How to build a product bundling strategy that moves the AOV needle

A product bundling strategy is the plan for which products you sell together, how you price them, where shoppers see them, and how you'll know they worked. Done well, bundles raise average order value without training customers to wait for a discount. When they go wrong, it's usually quiet: margin slips to people who were going to buy everything anyway, and the warehouse fills up with components that don't match.

Bundling is also one of the few growth tactics that works at almost any size. You don't need a personalization engine or a data science team to put a cleanser, serum, and moisturizer on one product page and price them as a routine. A 40-SKU Shopify store can launch its first bundle this afternoon with a bundles app. A retailer with 20,000 SKUs runs into the same questions at a bigger scale, which is why the steps below start simple and add complexity only where it pays for itself.

Why bundling works and the psychology of the whole picture

Most shoppers are after an outcome, like clear skin, a finished nursery, or a campsite that's ready by dark. A bundle sells that outcome directly, and research on how people make decisions helps explain why it lands.

A bundle turns several decisions into one

Every separate product a shopper has to pick is another chance to stall. Which serum goes with this cleanser? Do I need the travel case? A well-built bundle answers those questions for them, so the choice shrinks from "build a routine" to "yes or no on this routine." That's why bundles tend to do best in categories where shoppers feel unsure, like skincare, coffee gear, pet supplies, and first-time hobby kits.

People average a bundle, so a weak add-on can drag it down

It's tempting to throw in a cheap extra to make a bundle feel bigger. Research on what's called the presenter's paradox suggests that can backfire. In studies by Kimberlee Weaver, Stephen Garcia, and Norbert Schwarz, people judged a luxury cashmere sweater on its own as a more generous gift than the same sweater paired with a $10 gift card, even though the second option cost more. People averaged the value of the items, so the modest gift card pulled the whole package down.

For your bundles, that means every item has to pull its weight. A premium face oil packaged with a flimsy sample-size cotton pad set can read as cheaper than the face oil alone.

Bundled prices change how people use what they bought

Bundling also affects what happens after checkout. In a Journal of Marketing Research study by Dilip Soman and John Gourville, people who bought a four-day ski pass were less likely to ski on the final day than people who bought four single-day tickets. Once the price was bundled, the cost of each day got harder to feel, so skipping one felt cheaper.

Their research looked at ski passes and theater tickets, so treat it as something to test for yourself. It's still worth keeping in mind for consumables. If a three-month supply bundle leads customers to use the product less consistently, your reorder timing and satisfaction can slip even while AOV looks great.

Fixed bundles vs. dynamic, personalized bundles

Most bundle debates come down to personalized bundles vs. fixed bundles, and most stores end up using both at different points in the funnel.

Fixed bundles

A fixed bundle is a set combination at a set price: the starter kit, the holiday gift set, the "everything you need" camping cook set. Fixed bundles are easy to photograph, easy to explain, and easy to forecast because you know exactly which components move together. They're the right first bundle for almost every small brand.

The downside is that one combination can't fit every shopper. The person who already owns your cleanser doesn't want to pay for a second one.

Mix-and-match bundles

Mix-and-match bundles let the shopper choose, usually within rules like "any three candles for $60" or "build a box of six snacks." They solve the "I already have that one" problem and work well for replenishment categories with lots of flavors, scents, or colors. Most Shopify bundle apps support this format, so it's a realistic second step for smaller teams.

Personalized bundles assembled in the conversation

A personalized bundle is put together for one shopper based on what they've said, what they've bought, and what's in stock. AI is what makes this practical for a normal-sized team. An AI shopping assistant can ask a shopper a couple of questions and assemble the set on the spot, so you're no longer guessing which three combinations will cover most of your traffic.

Picture a shopper who types, "I'm going car camping for the first time in October and I have a two-person tent already." A fixed bundle would sell them a tent they don't need. A personalized one skips the tent and puts together a sleeping pad, a cold-weather bag, a headlamp, and a stove sized for two, then explains why each piece is there.

