For 15% at a 50% margin that is 0.15 ÷ 0.35, or about 43 extra orders for every 100 you would have had anyway. It is the same figure the calculator reports as "Additional orders needed to break even" (42.9% more volume) when you enter $75, 50% and 15%.

One assumption sits under all of this. A "new" sale must be one that would never have happened, not one that was only delayed. A shopper who would have come back next week at full price counts as a sale you would have had anyway.

Your margin sets the ceiling

The same discount is cheap on a high-margin product and expensive on a thin one. Here is a 15% discount on the same $75 order at five different margins:

Gross marginMargin beforeMargin left after 15% offShare of code orders that must be new
20%$15.00$3.7575%
30%$22.50$11.2550%
40%$30.00$18.7537.5%
50%$37.50$26.2530%
60%$45.00$33.7525%

If your catalog mixes products with very different margins, one flat number is a compromise. You can keep your thinnest-margin products or collections out of the discount, which is covered in how to exclude products from your Shopify cart discount.

Percentage or fixed amount?

A percentage discount costs the same share of every cart. A fixed amount does not. Here is $10 off at four cart values:

Cart value$10 off as a share of the cart
$4025%
$7513.3%
$1506.7%
$3003.3%

A fixed amount is generous on a small cart and barely noticeable on a large one. A minimum cart value caps the worst case: $10 off with a $50 minimum never costs more than 20% of the cart, because $10 ÷ $50 = 20%.

If your cart values vary a lot, you can also give each cart-value range its own discount. Tiered discounts by cart value walks through how the ranges work.

Check what stacks before you pick the number

If your offer can combine with a sitewide sale or a code the shopper already has, they can get both, and your margin check has to use the combined discount. In the worst case the two simply add up. A 10% sale plus a 15% offer is 25% off, which on the $75 order leaves $18.75 of margin: half of the $37.50 you would normally earn.

Shopify's own rules decide how discounts combine, and the settings are explained in how Shopify discount stacking works. The point for your sums is to know which offers can land on the same order before you choose a number.

When a discount is not the fix

A discount pays for a reason to buy now. It cannot fix a reason that is really a surprise cost. In Baymard Institute's survey of US online shoppers, 42% have abandoned a cart because they were "just browsing / not ready to buy". Once those shoppers are set aside, Baymard's distribution of the remaining reasons puts "extra costs too high (shipping, tax, fees)" at 40%.

If shoppers leave over a shipping fee that only appears at the last step, a discount is paying for a problem you could show earlier. Our guide to reducing cart abandonment on Shopify starts with that fix.

How to choose your number in four steps

  1. Work out your gross margin per order, after any per-order costs you pay.
  2. Pick a starting discount well below it.
  3. Divide the discount by the margin. That is the share of code orders that must be new sales.
  4. Ask whether you would believe that share. If you would be surprised that many code orders were genuinely new, go lower.

Then test two nearby offers instead of guessing. Be careful reading the result: revenue from orders that used a code is not the same as revenue the code caused, because some of those shoppers would have bought anyway. What a dollar of discount returns explains that difference.

How NavonaAI handles the offer

NavonaAI is built to prevent abandonment while the shopper is still on your store. It does not send emails or SMS. Here is how the offer behaves:

  • It shows only when a shopper has items in their cart and shows signs of leaving, such as heading for the tab bar, going idle, scrolling back to the top or hitting the back button. Shoppers without a cart never see it.
  • Each shopper gets a unique discount code that expires the moment they leave. A regular exit popup offers a generic code that anyone can use at any time.
  • The code is auto-applied at checkout, so the shopper does not copy and paste anything.
  • It does not ask for an email address or phone number.
  • Each shopper sees it once. If they leave without using it, it never comes back.

The settings map onto the sums above:

SettingThe part of the sum it controls
Discount type and amountThe discount in every formula on this page
Minimum cart valueThe smallest cart that can trigger the offer, which caps a fixed amount
StackingWhether the combined discount is the one to check
ExclusionsWhich products or collections stay out of the discount
Tiered OffersA different discount for each cart-value range

Offer tests let you compare two discounts, for example 10% off against 15% off. How to A/B test your Shopify cart abandonment popups covers how to run one. Tiered Offers and offer tests depend on your plan: see pricing.

Related reading

Frequently asked questions