How to Increase Average Order Value on Shopify: 8 A/B Tests ($265 → $673)

I grew the average order value on my own store by more than 150%: from about $265 to about $673 per order. The change that did it wasn’t a pop-up, a progress bar or a discount. It was something most stores never touch, and I’ll show you exactly what it was.

I’ve done this for brands from six to nine figures too. At Karma and Luck, a nine-figure brand, we grew average order value by 20%. What would 20% more per order mean for your store? On a store doing $100,000 a month, it’s about $240,000 more a year. Same traffic, same ads.

In this guide I’ll show you how to raise AOV on three levels (your products, your ads and every page of your website), with real A/B tests, the psychology behind them, and one tactic that can raise your AOV and still lose you money.

Why average order value is worth the work

Your revenue is three numbers multiplied together: visitors × the share who buy × how much each order is worth. That last number is your average order value.

Most store owners fight for the first two. More traffic costs money. A higher conversion rate takes a lot of work and testing. AOV takes work too, but it has one big advantage: these people are already buying. You’re just helping them buy a little more.

It changes your ads as well. If a customer costs you $50 and spends $100, your return on ad spend is 2. If the same customer spends $120, it’s 2.4. Same cost, more profit, and now you can afford to pay more for every customer, which means you can outbid your competitors. (More on that in how to lower your Meta cost per purchase.)

Bad, good and great AOV growth

How much can you realistically grow it? Here’s how I read the numbers:

AOV growth	What it usually means
Under 5%	Bad: you changed something small, or nothing at all
5–15%	Good: one strong change, like a well-built bundle or the right free-shipping threshold
20%+	Great: almost never one change; several winning tests stacked

In money, for a store doing $100,000 a month with a $100 average order (1,000 orders a month):

AOV increase	New AOV	Extra per month	Extra per year
+5%	$105	$5,000	$60,000
+10%	$110	$10,000	$120,000
+20%	$120	$20,000	$240,000

There’s one condition: all of this only works if your conversion rate doesn’t drop. Push people to spend more and get fewer of them to buy, and you can lose money. That’s why everything here is judged on revenue per visitor, not AOV alone.

Why people spend more: the psychology

None of this is magic. It’s a handful of effects behavioural researchers have measured for decades. Every tactic later in this guide uses one of them, so once you know them you can invent your own tests.

1. A goal (the goal-gradient effect). The closer people get to a goal, the harder they push to reach it. Coffee-shop customers with a loyalty card bought more often the closer they got to their free drink. A free-shipping threshold works the same way: “spend $75 and shipping is free” gives people a finish line, and they walk the last few steps. Studies of online stores confirm that shipping thresholds change how much people put in the basket.
2. The word “free”. People value something free far beyond its actual price. A free gift often pulls harder than a discount worth the same money.
3. An anchor. The first price people see changes how every other price feels. Put a three-pack next to a single item and the single item looks small while the three-pack looks smart. It’s also why the middle of three options wins so often.
4. Other people. “Most popular” or “Bestseller” makes a choice feel safer.
5. Fear of losing. People hate losing more than they like winning. “You’re $18 away from free shipping” makes paying for shipping feel like a loss.
6. Less effort. If adding a second product takes one click, people do it. And too many choices stop people completely: in a famous study, a table with 24 jams sold far less than a table with 6.

Level 1: product strategy, the biggest lever

Most AOV advice skips this part. Progress bars, upsells and cart bundles usually move your average order by 5, 10, maybe 15%. The thing that moved mine by more than 150% was none of those. It was changing what I actually sell.

No progress bar can turn a $30 product into a $100 order. Your products set the ceiling. Here’s what to look at.

Sell sets as real products

Say you sell coffee and a bag of beans costs $18. Most stores add a little box under it: “add a grinder and save 10%”. Almost nobody clicks it. Instead, create a new product called “Coffee Starter Kit”: beans, a hand grinder and a pour-over, with its own photos, its own page and its own price, $89. Now you’re not selling beans; you’re selling great coffee at home.

Or skincare: a cleanser costs $25, but “The Morning Routine” (cleanser, serum and moisturiser, step one, two, three) costs $79. The customer doesn’t have to work out what goes together; you did it for them. That’s why people buy the set.

Give your bestseller a better and a best version

Your bestseller is a $49 hair dryer. Keep it and call it Essential. Next to it, add a Pro version at $79 with a diffuser and travel case, marked “most popular”, and a Salon version at $129 with every attachment.