It's also what shoppers expect. McKinsey found that 71% of consumers expect companies to deliver personalized interactions, and 76% get frustrated when that doesn't happen. Recommendations also carry outsized weight when shoppers use them. In a Salesforce analysis of more than 150 million shoppers, visits where someone clicked a recommendation made up 7% of visits but drove 24% of orders and 26% of revenue.

Gladly recommends complementary products and upgrades in the flow of a conversation, draws on the customer's conversation history to keep suggestions relevant, and can add the items to the shopper's cart. It won't add an item that's sold out, which matters when you're pulling components from live inventory. The quality of a personalized bundle still depends on your product data, so if your attributes are thin, start by preparing your product feed for AI agents.

Which one fits your store right now

If you're a small team with under a few hundred SKUs, start with two or three fixed bundles built around your best sellers and add mix-and-match once you see which components shoppers swap. If you already have lots of pre-purchase questions coming through chat, personalized bundling is worth testing sooner, because those conversations are already telling you what each shopper needs.

Bundle pricing models and when to use each

There are three main bundle pricing models, and each one protects margin differently.

Discounted bundles

The bundle costs less than buying each item separately. This is the most common model and the easiest to understand, and it's also the one most likely to give away margin. Shopify's guide to bundling for retail suggests taking 10% to 20% off the subtotal if your average margins are 50% or higher, and 5% to 10% if they're lower.

Use discounted bundles when your goal is getting shoppers to try a second or third product they wouldn't have picked up otherwise.

Value-add bundles

The price stays the same as the hero product, or close to it, but the shopper gets something extra: a refill, a tool, a free accessory, or free shipping. Value-add bundles protect your price point, which helps if you're trying to break a discount habit that's eating into margin.

Remember the presenter's paradox here. The extra item should feel useful by itself. A refill pod people will use beats a branded keychain every time.

Premium bundles

Sometimes the bundle costs more than the parts. Curated gift boxes, limited-edition sets, and anything with special packaging or expert selection can carry a premium, because the shopper is paying for the curation and the unboxing as much as the products. Premium bundles do especially well in Q4 and around gifting moments, when the buyer cares more about looking thoughtful than finding the lowest price.

Run the margin math before you pick a discount

Take a skincare brand with a three-step routine. The cleanser sells for $28, the serum for $42, and the moisturizer for $36. Product costs are $7, $10, and $9, and pick, pack, and ship runs about $8 for a single item and $9 for all three.

• A shopper who buys only the serum brings in $42 minus $10 in product cost and $8 in fulfillment, which leaves $24.

• A shopper who buys all three at full price leaves $106 minus $26 and $9, which is $71.

• The routine bundle at $89, about 16% off, leaves $89 minus $26 and $9, which is $54.

So each bundle sale earns $30 more than a serum-only order, and $17 less than a full-price order of all three. The bundle pays off as long as fewer than about 64% of bundle buyers would have bought all three products anyway. You get that number by dividing the $30 gain by the combined $47 swing. If most of your customers already buy the full routine, a value-add bundle will likely serve you better than a discount.

Choosing which products belong together

The best bundles come from evidence you already have. You don't need special software to find them.

Start with the orders you already have

Export six to twelve months of orders and look for products that show up in the same cart. A spreadsheet pivot table is enough for a small catalog. You're looking for pairs that appear together often but not always, because pairs that always ship together are already selling themselves.

Also look at what customers buy within 30 to 60 days of a first order. Those second purchases often point to a natural starter bundle.

Read the questions shoppers ask before they buy

Your support conversations are some of the best bundling research you have. When shoppers keep asking "does this work with the other one?" or "what else do I need for this?", they're describing a bundle you haven't built yet. Go through a month of pre-purchase chats and tag the product pairs that come up.

Rules for a good pairing

• Each item does its own job. Two cleansers in one routine set feel like filler, and shoppers notice.

• Similar purchase cadence. A cleanser that lasts two months bundled with a moisturizer that lasts six creates awkward reorders.

• One clear hero. Build the bundle around a product people already want, then add items that make it work better.

• Stock you can count on. Don't build a headline bundle around a component with long or unreliable lead times.

