How to Sell More with Product Recommendations
Miguel Nicolás
02/06/2026
-
Do you want to know how to sell more with product recommendations in your eCommerce?
-
-
Did you know that Amazon estimates that its recommendation system accounts for 35% of its revenue?Of course, I do know that none of us are Amazon (yet), but it is easy to see the real potential of something that you can also apply in your own online store.Well, this is exactly what the post you are about to read is about. We are going to tell you how product recommendations work, how to integrate them into your business strategy and, as always, we are going to do it with some highly illustrative examples. It sounds interesting, right?Let’s get started.
-
What Exactly Are Product Recommendations?
-
Think about when you enter a physical store and the salesperson assisting you says: “If you take those waterproof Trail running shoes, some technical socks would suit you very well, as they are breathable and prevent your feet from ‘cooking’.” That is a traditional recommendation, but in eCommerce it works the same way, although in an automated and, even better, scalable way.Product recommendations are, basically, suggesting other items to your customer that may interest them based on what they are viewing, purchasing, or their purchase history.Although it sounds simple, if you systematize it and do it properly, you will see that it is a very powerful conversion tool.There are 3 main types of recommendation-based strategies:
- Upselling: you offer the customer a better version of the product they are viewing, but at a higher cost. For example: if someone is looking at a basic food processor, you show them the model with more power and more accessories.
- Cross-selling: in this case, you suggest complementary products to one they are purchasing (or have purchased before). The customer adds a camera to the cart and you propose a memory card, a case, or a tripod.
- Related products: similar items by category or theme. Because if someone is looking at primary school textbooks, they will probably also need a pencil case, notebooks, or a backpack.
-
Where Should I Place Product Recommendations in My eCommerce?
-
Well, where else? On the product page! Yes, of course, but not only there. This is a fairly common mistake and reality shows that, if you only recommend products there, you are missing strategic opportunities.Let’s look at some possibilities and placements:
- On the product page. Just because it is obvious does not make it any less effective. Below the description or on the side, a section such as “You may also like” or “Complete your look” (in fashion stores) works very well. In fact, although recommendations generate only 7% of an eCommerce site’s traffic, they are responsible for 26% of its revenue, according to data from Clerk.io. A children’s clothing store, to continue the example, can display the trousers that match the sweater the customer is viewing.
- In the cart. This is the moment of highest purchase intent. The customer has already decided to buy something, so they are receptive. Cross-selling fits perfectly here: “Customers who bought this also bought...”. An online stationery store could suggest a pen set when the customer has a planner in the cart.
- During the checkout process. Be careful, because this is somewhat delicate, as you do not want to distract the customer or lengthen the process. However, a small low-cost product suggestion (an accessory, an add-on) can work as a “Would you like to add anything else before paying?”. Perfect for encouraging impulse purchases.
- In transactional and post-purchase emails. Once the customer has received their order, you have a golden opportunity. They already trust you. An email sent one week later with products related to what they purchased has open and conversion rates far higher than generic newsletters.
-
The Logic Behind a Good Recommendation
-
Recommending anything without criteria puts the strategy at risk. You need to think carefully about this strategy. How do I select recommendations? That is a good question, the first of several you should ask yourself before configuring your recommendations:Does it make sense to group these products together? If someone buys cat food, do not recommend dog collars. Consistency is essential.Is the price compatible? In upselling, too large a price jump can scare customers away. If the base product costs €30, proposing one that costs €300 may not make much sense.
-
PRO TIP: The unwritten rule is that the upsell should not exceed between 25% and 40% of the original price (and even 40% seems high to me, but it depends on your niche or segment).
-
Does this recommendation add real value? Always think from the customer’s perspective. It is not only about context, but also about knowing your target audience, because some products may appear to be related, but in practice they are not.
-
A Real Example Step by Step
-
Imagine you own an online stationery and office supplies store. A customer visits the page of a 2026 weekly planner. What can you do?On the same page, you display as an upsell the leather-cover version of that same planner, which costs €8 more and includes additional features. As a cross-sell, you add a block called “Complete your desk kit” showing permanent ink pens, colored markers, and a bamboo pen holder.The customer adds the planner to the cart. In the cart, a suggestion appears: “You may also be interested in this sticker set to organize your planner.” Price: €4.95. An impulse purchase is almost guaranteed in many cases.One week after receiving the order, you send an email: “Are you enjoying your planner? These notebooks match it perfectly.” And you attach three options with direct links.This type of flow, when properly executed, can increase the average cart value by between 10% and 30%, according to data from Forrester Research. You extend customer lifetime value, recurrence, and profitability within your current customer base, even without acquiring a single new customer.
-
Mistakes You Should Avoid
-
The first and most common one is recommending too many products. If you display 20 options, the customer becomes overwhelmed and chooses none of them. Three or four well-selected products work much better than an endless list.You should also avoid recommending what interests you rather than what interests the customer. Recommendations are not intended to help move unsold inventory (there are other tactics for that). Recommendations in an eCommerce store are meant to improve the customer experience. If you do the opposite, customers notice and it creates distrust.Another very common mistake is not paying attention to data. Over time, your own sales provide extremely valuable information: which products are purchased together and which accessories are bought with which items. Use it.
-
Where to Start Step by Step
-
If you do not yet have an active recommendation system, do not worry; let’s create a realistic and easy-to-implement action plan:1.- Review your 10 best-selling products and manually configure between 3 and 5 related products for each one. This alone will cover most of your traffic.
2.- Activate or create a post-purchase follow-up email with personalized recommendations according to the category purchased by the customer. This simple automation is a fairly refined adjustment.
3.- Once you have data on which products are bought together, adjust your recommendations so that they become increasingly accurate. Optimize continuously in an ongoing cycle to achieve the best possible results.Product recommendations are not a trend, and they certainly are not a feature reserved for large companies.They are one of the simplest and most effective sales levers that exist. And best of all, they improve your customers’ experience while increasing conversion.Everyone wins. -
Do you already use product recommendations in your store? Share in the comments what is working best for your business.