Did you know that 70% of Amazon.com is devoted to product recommendations?
When such a big, successful e-commerce company focuses that much of its resources on something it must be important.
So what is it?
Similarly to an in-store sales advisor asking you about your preferences and then showing you items accordingly, in e-commerce a software does this automatically. This is one of the core activities in website personalization. This feature is called personalized product recommendation, which displays suggested items that are specifically relevant to each individual shopper, according to algorithms and data collected.
How does it work?
Barilliance’s personalized product recommendations engine combines three factors.
1. The intent of the customer
The software uses data collected about the visitor’s recent activity, overall shopping behaviors, past purchases and browsing history.
2. The wisdom of the crowd
It uses aggregated data from the site that makes correlations between people who bought similar items and the individual shopper.
3. Customized set of variables
Lastly, you can set some categories you would like to include, such as top sellers, items with a good click-through-rate and manually selected items.
Where is it displayed?
Product recommendations on-site can be displayed on any page. The most common ones are the home, category, product detail and cart pages. It is also effective to use it on out-of-stock pages.
It is also possible to have more than one type of product recommendation on one page.
When is it generated?
The recommendations are generated according to changes of the shoppers’ preferences and items’ availability, in order to be accurate.
The product recommendations are generated and changed according to real-time shopper engagement.
For example, if a shopper has sorted a category from low to high, it is likely they are price-sensitive. Therefore, from then on, the product recommendations will present only cheaper items. Additionally, once the shopper has selected a size, the product recommendations will only include items that are in-stock for that size.
2. In emails
The product recommendations are generated when the customer opens the email, rather than when it is sent.
What is displayed?
There are many types of product recommendations that suggest different types of product sets.
New, anonymous shoppers the system has no data on will see common generic recommendations, which include new arrivals, best sellers, top rated and sales.
Returning customers and customers who have engaged with the site for some time and the system has collected data, will see personalized product recommendations. The common ones include similar items (upselling), complementary items (cross-selling) and recently viewed.
Similar items recommendations help customers to find items that fit more of their demands and desired, and by this increase the likelihood of a purchase. Complementary items help customers to find different items of interest and as a result, increase their average order value.
These can be titled in different ways that affect the shoppers’ perception. For example, “Customers who viewed this product also viewed…” and “inspired by your shopping cart”. Involving other shoppers in the title provides social proof to shoppers, which encourages purchases.
Besides images and names of the items themselves, you can also display other information in the recommendations, such as price, colors, and sizes available, social proof notifications and more.
Why is it useful?
Product recommendations create a “shortcut” to items that are specifically relevant to the individual shopper. They transfer the right message to the right shopper at the right time.
Not finding desired items and spending too much time and effort browsing through options can cause a poor customer experience and browse or cart abandonment.
As a helpful tool, product recommendations are a quick and easy way to finding items and optimizing the product discovery process, enhancing and focusing the shopper’s choice, increasing engagement and loyalty. They also trigger impulse purchases and can induce a higher average order value.
Consequently, the feature induces higher revenues and profit margin. Customers who click on product recommendations have a 5.5 times higher conversion rate than those who do not. On-site, personalized product recommendations account for 12% of total revenues. In remarketing emails they can lead to 30% increase in sales conversion rates and 35% in click-through rates.