Walk past any store in the world and you’ll see the same tactics in effect: window displays showing the brand’s most attractive, best-selling products. Entering the store will see smaller displays like mannequins, collections, or recommended setups.
These are traditional methods of recommending products to shoppers – of showing potential customers what is currently in fashion to form a connection and suggesting what they might want to buy.
While eCommerce doesn’t have glass windows, well-executed product recommendations can offer even more to digital brands. In this article, we’ll look at what product recommendations are and the best practices for common site pages including Product, Cart, Search and Home, Blog, 404 and Collection pages.
What are Product Recommendations and why are they powerful?
Product recommendations are exactly as described: a select list of products presented to the customer, physically or digitally. If you are using an AI platform like LimeSpot, those online recommendations are based on analysis of what your shoppers are most likely to engage with and purchase.
In today’s marketing landscape, personalized recommendations have quickly become the gold standard. That’s no surprise, given that in-house marketers personalizing web experiences using product recommendations, report an average 19% uplift in sales.
Furthermore, customers expect brands to create customized shopping experiences:
- 78%-87% of customers want personalized content, and
- 85%-92% cite it as a major influence on whether to buy
Figure 1 presents three common examples of displaying product recommendations on Product pages:
Figure 1a- “Shop the look / Related items” (BonnieChic)
Figure 1b – “Best Sellers” (Over the Ocean)
Figure 1c – “Frequently bought together” (Osbornee)
Taking Product Recommendations to the Next Level
Product recommendations can have an incredible effect on a store’s sales. However, most brands only scratch the surface of what’s possible.
Most companies manually curate their recommendations or just don’t use attribution and targeting information. While these can drive a small lift, they aren’t taking full advantage of the benefits of personalization and their strategy doesn’t take into account users needs and intents.
For example, they pull products together based on interrelation – recommending a baseball if someone buys a bat – sell them as a package, and call it a day. Not a bad technique, but very predictable.
This strategy does not focus on the user’s needs or desires, rather, it hopes that this combination of products corresponds with the activities and needs of a very general cohort of shoppers. It may have some impact, but it’s not enough.
Great product recommendations – ones that not only increase your Average Order Value (AOV) but also persuade indecisive shoppers to click the buy button – require a much deeper level of personalization. To get to that level, you need to understand both connections between your user’s behavior, needs, and desires – and the connections between your products. Make sure you have both data components in any platform you consider. Figure 2 illustrates how this process works with LimeSpot’s AI.
Figure 2: How LimeSpot Personalization Combines Product Attributes with Customer Attributes
Recommendations based on this approach connects customer intent and interests with the best- fit product from your catalog to showcase products that are ideal for the user. It’s the difference between having a general window display and introducing a personal shopper to each of your visitors. And it’s no surprise which one helps your store make more sales.
However, it’s difficult to know how best to implement product recommendations. 83% of marketers say that creating personalized content is their biggest challenge, so you’ll certainly want to use something automated vs. manual curation. You’ve only got so many hours in a day to optimize your store, and the best recommendations will come through AI.
What about placement? Should you put them on the home page? Attempt to upsell to people already looking at a product? Add something to your checkout experience?
Below we take you through the best practices from more than 85 million impressions we’ve collected, and share how to maximize sales gains with personalized recommendations.
While product recommendations are critical, they can also be a huge time investment for your team. And without AI, you won’t get optimal results. It doesn’t have to be that way. Try out LimeSpot’s easy publishing and sophisticated AI recommendations and start reaping the benefits today.
Best Practices for Product Recommendation
The first step is identifying the best placement for your recommendation boxes. Figure 3 below shows the % of people purchasing through product recommendations on each page. This data is based on an analysis of 85,711,116 unique visitors on online stores using LimeSpot personalization.
Figure 3: Conversion Rates from LimeSpot Product Recommendations on Different Pages
It’s not surprising to see that the product page itself sees the highest conversion rate, of 29.1%. But what is also interesting is that the home page sees a strong conversion rate, of 6%.
Here are the best practices on the top 7 locations on your site.
1. Product Page
As seen in Figure 3, 29 out of every 100 unique visitors who have been on Product pages ended up engaging with a LimeSpot recommended product and making a purchase, so let’s start with the highest performer.
Figure 4: Example of Recommendation Boxes on the Product Page
Drawing from our 85m+ unique visitors, we have found that the boxes shown in Figure 5 drive the most sales on Product pages.
