Frequently Bought Together on Shopify: How to Show Product Pairings Customers Actually Add
A "frequently bought together" block is one of the oldest ideas in e-commerce, and one of the easiest to get wrong. Set up frequently bought together on Shopify with the wrong data, in the wrong spot, or with the wrong products, and shoppers scroll past it. Set it up well and it quietly adds a second item to orders that would otherwise ship with one.
This guide covers where pairs come from, what to do when a product has no sales history, where to place the block, how many products to show, and the mistakes that make shoppers ignore it. If you are still deciding whether a slot should round out the basket or trade the shopper up, start with cross-sell vs upsell as adjacent reading.
What Does "Frequently Bought Together" Actually Mean?
Frequently bought together (FBT) recommends products that real customers have purchased in the same order as the item a shopper is looking at. It is a cross-sell: the suggestion sits next to the product, it does not replace it.
The key word is bought. FBT is not "similar products" and it is not "you might also like". A similar-products block shows alternatives (another pair of running shoes). An FBT block shows companions (the socks, the insoles, the waterproofing spray). Mixing those two jobs in one slot is a common reason FBT underperforms.
How Are Product Pairings Derived?
There are three ways to decide which products belong together, and strong setups usually combine them.
1. Order co-occurrence
Count how often product A and product B appear in the same past order, and rank B by that count for A. Better versions correct for popularity, because your bestseller appears in lots of orders simply because it sells a lot, not because it belongs with everything.
Strength: it reflects what customers actually do, including pairings you would never guess. Weakness: it needs volume. A product with a handful of orders produces noisy pairs.
2. Manual rules
You decide: "when the espresso machine is in the cart, show the tamper, the milk jug and the descaler." This is the right call for compatibility, new launches, and pairings you want to push for margin.
Strength: full control, works from day one. Weakness: it does not scale to a large catalog, and it goes stale the moment your range changes.
3. AI recommendations
Machine learning models learn product affinities from your orders and catalog, then generalize them, so they can rank companions even for pairs that rarely appeared together. For a deeper comparison of this approach against hand-picked pairings, see how AI-powered upsells outperform manual product recommendations.
Strength: scales across the whole catalog and keeps up as buying patterns shift. Weakness: like co-occurrence, it learns from history, so it still needs some order data to be confident.
Can You Do Frequently Bought Together on Shopify Without an App?
Partly. Shopify gives you two native building blocks:
- Related products. Many themes include a product recommendations section powered by Shopify's own algorithm. According to Shopify's developer documentation, it considers products historically purchased together, products with similar descriptions, and products from related collections. Because it blends "bought together" with "similar", it often behaves more like "you might also like" than true FBT.
- Complementary products. In Shopify's free Search & Discovery app, you can hand-pick up to 10 complementary products per product. Compatible themes show them as a block on the product page.
Both have limits worth knowing. Shopify's documentation notes that out-of-stock products, products priced at 0, gift cards and items already in the visitor's cart are excluded from recommendations. But the native blocks live on the product page, complementary products are fully manual, and neither puts pairings in the cart drawer or at checkout. For a small catalog that can be enough.
The Cold Start Problem: New Stores and New Products
Every data-driven method has the same blind spot: a product with no orders has no pairs. That hits new stores and new launches alike. A practical cold start plan:
- Hand-pick pairings for launches. For the first few weeks, a manual rule beats any algorithm.
- Pair new products with proven companions. Point a new item at accessories that already sell well with similar products. That borrows the demand you already have.
- Use bestsellers as the safety net. When there is no signal, popular products beat an empty slot.
- Hand over to data once orders arrive. Revisit manual rules after a month or two and let the data win where it disagrees.
Where to Show Frequently Bought Together Products
The same pairing converts differently depending on where the shopper is and what they are thinking about.
- Product page. The shopper is still deciding. Show items that make the main product usable or better, below the add-to-cart button so they support the decision rather than compete with it.
- Cart drawer. The shopper has committed. This is the natural home for low-friction companions added in one tap, and FBT here can respond to everything in the cart, not just one product.
- Checkout. The shopper is paying. Only one or two small, obvious add-ons belong here.
Different placements usually deserve different pairing logic. The guide to different product recommendations for cart, product page and checkout walks through that in detail.
How Many Frequently Bought Together Products Should You Show?
Show three to four on the product page and in the cart, and one or two at checkout. Shopify's own theme guidance suggests limiting recommendations to four products per product page, so only the most relevant ones get attention.
More options do not mean more adds. Past a handful, shoppers start skimming, and your strongest pairing gets the same attention as your weakest.
Common Frequently Bought Together Mistakes
Recommending items already in the cart
Showing a shopper the exact product they just added looks broken. Any FBT block should filter out products already in the cart. The exception is consumables, where "add another" is a real offer, and that should be a deliberate choice, not an accident.
Recommending sold-out items
A sold-out pairing is a dead end that wastes a slot. A product can also be in stock overall but sold out in the shopper's size, so check availability per variant.
Near-duplicates instead of companions
If a shopper is looking at a navy linen shirt and your FBT block shows three other linen shirts, that is a similar-products block wearing the wrong label. It asks the shopper to reconsider, not to add. Pairings should usually come from a different product type than the item they are paired with.
Set it and forget it
Catalogs and seasons change, and pairings drift. Review your top products' pairings at least quarterly, and keep most suggestions cheaper than the main item unless you are deliberately upselling.
Frequently Bought Together With EliteCart
EliteCart covers all three pairing methods and shows them in the cart drawer, the two-step cart, the product page (on plans that include product-page upsells) and checkout modules.
- Manual upsell flows handle launches and compatibility. Under Upsells → Manual upsells, you trigger a flow on specific products, variants or collections and pick up to 12 upsell products. Products already in the cart are hidden by default, and sold-out or unavailable products are filtered out automatically.
- EliteAI™ Ultra Original is the frequently bought together engine. It is trained on your store's order history and recommends the products customers most often buy together with the item in the cart.
- EliteAI™ Ultra CrossCategoryBoost recommends complements from other product types than the cart item, which avoids the near-duplicate problem by design.
- Bestsellers can be chosen as a source for any location, which is a useful safety net while a store builds order history.
Under Upsells → AI upsells → Customize by display location, you choose which of these each location uses, so the cart can run frequently bought together while the product page runs something else. If you want tighter control, Fine-tune EliteAI™ Ultra lets you build your own versions with filters and boosts, for example a "different product type than the trigger item" filter plus a boost toward cheaper products for classic add-ons. Setup details are in the Understanding EliteAI™ and Fine-Tuning EliteAI™ Ultra help articles.
FAQ
Does Shopify have a built-in frequently bought together feature?
Not as a dedicated block. Shopify offers related products (algorithmic, mixing bought-together and similar items) and complementary products (hand-picked in the Search & Discovery app), both on the product page.
How much order history do I need for data-driven pairings?
There is no hard minimum, but pairings get more reliable as orders accumulate. Until then, use manual pairings for important products and bestsellers as a fallback.
Is frequently bought together the same as cross-selling?
It is one type of cross-selling, specifically the kind based on what customers have actually purchased together.
Start with your top five products. Check what each one is paired with today, remove anything sold out, already in the cart or basically the same product, and make sure the cart shows pairings too, not just the product page.