Product Recommendations That Actually Lift AOV (and the Ones That Don't)

Related is not complementary

Shopify's free Search and Discovery app sets two kinds of recommendations, and the difference between them is the whole game. Related products are similar items, generated automatically, and a shopper who clicks one usually buys it instead of the product they were looking at. Complementary products are bought in addition, like the case for the phone or the refill for the bottle, and you choose them yourself, up to ten per product. For AOV, complementary is the one that matters, because related swaps a purchase and complementary adds one.

Shopify suggests complementary items that cost the same as or less than the product they sit on, which is the right instinct: the add-on should feel like a small decision next to the big one. Pick the pairs from your orders, not from hunches. The Items bought together report, under Analytics then Reports, lists the combinations your customers actually buy, and the pairs at the top of that report are the recommendations that will work.

Placement beats the algorithm

Where the recommendation sits matters more than how clever it is. A frequently bought together module on the cart page or in a slide-out cart consistently outperforms the same module on the product page, because the shopper on the product page is still deciding whether to buy at all, and the shopper in the cart has already decided. Customers also bought outperforms you might also like for the same reason: the first is evidence from other buyers, and the second is a guess.

That is also why the free shipping threshold is usually a stronger lever than any recommendation widget. Set the threshold 15 to 20 percent above current AOV, show progress toward it prominently, and the recommendations sell themselves, because the shopper is now actively looking for something to add. The free shipping threshold math is the starting point, and the recommendations are what fill the gap it creates. A threshold with nothing to add toward it is a wall. A threshold with three relevant add-ons in the cart is a nudge.

The test that tells you if it works

Test one change at a time and track profit per order, not just higher AOV, because discounts and incentives can raise the average while cutting the margin. Run the complementary block in the cart drawer for two weeks, with add-ons priced at roughly 10 to 20 percent of AOV, and compare revenue per session against the two weeks before. Accessories and refills work better than unrelated best sellers, and a bundle works when it solves a clear use case, like matching sets or consumables, not when the products feel randomly grouped.

If the test shows nothing, the problem is usually upstream. Recommendations work best after the basics are fixed: clear product information, variant and compatibility clarity, and easy navigation. The category page audit is where that work starts, because a shopper who cannot find the right product will not buy the add-on either. Fix the page, then tune the recommendation. The other order wastes the test.

Frequently asked questions

Do product recommendations increase AOV?

They can, but results depend on execution. Complementary products in the cart or cart drawer, chosen from real order data, lift AOV. Generic related products on the product page usually show little or no lift, because they swap one purchase for another.

What is the difference between related and complementary products?

Related products are similar items, generated automatically, and a shopper usually buys one instead of the original. Complementary products are bought in addition, like a case for a phone, and you choose them yourself in Shopify's Search and Discovery app.

Where should recommendations go?

In the cart or the slide-out cart drawer. A frequently bought together module there consistently outperforms the same module on the product page, because the shopper in the cart has already decided to buy.

How do I pick which products to recommend?

From your own orders, not hunches. The Items bought together report in Shopify Analytics lists the combinations customers actually buy. Start with the pairs at the top of that report.