The best performing box on your Product page is the “Frequently bought together”. It requires purchase data from your store in order to figure out what products to recommend. If your store is relatively new, there might not be enough information for the AI engine to recommend products that are frequently bought together.
“Related products” is your second best choice. It showcases the products with similar attributes to the item that is currently being viewed or purchased.
“Trending products” sits in the third place as it capitalizes on quick changes in the Internet’s landscape by showing which of your items are going viral. This is similar to the Most Popular items but it is dependent on the store’s most recent traffic.
Figure 5: Best types of Recommendation on the Product Page
2. Cart Page
The Cart page is the perfect location for a cross-sell or upsell. At this point, the user is ready to purchase, with only one step left before checking out. It’s because purchase intent is high that this page’s product recommendation boxes achieve a conversion rate of over 9%. Figure 6 shows how they might look:
Figure 6: Example of Recommendation Boxes on the Cart Page
Product recommendations on the Cart page are much more sensitive than the other pages. Poor recommendations can easily break the sales process and lead to abandoned carts. However, if you offer the right product to the right person on the Cart page it would be quite rewarding as it is the best place for upselling and cross-sells.
Using AI and machine learning algorithms is the only way to have real-time recommendations on the Cart page, personalized for each shopper based on their browsing and purchase history, intent and the items in their Cart. If you’re thinking of implementing recommendations, Figure 7 below shows our top 3 boxes.
Figure 7: Best Types of Recommendations on the Cart Page
3. Home Page
For direct traffic, the Home page is the first page seen. Making the best use of this prime real estate is key to increasing your sales. The Home page works as the window display of your store, so by adding the “Trending Products” recommendation box to your Home page, you are keeping your window display automatically updated by showcasing your most attractive products, always.
To create a better personalized experience for your window display you should add a “You May Like” recommendation box to your Home page as well. It shows your customers products based on their actions and behavior history on your site. As a combination and cross-section of related items for the recently viewed products, this recommendation box dramatically improves the discovery process for your online shoppers. Figure 8 shows the product recommendations that perform best on the Home Page.
Figure 8: Best Types of Recommendations on the Home Page
4. Search Page
A big portion of online purchases begins with a simple search on Search Page. Having “Trending products” recommendation box in the search page provides you an excellent opportunity to showcase your most attractive products to your shoppers as they go through the search process.
The search process also reveals customer intent and highlights the products that the customer is highly interested in. They’re actively searching for something and if you can show them what they’re after you have a higher chance that they will buy. That said, the “You May Like” recommendation box performs almost as good as “Trending products” in Search pages.
If you’re thinking of implementing product recommendation on your search page, you might want to start with our data which indicates the top-performing recommendation boxes, as shown in Figure 9.
Figure 9: Best Types of Recommendations on the Search Page
5. Blog Pages
If you’re on top of your SEO and content marketing game you should see a good number of people stopping by your blog. With that high level of traffic, your Blog pages are the perfect place to put a product recommendation. Figure 10 shows the best product recommendation boxes for your Blog pages.
Figure 10: Best Type of Recommendations on the Blog Page
6. 404 Page
404 pages are often a forgotten element of your site. But as you saw in Figure 3 above, our analysis shows that there’s an average click-through of 3.1% on recommendations for on 404 pages. If you’re using something like LimeSpot, adding those boxes is as easy as placing the box and letting the system do the rest. Figure 11 shows how they might look.
Figure 11: Example of Recommendation Boxes on the 404 Page
Figure 12: Best Type of Recommendations on the 404 Page
7. Collection Page
Collection pages are great because they’ve already part-segmented your audience. If someone’s looking at your summer collection then you know they want summer clothes. It makes your job a little easier because the user is already more likely to buy products within that collection.
Adding some product recommendation boxes onto Collection pages is a no-brainer. Figure 13 shows the recommendation options that provide the best return.
Figure 13: Best Type of Recommendations on the Collection Page
These best practices give you an idea of the areas of your site where recommendations are most likely to drive more sales. That said, each website is unique. To find out how personalized recommendations perform for your site we recommend running ongoing A/B tests to maximize the impact.
If you’d like to get started experimenting with different product recommendations on your store, LimeSpot can provide you both the recommendations and the A/B infrastructure to quickly measure how the potential sales you are leaving on the table.
To get started, sign up for a free LimeSpot account today.