Most Shopify merchants who install a product recommendation app see their average order value (AOV) nudge up slightly, then flatline. The problem is not the app. The problem is that "you might also like" widgets showing random products from the same collection add almost no value. Customers scroll past them because they have learned to associate those sections with irrelevant suggestions.

Cross-selling works when the recommendation is genuinely useful: the product complements what the customer is already buying, it is shown at a moment when the customer is open to adding to their order, and it is framed in a way that makes the value obvious. Done correctly, cross-selling increases AOV by 10 to 30%, which on most Shopify stores is the difference between a profitable business and a marginal one.

Cross-Sell vs Upsell: What Is the Difference and Why Does It Matter?

These terms are frequently conflated. The distinction matters because the tactics, placement, and copy differ.

An upsell is an offer to buy a more expensive or premium version of the product the customer is considering. On a product page for a 64GB phone case, an upsell suggests the same case for the 256GB model. In a checkout context, an upsell might offer an extended warranty or a premium tier.

A cross-sell is an offer to buy a related product that complements the item already in consideration. On the same phone case product page, a cross-sell suggests a screen protector, a charging cable, or a phone stand.

The strategic distinction: upsells require the customer to reconsider their existing choice (higher friction). Cross-sells add to it (lower friction). Cross-sells work better at the cart stage and post-purchase because the primary decision is already made.

Where Do Cross-Sells Work?

Not every location is equal. Placement determines both visibility and receptivity.

Product pages are the earliest point to show cross-sells. A "frequently bought together" section or "complete the look" module below the product description works well for complementary categories: apparel + accessories, electronics + peripherals, skincare + complementary products. The risk here is distraction: a cross-sell shown too prominently before the primary add-to-cart can pull attention away from the conversion you are trying to achieve. Keep cross-sells below the fold on product pages.

Cart pages and cart drawers are the highest-converting location for cross-sells. The customer has already committed to a purchase. They are in buying mode. A short list of complementary products with one-click add-to-cart functionality at this stage typically converts at 3 to 8% of cart sessions. Keep the selection tight: two to three products maximum. More than three creates decision paralysis.

Post-purchase pages (the thank-you page and order confirmation screen) are underused for cross-selling. The customer has just completed a purchase, their card details are already saved (on Shopify Payments), and their guard is down. A post-purchase upsell offering a complementary product at a discount, with one-click purchase enabled, can achieve 5 to 15% acceptance rates. This is higher than most pre-purchase cross-sell placements.

Cross-Sell Location	Typical Conversion Rate	Best For
Product page (below fold)	1-3%	Complementary categories
Cart drawer / cart page	3-8%	Low-cost add-ons
Post-purchase offer	5-15%	Discounted repeat buy
Order confirmation email	2-5%	Long-consideration products

Manual vs Algorithmic Recommendations

For stores with small catalogues (under 100 products), manual recommendations are often more accurate than algorithmic ones. You know which products go together. Setting explicit "pair with" relationships using product tags or metafields, and surfacing these in your theme, gives you precise control.

For larger catalogues, manual curation becomes impractical. Algorithmic recommendations use purchase history data ("customers who bought X also bought Y") to surface relevant products. The accuracy of algorithmic recommendations improves with transaction volume: a store processing fewer than 200 orders per month may not have enough data for meaningful algorithms, and manual curation will outperform automated suggestions.

A hybrid approach works well for most mid-size stores: manually curate cross-sell relationships for your top 20 products (which typically represent 60 to 80% of revenue), and let algorithms handle the long tail.

Which Apps Drive the Best Cross-Sell Results?

Three apps are worth considering, each suited to different store sizes and needs.

LimeSpot Personalizer (free plan available, paid from $18/month) is one of the most complete recommendation engines on the Shopify App Store. It uses purchase and browsing data to build personalised recommendation widgets that can be placed on product pages, collection pages, cart pages, and the thank-you page. Its strength is ease of setup and visual editor for customising widget appearance without code.

Rebuy Personalisation Engine (from $99/month) is the premium option for stores with serious cross-sell ambitions. Rebuy powers smart cart functionality (a slide-out cart with built-in recommendations), post-purchase offers with one-click checkout, and personalised email recommendations. Its data model is more sophisticated than LimeSpot's, and it integrates tightly with Klaviyo and Attentive for post-purchase flows. The price is justified only for stores generating enough volume to see returns: typically £500,000 per year or above.

Also Bought Product Recommendations (from $9.99/month) is a focused, single-purpose app. It does one thing well: displays "customers who bought this also bought" widgets on product pages, powered by actual purchase data from your store. It is more affordable than LimeSpot for smaller stores that only need product page cross-sells and do not require cart drawer or post-purchase functionality.