If you're debating whether an add-on belongs in a bundle or in a separate recommendation, it helps to know the difference between cross-sells and upsells and pick the one that fits the moment.

Merchandising bundles on PLP, PDP & cart

Where a bundle appears changes what it needs to say. The same set can win on one page and fall flat on another.

Product listing pages

On a PLP, shoppers are scanning. Give bundles their own tile with a clear name like "The morning routine" or "First-time camper kit," show every item in the photo, and put the savings or the extra item right on the card. Many brands add a "Sets & bundles" filter or collection so shoppers who prefer them can find them fast.

Product detail pages

The PDP is where most bundles get made. Show a "complete the routine" or "frequently bought together" module right under the add-to-cart button, and let shoppers add the full set with one click. On a bundle's own PDP, list each component with a short line on what it does, and show the separate total next to the bundle price so the value is obvious.

Cart and checkout

The cart is the place for small, low-risk add-ons. Shipping costs are also where a lot of orders die. Baymard Institute's research found that 40% of US shoppers who abandoned an order during checkout did so because extra costs like shipping, taxes, or fees were too high. A cart message that says "add the travel size for $12 and your order ships free" solves the shopper's problem and raises AOV in the same move.

Seasonal bundle merchandising

Seasonal bundles give you a reason to refresh your merchandising without new products. Plan them on a calendar: holiday gift sets need to be photographed and stocked by early fall, back-to-school kits by July, and summer travel sets by May. Rename and re-photograph an evergreen bundle for the season, keep the components mostly the same, and swap in one limited item so it feels new. When the season ends, break leftover kits back into sellable units or move them into a clearance bundle before they sit.

Inventory & fulfillment considerations for bundles

Bundles look simple on the storefront and get complicated in the warehouse. Sorting out bundle inventory management early keeps you from overselling a set you can't ship.

Kitting vs. bundling

People use these words interchangeably, but they mean different things for your operation. As one fulfillment provider explains it, a bundle can exist entirely in software, while a kit is a new physical unit that's assembled and stored under its own SKU before an order arrives.

Most small and midsize ecommerce brands should start with virtual bundles and only move to kitting once a bundle sells steadily enough to justify pre-assembly, or once the packaging itself is part of the product, as with a gift box.

Bundle inventory management without overselling

The most common bundle failure is selling a set when one component is out of stock. A few habits prevent it:

• Track inventory at the component level and let the bundle's availability come from its scarcest item.

• Decide ahead of time what happens when a component runs out: hide the bundle, swap in a substitute, or show it as unavailable.

• Reserve safety stock for components that appear in several bundles, since one popular set can drain a product that other bundles also need.

• If you sell on marketplaces, check each channel's rules for bundles before you list them.

Plan for partial returns

Returns get messier with bundles because customers often send back one item and keep the rest. The NRF estimates that 19.3% of online sales were returned in 2025, so a bundle strategy without a returns policy is only half a strategy. Decide how you'll refund a single item from a discounted bundle, usually at its share of the bundle price, and put that rule where shoppers will see it before they buy.

Measuring bundle performance beyond just AOV

AOV is the number that gets bundles approved, and it's also the easiest one to fool yourself with. A bundle can raise AOV and still lose money, cannibalize full-price sales, or bring in shoppers who never come back. Track these alongside it.

How to read a bundle test

Run a new bundle against a holdout if your traffic allows it, even if that's just showing it to half your visitors for four weeks. Compare contribution margin per visitor alongside AOV, and check whether the hero product's standalone sales dropped. For smaller stores that can't split traffic, compare the four weeks before launch with the four weeks after and adjust for seasonality as best you can.

Then look at what happens next. Customers who start with a bundle often behave differently from single-item buyers over the following months, and those metrics that predict customer lifetime value will tell you more about a bundle's real worth than one order ever will.

Where to start this week

Pick one hero product. Pull the items that most often land in the same cart with it, build one fixed bundle, and run the margin math before you set the price. Put it on the hero's PDP and in the cart, track attach rate and contribution margin for a month, and use what shoppers ask in chat to decide what your second bundle should be.