Many people who would have bought the $49 dryer now pick the $79 one, and the $129 version exists mostly to make $79 feel reasonable. It’s the same reason coffee shops sell small, medium and large: most people take the medium.

Bigger sizes and multi-packs

If people use your product up, sell more of it at once. One pair of socks costs $15; a three-pack $39 ($13 a pair); a six-pack $69 ($11.50 a pair), marked “best value”. Same with protein powder: 1 kg, or 2.5 kg at a lower price per kilo. Always show the price per unit and how much they save.

Products that go with what people already buy

Running shoes at $120? Add running socks, insoles and a shoe cleaner. Coffee machines? Descaler and a milk frother. Candles? A wick trimmer and refills. Each one is cheap compared with the main product, so it’s an easy add, and refills bring customers back.

Look at the other side too. If your range is full of cheap entry products, they pull your average down. Sometimes the smartest move is a higher price or fewer cheap items. Watch your margin, and test it like everything else (see the price test that made $37,219 more).

Level 2: your ads set your AOV

Your average order starts before anyone reaches your store. It starts with what you show in your ads.

• An ad for your cheapest product attracts people who want your cheapest product, so first orders are small.
• An ad for a set, a bundle or a higher-priced bestseller makes the first order bigger from day one.
• The offer in the ad sets a target before people land: “free shipping over $X”, “buy two, get one free”, “a free gift when you spend more”.

Then there’s how Meta delivers your ads. Most stores optimise for the number of purchases. You can instead tell Meta to optimise for the value of purchases (“maximise value of conversions”), optionally with a ROAS goal, so Meta looks for people who tend to spend more. Meta’s own study reports about 12% higher ROAS on average for value optimisation. That’s Meta’s number, so test it yourself, and it only works if your pixel or Conversions API sends purchase values and you have enough purchases. If you sell one product at one price, it won’t help.

How to measure AOV per ad

In Ads Manager, open your columns and add “Average purchases conversion value”. That’s your AOV per ad.

Ad	Cost per purchase	AOV	ROAS
Ad 1	$40 (cheapest)	$78	under 2
Ad 2	$50	$142	almost 3

Most people would scale Ad 1 because it has the cheapest cost per purchase. Ad 2 is the better ad, and you’d never see it by looking at cost per purchase alone. So always look at cost per purchase and AOV together, or simply at ROAS, which already includes both.

Then do more of what works: if bundle ads bring bigger orders, make more bundle ads. And send each ad to the matching page. If the ad shows a bundle, link straight to that bundle, not your homepage. (Catalog ads do this automatically: my catalog ads strategy.)

Check where your AOV comes from

In Shopify Analytics, look at AOV by channel and for new versus returning customers. Paid social often brings smaller orders than search or email, and new customers often spend less than returning ones. If Meta traffic brings the smallest orders, start with those ads and the pages they land on. If new customers spend much less, focus on the first order: bundles and the free-shipping threshold.

Level 3: your website, page by page

I keep a library of more than 1,000 published e-commerce tests. These are ones that moved average order value. Don’t treat any of them as something that will work for you: the same change can win on one store and lose on another. Use them as ideas to test.

Homepage: set the target early

The simplest move is a clear free-shipping threshold at the top. A skincare brand added “Free shipping over $75” right above the Shop button: orders +90%, AOV +7%.

The second lever is what you show first. A plant nursery added smart recommendations and its own merchandising rules on the homepage: AOV +20%. Another store moved its category section above the big promo banner so people found the right products faster: AOV +11%, revenue per visitor +13%. If the first products people see are your cheapest, that’s what they’ll buy; lead with bestselling sets and bundles.

Collection pages

1. Sorting. Never default to “price, low to high”; you’re training people to buy your cheapest item. Sort by bestsellers.
2. Sets inside the grid with a clear badge (“Set, save 15%”). A clothing brand did this and mobile revenue per user rose more than 3%.
3. Price per unit for anything people use up, so the bigger pack looks smart.
4. Social proof on the card. A memorial-products store added star ratings and review counts to every card: conversion +23%, AOV +70%+. Extreme, and not fully conclusive, but it shows how much trust changes what people will spend.

Product pages: bundles and sizes

The two that work best are bundles (“frequently bought together”, one button to add all) and quantity or size options, with the middle option marked “most popular”.

One of my favourite tests: a brand selling products in different bag sizes showed the price per kilo and the saving on bigger sizes. 50/50 split, 22 days: AOV $100.43 → $111.33, and because conversion went up too, revenue per visitor +27.5%.

