How Amazon set the standard for product recommendations
Amazon’s product recommendation feature is the gold standard in eCommerce and continues to be the leader in the market to this day. Product recommendations can boost average order value by 50%, increase revenues by 300%, and improve conversion. In addition to being a powerful tool for increasing revenues, product recommendations are so essential that customers now expect to see similar features on all other eCommerce sites.
A product recommendation feature provides a way to better equip customers to make a purchase. Customers who make informed purchases are often seeking relevant information surrounding their purchase. 79% of online shoppers spend time 50% of their time researching prior to making a purchase. Whether it’s items that other people have historically purchased together, or just similar products at varying price points, these options help to educate customers and empower them to make the purchase. And the more relevant the recommendation is, the more a customer will develop loyalty and trust to the company and brand. That’s how Amazon succeeds in their vision to be the “everything store.” The recommendation feature highlights products that the customer wants at varying price points, as well as products that the customer might not have realized would be a great companion to their intended purchase.
Not all of Amazon’s personalizations are created equal, which is what makes Amazon’s approach unique. Some of the recommendations are generated directly from the customer’s browsing and purchase history, which are featured by a “Recommendations for You” tag.
Others are less directly personalized but still relevant. Often, Amazon includes a “Frequently Bought Together” recommendation, which shows the customer the price for adding a highly relevant product to the purchase. Otherwise, Amazon also features product recommendations based on the related browsing history of other Amazon users.These last two are significant in that they are, in a way, not directly related to the customer. It could even be considered a bit risky: what if the items other people bought are not actually relevant?
From personal experience, they’re almost always relevant – and here’s why. Amazon doesn’t take into consideration browsing history only to generate recommendations; they take in the attributes of the customer based on the data they have. This is apparent in the “Recommended for You” section – if the customer is a male in his twenties who looked at bicycles, they are more likely to recommend a mountain bike than a beach cruiser or tandem bike. When providing that same customer recommendations based on “What Other Customers Viewed,” they take those attributes into consideration. Customers see a filter of recommendations based on other shoppers who share the same attributes – in this case, males in their twenties who like bikes – which are more likely to be accurate, even if not based on your personal shopping history.
Perhaps one of the best things Amazon does to personalize recommendations is introduce new and similar products in follow-up emails. This additional detail enhances the shopping experience, wraps up the personalization nicely, and gives the customer a reason to come back. Being able to discover and scroll through an unending selection of products is great, but putting it all together in an email targeted to the specific customer is powerful.
While not every company has the capacity to create a product recommendation engine like Amazon, there’s a lot to learn from the first-in-class platform. Almost every leading retailer has some kind of product recommendation or discoverability feature. Lacking similar features may severely reduce the appearance of credibility and cause customers to look elsewhere. However, creating a product recommendation section on a webpage might also backfire. If there aren’t enough products to develop relevant recommendations, it can become too distracting for the customer.
Amazon’s product recommendation engine revealed that customers want to know what’s available to them, and are ready to purchase if directed to the right products. Empowering customers to make the right purchase can make a huge impact in sales. Whether it’s updating the recommendation engine, adding a great series of emails, or making discovering products more customer-friendly, there’s a lot retailers can learn from Amazon.
Want to learn more about data-driven marketing? Download our guide!
ReSci is a team of marketers and data scientists on a mission to democratize AI. We make powerful recommendations and predictions accessible to brands. Find out how we can help you connect with your customers.
