How to increase AOV on Shopify: bundles, upsells and free-shipping thresholds

Will bundles, upsells or free shipping actually raise your AOV?

To calculate your average order value (AOV), divide your revenue for a period by the number of orders in that same period.

Bundles, upsells, and free-shipping thresholds can increase that number. The problem is that they can also reward customers for something they were going to do anyway, such as giving a discount to a shopper who was going to buy two items anyway.

Let’s say a customer was always going to buy two things. They see a bundle discount and they take it. In doing so you have given away the margin on an order that would have happened at full price. Free shipping is the same thing when you pay for the label on orders that would clear the line anyway.

AOV itself can also be misleading; a change that increases AOV but reduces the number of orders can still mean an overall loss of money.

This is the main reason why we judge store tests on revenue per visitor first and only then check AOV and conversion rate and profit per order underneath it.

What this approach means in practice is that every offer is a guess that has to prove itself. You should write down the number it has to beat before launch, run it as a split test and keep it only if it clears that number.

Why is redemption rate a misleading success metric?

The redemption rate tells you how many shoppers used the offer, not whether they bought more than they would have otherwise. An offer can look great in a weekly report while basket size stays flat underneath it.

That happens when the team measures the offer instead of what shoppers did. The promotion 'worked' on paper, but the store was just giving away margin on orders it would have gotten anyway.

What to track instead

Rather than tracking how many shoppers used the offer, make sure to compare the store’s numbers before and after the offer went live. You will need four numbers for this and you will have to read them together, not one at a time.

Revenue per visitor is the most important of the four, as it increases when shoppers buy more per order and when more visitors buy.

Next, look at how many of your orders are made up of two or more items and what the average number of units is per order. These two figures will tell whether the offer actually made people buy more products or if the baskets look the same as before.

Finally, review your profit per order. This is where you find out whether the extra sales were worth the cost of the discount or free product you gave away. If the share of orders with 2+ items remains unchanged after the offer goes live, it means the offer has not changed the way people shop. All it is doing is taking your margin down, even if many shoppers are using it.

Where should upsells go on a Shopify store?

You should place them wherever a test indicates they generate more per visitor and watch for anything on the page that allows shoppers to skip them. The most common cause of this behavior, in our experience, is express checkout buttons.

Wedge Guys, a golf grip brand we work with, tested adding express checkout buttons for Shop Pay, PayPal and Apple Pay to the cart. Over 30 days, the version with the buttons produced 9.1% less revenue per visitor than the version without them.

Orders were flat (543 vs 542) but AOV was 5.7% lower and revenue from cart add-ons was down 80%. The shoppers who went straight to checkout didn’t even see the add-ons.

Next, we tested hiding the Shop Pay express button from the Buy Box on the product page. That particular version won on revenue per visitor, generated more orders at a similar conversion rate and sent more shoppers to the frequently bought together offer after adding a product to the cart.

Confidence on this test was 86%, below the 95% many teams wait for before they act. It pointed the same way as the cart test, so we recommended rolling it out.

A faster checkout and a better checkout are not always the same thing for your store. Every shortcut that gets shoppers past your upsell (if it lives in the cart) ends up costing you some of that add-on revenue.

How do you set a keep or revert rule before you test?

You have to put the decision rule down before the offer is live. If the rule is established up front, then nobody on the team can look back at a small lift and say that it’s a win. We follow the same five steps for any bundle, upsell or threshold change.

1. Pull a baseline from a comparable period

Write down your revenue per visitor, AOV, share of 2+ item orders, and profit per order from a period that looks like the one you are about to test.

2. Pick one primary metric

Most often the primary metric will be revenue per visitor. If you have a main number, it stops the team from choosing whatever number moved once the results come in.

3. Write down the keep and revert numbers

For example, you could keep the offer at +10% revenue per visitor and revert it if the lift comes in under +3%. Base the keep line on your margin, because the offer has to generate enough extra profit to pay for itself.

4. Set the end date

Make sure you run the test for at least two full weeks, even longer if your traffic is low. The Wedge Guys cart test above ran for 30 days.

5. Decide on that date using only the rule

If the result lands somewhere between the keep and revert numbers, re-run the test before scaling anything.

This also helps to run only one test per lever at a time. If you change both the offer and the shipping threshold in the same week, you won’t know which one moved the numbers.

Bundles vs. flat discounts: which one lifts basket size?

The main difference between the two is the timing of their payouts. You pay for a bundle only if the shopper adds more units, but a flat discount applies to every order, even if they only contain one item. That’s the difference that lets you actually move your AOV.

The depth of the bundle matters too. One DTC brand we work with has a “Buy 2, Get the 3rd Free” offer, which gives away a third of the price on every 3-unit order.

We proposed replacing it with tiers, whereby shoppers save 15% when they buy 2 and 25% when they buy 3. It still looks like a real bundle deal to the customer but the brand gets a better margin on it.

Check that your discounts do not fight your threshold

Quantity discounts can turn shoppers away from your free-shipping line rather than toward it.

On Wedge Guys’ product page, a $9.49 handle read, “Buy 7+ save 10%." The seven grips come in at $66.43 or $59.79 with the discount. So a shopper who followed the store’s own prompt actually moved further from the $75 free shipping threshold, instead of closer to it.

We designed a test with quantity presets that are tied to the threshold, and the brand added a useful warning while we were setting it up. Many of their customers stack the 10% quantity discount with a 15% coupon, so on one grip a shopper would need 17 units to clear $75.