And an honest one: another brand added a “best value: three for the price of two” tab. On more than 40,000 orders, AOV rose just 2%. Not every bundle is a big win.

Don’t guess what to bundle: open your orders and see what people already buy together. And if bundles don’t fit your products, try a free gift above a certain amount (“spend $150, get a free travel case”). Thanks to the zero-price effect, a gift often feels more valuable than a discount and costs you less.

The cart: goals, progress bars and one add-on

The cart is where goals work best, because people are already buying.

• A free-shipping progress bar: “You’re $18 away from free shipping.”
• Stacked goals: free shipping, then 10% off, then 15% off, then a gift. Every step pulls people to the next, but too many steps can confuse, so test it.

Real results: a skincare brand added a second goal (a free gift after free shipping) and, on more than 100,000 visitors, AOV went from $66 to $71. Another brand used several gift thresholds: revenue per visitor +12.2%. A sportswear brand showed tiered discounts in a progress bar: mobile AOV almost +5%. A big cosmetics brand showed how much was left until a discount: average basket +8%.

The second cart lever is one relevant add-on, not ten. A fashion store replaced its basic cart with one that had a shipping bar, recommendations and a gift tracker: AOV +15%, revenue per visitor +13%. A furniture store offered a leather care kit with one click to people buying leather sofas: +$55 per order, estimated at more than $2 million a year.

The other side: a controlled test found that switching to a cart drawer lowered conversion 4.4% and revenue per visitor 6.6%. And on one store I worked on, a customer review inside the cart drawer lowered revenue per visitor, so we paused it (that test is here).

Checkout and after the purchase

One-click add-ons inside the checkout steps are only possible on Shopify Plus (is Plus worth it?), and anything before payment can distract people from paying.

Two small checkout tests worth knowing:

• “Your order is reserved for 10 minutes”: on about 90,000 visitors, revenue per visitor +10%, AOV almost +5%.
• A store that had always shipped free tested a free-shipping threshold instead: AOV $50 → $53, revenue per visitor +13%.

The safer option is the post-purchase upsell: the customer has paid, you show one offer before the thank-you page, and they add it with one click without typing their card again. It works on every Shopify plan. A supplement brand featured on Shopify’s blog reported about 30% of customers took the offer and AOV rose 25%. Most stores see 5–10% from post-purchase offers; the best reach about 20%.

Three mistakes that quietly lower your AOV

1. A free-shipping threshold below your current average order. If most people already spend more, it pushes nobody to add anything; it just gives shipping away.
2. Discount codes everywhere. Pop-ups, influencers and emails all offering 10–15% off shrink every order and teach people to wait for a code.
3. Too many offers at once. A bundle, three cart upsells and two more at checkout. Remember the jam study: one clear offer per step.

Why you can’t skip A/B testing

Every change above can raise your AOV and still lose you money. Say a new bundle lifts AOV by 12%, but the page becomes more confusing and 15% fewer people buy. You now earn less per visitor. On your AOV report it’s a win; in your bank account it’s a loss.

So never judge an AOV test on AOV alone. Judge it on revenue per visitor: total revenue divided by visitors. It already includes conversion rate and order value. If it goes up, it’s a win; if it goes down, it’s a loss, whatever your AOV says.

How to run an AOV test

1. Change one thing at a time. Add a bundle and a new threshold together and you’ll never know which one worked.
2. Split your traffic 50/50, at the same time.
3. Run it long enough: at least two full weeks (weekdays and weekends) and until each version has enough orders. Don’t stop a test the moment it looks good; early winners often end up losers. Don’t start one in the middle of Black Friday.
4. Decide on revenue per visitor. Keep the winner, remove the loser, move to the next test.

This is how you reach 20%: not with one magic change but with three or four good ones stacked. A better shipping threshold (+6%), bundles on product pages (+8%) and a post-purchase offer (+6%) together make about +21% per order.

Where I’d start tomorrow

1. Write down your AOV and revenue per visitor for the last 30 days. That’s your baseline.
2. Look at your product range, where I saw the biggest jump. Create one set, or a better version of your bestseller.
3. Add the AOV column in Meta and compare your ads. Make more of the ones that bring bigger orders.
4. Set a free-shipping threshold 20–30% above your current average order and show it with a progress bar in the cart.
5. Test one bundle on your bestseller, built from what people already buy together.
6. Add one post-purchase offer.

One at a time, each one A/B tested, all judged on revenue per visitor.