On most PrestaShop 8 stores we audit, the built-in cross-sell block runs in decorative mode: 4 random products from the same category as the items already in the cart, with no weighting, no learning, no analytics. Nobody knows whether it converts, and the honest answer is: barely at all. Yet on a store doing €10,000 in monthly revenue, a genuinely data-driven cross-sell with an 8% click-through rate and 5% conversion adds several hundred euros of average order value every month. Over 12 months, cross-sell can represent the equivalent of a new acquisition channel — without any advertising spend.
This guide details the 7 cross-sell strategies that actually work on PrestaShop 8 in 2026, how to weight them so they work together, and how to measure what genuinely contributes to average order value versus what simply takes up space on the cart page.
Why the native PrestaShop 8 cross-sell is no longer enough
PrestaShop exposes a displayCrossSellingShoppingCart hook that renders the native block at the bottom of the cart. The code is straightforward: it takes the categories of products already in the cart, selects 4 random products from the same category (excluding those already added), and displays them.
This design made sense ten years ago, when buyer expectations were lower and competition for average order value was less intense. In 2026, that is no longer the case. Stores that genuinely push their average order value upward combine several recommendation logics, weight results according to what converts on their specific audience, and continuously learn from observed behaviour.
The limitations of the native block today are:
- no memory of product pairs that have actually been co-purchased;
- no notion of accessory beyond raw co-presence in the same category;
- no measurement data: there is no way to know the click-through or conversion rate the block produces;
- no bundle logic (offering a discount when a customer buys several products together);
- no segmentation by manufacturer, price range, popularity or catalogue freshness.
The result is predictable: a block that occupies 200 to 400 pixels of height on the cart page with no measurable contribution to revenue.
How much a smart cross-sell can actually generate
Let us do the maths. A PrestaShop store with €10,000 in monthly revenue, an average order value of €50 (200 orders per month) and 5,000 unique monthly visitors.
Today, the native cross-sell converts at around 0.5% (a low but realistic estimate). Across 200 orders, that represents perhaps 1 additional order per month — effectively nothing.
With a multi-strategy data-driven cross-sell achieving:
- a click-through rate (CTR) of 8% on displayed products;
- an add-to-cart rate of 25% among those who clicked;
- a completion rate of 60% among those who added to cart.
The calculation gives: 200 orders × 8% × 25% × 60% = 2.4 additional orders per month per cross-sold product. With 4 products displayed on average, and even dividing by a cautious factor to avoid double-counting, that comes to 5–10 additional orders per month. At €50 average order value, that is €250 to €500 in additional monthly revenue from cross-sell alone.
But the most powerful lever is not only the additional orders: it is the effect on average order value itself. When a successful cross-sell pushes the customer to add a €15 complementary product to a €50 basket, average order value rises to €65. Across 200 orders per month, that €15 delta represents €3,000 in additional revenue — the equivalent of a junior salesperson’s full salary.
Of course, these figures are not guaranteed. Cross-sell does not work with the same intensity across all sectors (highly technical B2B has less basket effect than mass-market B2C). But the order of magnitude is consistent with what we observe on stores that have seriously implemented the mechanism.
The 7 cross-sell strategies that work in 2026
No single strategy is universally winning. What works is combining several weighted logics and measuring which one genuinely converts on your audience. Here are the seven that complement each other intelligently.
1. Native PrestaShop accessories
PrestaShop exposes an explicit relationship between a product and its accessories, defined in the back office on the product page. This relationship is valuable because it is editorial: you decide which products are logically associated (an HDMI cable for a television, a case for a phone, an ink cartridge for a printer).
This is the most reliable strategy because relevance is guaranteed by your manual work. The only drawback: it assumes your accessories are up to date in the back office, which is not always the case on large catalogues.
Recommendation: high weight (8–10 out of 10) if your accessories are well populated; moderate weight (3–5) otherwise, while you work on filling them in.
2. Frequently bought together (order learning)
This is the most powerful strategy in 2026 because it learns from your real data. The principle: with each validated order, all product pairs present in the order are recorded in a dedicated table with a frequency counter. The more often a pair recurs, the higher it ranks.