Because of this, you should model your threshold on the price a shopper pays after every discount they can combine, not on the list price.

Never stack offers inside one test

The stacked deals are what make it impossible to know which one changed the basket size and on top of that, they let shoppers take both.

On one store we manage, a 10% welcome code conflicted with bundle discounts already in the cart. Shopify does not combine these two discounts by default, so we checked with the brand before switching combining on, because it would cut the margin on every bundle order.

When you compare a bundle versus a flat discount, follow these steps:

1. Split your traffic into a bundle cell and a flat discount cell without an overlap so that one offer cannot leak into the other cell.
2. Keep both cells live for the same window, because running them at different times mixes in seasonality and traffic changes.
3. When the window closes, compare the revenue per visitor and profit per order for each cell. Here profit is critical, because the flat discount also applies to single-item orders.
4. Use the rule you wrote before launch and only pick the winning cell if it clears that number.

Free-shipping thresholds: what if your AOV sits just above the line?

When this happens, you end up paying for shipping on almost every order. Free shipping matters to shoppers: according to Baymard Institute’s 2026 data, 40% of US shoppers who abandoned their cart (excluding those who were just browsing) did so because extra costs, such as shipping, taxes and fees, were too high.

But besides helping you win the sale, the threshold has a second job: protecting your margin.

A DTC brand that we work with had an AOV of around €53 against a €50 free-shipping threshold so it was paying for the label on almost every order. Their products were small enough to fit through a letterbox, so the fix was cost-related: moving to letterbox parcels, which are much cheaper than regular parcels.

The cheaper shipping raised the margin on each order and also lowered the break-even return on ad spend the brand needed from its ads.

It is also worth checking how Shopify calculates your threshold. For example, for one of the pet brands that we work with, free shipping was calculated on the cart total after a 10% subscription discount was already applied. Carts that started between $54 and $59.99 before the discount slipped below the $54 mark and lost free shipping.

We built a Shopify Function to set the threshold based on the pre-discounted price and added a progress bar on the product page.

How to test a new threshold

Shopify’s free shipping guide offers an easy way to price a threshold. You take the difference between the new threshold and your AOV, multiply that by your gross margin and compare that to your average shipping cost.

In its example, with a $40 AOV, $8 shipping and a 40% margin, a $45 threshold adds about $2 of gross profit per order and still leaves $6 of shipping uncovered.

When raising the threshold costs you orders

Once you raise the threshold, you may lose orders immediately. We work with another DTC brand who increased free shipping from €40 to €50 and increased their standard shipping fee on the same day.

On that particular day, the orders went down from 38 to 25 with the same offer live and the same ad spend. One of its best Meta ad sets went from a 2.1 to a 0.5 return on ad spend.

We suggested going back to €40, or €45 at most, but the brand kept €50 because it expected its “buy 2, get the 3rd free” bundle to push more than half of carts over the line. Both are reasonable bets, but because the fee and the threshold changed together, there is no way to tell which one cost the orders.

Before you raise the threshold, look at what share of your carts already reach the new number and which offer gets them there, and change one thing at a time. Shipping costs also differ by sales channel, which we break down in our Amazon vs DTC comparison.

From our accounts: what each change showed

Most of these findings trace back to the same problem. It can be a small change on the page, like a single button or a single line of discount, but it can change what shoppers put in the cart and how close they come to free shipping.

Frequently asked questions

What counts as a good result for a bundle test?

The number you agreed on before launch is a good result and that number has to cover the margin the offer gives away. And you should judge it by revenue per visitor and profit per order instead of by how many shoppers purchased the bundle.

How long should an AOV test run?

Run it for at least two full weeks, or longer if your store has low traffic. End it on the date you set at the start, so an offer that isn’t working doesn’t keep eating your margin.

Should Shopify brands use flat discounts or bundles?

We usually begin with bundles, since they only pay out when shoppers add more units, and then confirm the choice with a split test. A flat discount, by contrast, applies to every order, including single-item ones.

Do express checkout buttons hurt AOV?

Yes, they can. Adding Shop Pay, PayPal and Apple Pay buttons to the cart resulted in flat orders but a 9.1% decrease in revenue per visitor over a 30-day test, as shoppers skipped the cart add-ons on their way to checkout.

Where should upsells go on a Shopify product page?

Put upsells wherever testing shows they make more money per visitor. The most common spots we see are under the add to cart button, in the cart drawer and right after a shopper adds a product to the cart. Whichever you choose, make sure express checkout cannot bypass it.

How much does ecommerce conversion rate optimization help cost?

Our services start from $2,000 per month and are custom for each brand, based on the complexity and the amount of work.

Usually a store that requires 1 or 2 offers to test is going to be closer to the starting price than a store where we are rebuilding the cart, the product pages and the shipping setup as well.

The easiest way to get an exact number is to tell us what you want to test and we can put together a quote based on that.

What to check before your next AOV test

Before you launch anything, make sure you write the keep and revert numbers, and don’t forget to make the revenue per visitor your primary metric. Also, track the share of 2+ item orders, as the redemption rate alone will not tell you if the offer changed what people buy.

It is also worth double-checking that express checkout does not allow shoppers to skip your upsell, and you should compare your AOV to your free-shipping threshold every month using prices after discount. And most important of all, change one lever at a time and don’t stack offers in a test.

If your offers are giving away margin on orders you would have had anyway, our ecommerce conversion rate optimization team can design and run your next AOV test, so get your free marketing plan from Olifant Digital today and find out which offers are actually paying for themselves.