Product recommendations

Product recommendations are the blocks that suggest other items: "you may also like" on a product page, "frequently bought together" near the add to cart button, "complete the look" in the cart and "picked for you" on the home page. Done well, they help a shopper find the right product faster or add something they would have wanted anyway. Done badly, they fill the page with unrelated items and push the main product further down.

The tools here choose those products for you. Most combine several methods: items often bought together, items viewed by people who viewed this one, items similar in attributes, and products picked from the shopper's own history. Merchandisers can add rules on top, such as hiding out-of-stock items, excluding certain brands or always pairing a product with its accessories.

Placement matters as much as the algorithm. Recommendations on the product page help shoppers compare and choose. Recommendations in the cart and after checkout suit add-ons and refills. Home page and email recommendations help returning customers. Start with one or two placements where the reason for the suggestion is clear, and add more only when you can see they help.

Many apps in this category also handle bundles, upsells and post-purchase offers, where a shopper can add an item to an order they have already placed. These can add order value, but watch for offers that feel pushy or slow down checkout.

Featured AI conversion tools: product recommendations

SponsoredProduct recommendations This position is open Put your tool at the top of the product recommendations page. Paid spots are always labeled Sponsored. From $99 a month. Get featured

What a featured listing costs ยท from $99 a month

6 AI conversion tools in product recommendations

Nosto

Commerce experience platform for product recommendations, personalized category pages, search and user content.

  • Works with Shopify
  • Works with BigCommerce
  • Developer API

Rebuy

Shopify personalization app for upsells, cross-sells, smart carts and post-purchase offers.

  • Works with Shopify

LimeSpot

Product recommendation and bundling app for Shopify stores, with personalized blocks on product, cart and home pages.

  • Works with Shopify

Clerk.io

Ecommerce personalization for recommendations, search, email and audience segments, aimed at mid-size stores.

  • Works with Shopify
  • Works with WooCommerce

Recombee

Recommendation engine as a service, with an API for product, content and search recommendations.

  • Free plan or free tier
  • Developer API

Barilliance

Personalization tool for product recommendations, cart abandonment emails and on-site messages.

When you compare tools, ask how they decide what to show for new products or new visitors with no history, since that is where simple algorithms fall back to showing best sellers. Ask how they count the sales they credit to themselves, because a click on a recommendation followed by a purchase days later may or may not have been caused by it. A holdout group, where some visitors see no recommendations, is the clearest test.

Keep an eye on what the blocks actually show. Algorithms can get stuck on a handful of best sellers that appear everywhere, or suggest a product the shopper already bought. Browse your own store as a new visitor and as a returning customer every few weeks, and adjust the rules when the suggestions stop making sense to a person.

Check page speed and layout on mobile as well. A recommendation carousel that loads late can push content around as the page renders, which frustrates shoppers and can hurt search rankings.

Questions about product recommendations

Where should I put product recommendations?

Start with the product page and the cart, where the reason for a suggestion is clearest: alternatives on the product page, add-ons in the cart. Add home page and email placements once you can measure that the first ones help.

How do recommendation tools handle new products?

New products have little behavior data, so many tools fall back to attribute similarity or best sellers. Ask each vendor how they handle new items and new visitors, and check that your product attributes are complete enough to match on.

Can I trust the revenue a recommendation app reports?

Treat it with care. Apps credit themselves with orders using their own rules, and other tools may claim the same sale. A holdout test, where some visitors see no recommendations, gives a clearer answer.

Are post-purchase upsells a good idea?

They can add order value with little risk to the first sale, since the order is already placed. Keep offers relevant, make declining easy, and check that the step doesn't slow down or confuse checkout.

Related categories