Major advantage: you discover associations you would never have thought to code editorially. On a cosmetics store, the algorithm may detect that customers who buy serum X also regularly buy cream Y — not because they are logically linked, but because it is a real purchasing behaviour.
Constraint: a minimum order volume is needed for the index to be meaningful. On a new store with 50 orders per month, wait 6 to 12 months before the index produces genuinely statistical recommendations. On a mature store with 1,000 orders per month, it is usable within 30 days.
Our DataFirefly Cross-Sell module calculates this index automatically with each validated order and exposes a minimum threshold (3 occurrences by default) to avoid statistical noise.
Recommendation: weight 9–10 as soon as you have sufficient volume.
3. Same category
The obvious one: if someone has a product from the “shirts” category in their cart, suggesting other shirts is a basic logic. This is what the native PrestaShop cross-sell does, but often indiscriminately.
Limitations: activate with a moderate weight (4–7), because some use cases make it counterproductive. If a customer has already put 3 shirts in their cart, suggesting 4 more similar ones risks cannibalising their choices instead of growing the basket. Always pair the strategy with an exclusion filter for products already in the cart — something the native block does not handle consistently well.
Recommendation: weight 6–7, lower on stores where the typical cart contains several items from the same category (fashion, accessories).
4. Same manufacturer
The brand loyalty lever. If a customer has an Apple product in their cart, suggesting other Apple products has two effects: it recognises an existing preference (the customer likes this brand) and exploits ecosystem coherence (products from the same brand are often compatible with each other).
Particularly effective in: consumer electronics, branded cosmetics, branded clothing, estate wines, author books. Less effective on unbranded products or generic private labels.
Recommendation: weight 5–7 depending on the strength of the brands in your catalogue.
5. Period bestsellers
Passive social proof. Suggesting last month’s bestsellers captures two signals: these products convert well (so statistically more likely to convert on a new customer too), and they are typically perceived as “safe bets” by buyers.
Use sparingly: if you put bestsellers in every cross-sell block, you risk cannibalising the long tail of the catalogue. Prefer a moderate weight so they appear as a complement to the more contextual strategies, not as a replacement. A “Bestseller” badge on the product page itself is an effective complement: it signals the status visually without depending solely on the carousel.
Recommendation: weight 4–6 by default, increase for new visitors who have no history yet.
6. New products
The “you haven’t seen this yet” effect. New catalogue arrivals capture the attention of repeat buyers who already know your core range. It is also an indirect SEO mechanism: pushing new products into visibility accelerates their indexing and discovery.
Particularly useful for stores that frequently refresh their catalogue (fashion, home décor, food). Less relevant for slow-rotation catalogues (premium furniture, white goods).
Recommendation: weight 3–5 in general, higher (6–7) on seasonal fashion and home décor stores.
7. Similar price range
Budget coherence. If a customer has added a €45 product to their cart, suggesting a €350 product will probably lose the sale. Suggesting a product in the €30–60 range (±30% for example) stays within their mental budget and increases the probability of an add-to-cart.
This strategy works better as a cross-cutting filter than as a standalone strategy: rather than proposing “other products in the same price bracket”, filter the recommendations from other strategies to exclude those that fall outside the budget range.
Recommendation: weight 3–4, or use as a filter rather than a primary strategy.
Combining 7 strategies without losing control: weighting and cumulative scores
The trap when you have 7 strategies is stacking them without logic. The right design is a cumulative scoring system where each candidate product receives a score equal to the sum of the weights of the strategies in which it appears.
In practice: run the engine, fire the 7 queries, aggregate the results with their weights. A product that appears as an accessory (weight 10), as frequently bought together (weight 9), and as same category (weight 7) scores 26. A product that appears only as a bestseller (weight 5) scores 5. Sort by descending score, keep the top 4 or 6, display.