Using Product Tags and Collections to Power Recommendations

If you want to avoid or delay adding another app, you can power cross-sells using Shopify's native product metafields or tag-based logic.

The approach: add a metafield to each product (for example, custom.cross_sell_products) that stores a list of product handles for related items. Display those products in a "Goes well with" section on the product page using your theme's metafield display functionality.

This requires a small amount of theme customisation but gives you complete manual control over recommendations without app overhead. It works well for stores with fewer than 200 products where every cross-sell relationship is intentional.

For collection-based cross-sells, use Shopify's "Related products" theme section (available in Dawn and most OS 2.0 themes), which surfaces products from the same collection. This is better than nothing but less powerful than purchase-data-driven recommendations.

Cross-Sell Copy That Does Not Feel Pushy

The framing of cross-sell recommendations determines whether customers perceive them as helpful or salesy. The best-performing cross-sell copy is specific and utility-focused.

Avoid:

"You might also like" (too vague, customers ignore it)
"Don't forget these!" (pressure-based, off-putting)
"Complete your purchase" (transactional, no value statement)

Use:

"Customers also buy these together" (social proof, implies it is a known combination)
"Goes well with [Product Name]" (specific, relational)
"Frequently bought with this" (purchase data-backed, trustworthy)
"Complete the look" (for apparel, gives a reason to buy multiple items)

The more specific the copy, the better it performs. If you can name the relationship (for example, "Most customers add a screen protector with this case"), you are giving a reason to consider the cross-sell rather than just presenting an item.

Measuring Cross-Sell Revenue Contribution

Most cross-sell apps include revenue attribution reporting. The key metrics to track:

Cross-sell conversion rate: percentage of sessions where a cross-sell widget is viewed that result in the additional product being added to cart
Revenue attributed to cross-sells: how much additional revenue per month is generated through cross-sell interactions
AOV lift: average order value for orders that include a cross-sell item vs orders that do not

Review these monthly. If your cross-sell conversion rate is below 2% on the cart page, the recommendations are not relevant enough. Check whether the algorithm has enough data to work with, and consider adding manual overrides for your top products.

Common Cross-Selling Mistakes

Too many recommendations. Showing 8 products in a "you might also like" widget is not 4 times better than showing 2. It creates decision paralysis and trains customers to skip the section entirely. Keep it to 2 to 4 products.

Recommending the same category, not a complementary one. Showing three more t-shirts on a t-shirt product page is not a cross-sell, it is a navigation aid. A genuine cross-sell offers something different that makes the primary purchase more valuable.

Cross-selling out-of-stock items. There is nothing more frustrating than being tempted by a recommendation that is not available. Filter out-of-stock products from all recommendation widgets.

Ignoring the thank-you page. The post-purchase page is consistently the highest-converting cross-sell placement and the least used. If you are not showing an offer here, you are leaving revenue on the table.

Key Actions to Take Now

Audit your current cross-sell setup. Identify where recommendations currently appear, what is being shown, and whether those products are genuinely complementary or just popular items from the same category.
Install LimeSpot (free plan) or Also Bought and configure product page and cart page recommendation widgets. Start with your top 10 selling products.
Set up a post-purchase offer on your thank-you page. Offer a discounted complementary product. Test a 10 to 15% discount to incentivise immediate add-on purchase.
Review your recommendation copy. Replace generic labels with specific relational language like "Goes well with" or "Customers also buy these together".
Filter out-of-stock products from all recommendation widgets. Check this setting in your app's configuration.
After 30 days, review cross-sell conversion rate and AOV lift data. If performance is below 2% on cart, check product relevance and reduce the number of recommendations shown.

Frequently Asked Questions

How many cross-sell products should I show at once? Two to four is the optimal range. Below two, customers may not find a relevant option. Above four, decision fatigue reduces click-through. For cart page cross-sells specifically, two to three tends to perform best.

Should I discount cross-sell products to encourage add-on purchases? Discounting on product pages and cart pages can train customers to expect discounts on add-ons, which erodes margin over time. On post-purchase offers, a time-limited discount (for example, "Add this for 15% off in the next 10 minutes") is more justifiable because the customer has already paid and the offer does not affect the primary purchase economics.

Do cross-sells work for B2B Shopify stores? Yes, and often more effectively than for B2C. Business buyers who replenish consumables or buy equipment benefit from "frequently ordered together" suggestions because they reduce search time. Manual curation works particularly well in B2B contexts where you understand the purchase workflows of your customer types.

How long does it take for algorithmic recommendations to become accurate? Most recommendation engines need 100 to 200 orders on a product before purchase-based recommendations are reliable. For new products or low-volume SKUs, fall back on manual tagging or collection-based recommendations until enough data exists.

Cross-Selling on Shopify: How to Show the Right Products at the Right Time