The advantage of this approach: it is explainable and debuggable. If you see an unusual product in first position in the carousel, you can trace why (“it came up as both an accessory AND a bestseller, so score 15”). This is very different from an opaque AI system where recommendations are produced by a machine learning model with no explainability — often sold at a high price and impossible to explain to merchants when they ask “why that product?”.
To set the weights, start from a proven base:
- Accessories: 10
- Frequently bought together: 9
- Same category: 7
- Same manufacturer: 6
- Bestsellers: 5
- New products: 4
- Price range: 3 (or as a filter)
And adjust based on what your analytics show you, which brings us to the next point.
Measuring what actually works
Without measurement, you are flying blind. The minimum analytics to put in place on your cross-sell:
- Impressions: every time a product is displayed in the carousel, record the event. This is the basis for all other metrics.
- Clicks: every click on the product card. Gives CTR (clicks divided by impressions).
- Add-to-cart: every add to cart from the carousel (not from the product page, which is counted separately). Gives the add-to-cart rate.
- Purchases: the recommended product made it into a validated order. Gives the final conversion rate.
Over 30 days, these aggregated metrics give you a global view. But the real gain comes from the breakdown by strategy. If the “bestsellers” strategy shows a 5% CTR and the “frequently bought together” strategy shows a 12% CTR on your store, you have the data to increase the weight of one and reduce the other. Without this granularity, you are adjusting by instinct.
Signals to watch:
- CTR below 3% on a strategy: it is contributing nothing — lower its weight or disable it.
- Add-to-cart rate above 30%: the strategy is relevant — increase its weight.
- Final conversion rate (orders divided by impressions) above 1%: that is an excellent ratio.
The DataFirefly Cross-Sell module includes these per-strategy analytics natively in the admin dashboard — no need to dig the data out of GA4 with poorly configured custom events.
Beyond the carousel: bundles, wishlists and side carts to amplify the effect
The cross-sell carousel is the entry point of the subject, but it is not enough on its own. Three complementary levers significantly amplify the effect on average order value.
Frequently bought together bundles. Alongside the classic carousel, a separate block that proposes a bundle with a discount when the customer has a product in their cart that appears in frequent pairs (minimum threshold: 3 occurrences). This is Amazon’s “Frequently bought together” mechanic, which combines recommendation and price incentive. A 5–10% discount on the bundle is generally enough to trigger a grouped purchase without cannibalising margin.
The wishlist as a deferred lever. Not all visitors are ready to buy immediately. A well-integrated wishlist captures the products they are interested in, lets them come back later, and — importantly — can serve as a basis for targeted follow-up emails (price alert, stock alert). A visitor who has wishlisted a product has expressed a strong intent; retargeting these visitors converts better than generic retargeting.
The side cart. When a customer clicks “Add to cart” from a product page or a cross-sell carousel, there are two options: redirect them to the cart page (flow interruption), or display a side panel that slides in from the right with the added product plus complementary suggestions (embedded cross-sell). The side cart keeps the customer in the browsing context instead of disconnecting them from the category they were exploring. Our DataFirefly SideCart module implements this mechanic with integrated cross-sell.
Together, these form a coherent system: cross-sell carousel on the cart page, side cart for adds from elsewhere, bundle for frequent pairs, wishlist for deferred purchases. The more levers work together, the stronger the cumulative effect on average order value.
Conclusion: cross-sell is an investment, not decoration
The native cross-sell in PrestaShop exists because a minimal feature was needed. The cross-sell that genuinely contributes to revenue is something else: a data-driven system where several weighted strategies work together, where every displayed product is measured, and where bundles and wishlists amplify the carousel effect.
On a store doing €10,000 in monthly revenue, turning cross-sell from “decoration” to “average order value engine” represents several hundred euros of additional revenue every month. Over 12 months, that is the equivalent of a paid acquisition channel — without the advertising costs.
To go further on related conversion levers, browse our Conversion & UX category, or our PrestaShop tutorials for the technical side. And if you are ready to move from decorative cross-sell to one that drives average order value, the DataFirefly Cross-Sell module natively implements all 7 strategies, per-strategy analytics and frequently bought together bundles — 3-minute install, proven default configuration, unencrypted source code.