Marketing Glossary

Recommendation Engine

What is a Recommendation engine?

A recommendation engine is a technology that matches what customers are browsing with products or content tailored to their interests.

Recommendation engines rely on carefully considering a person’s browsing habits to offer them what they want. This not only improves general usability but also allows you to optimize your results live.

What are Product Recommendation Engines?

A product recommendation engine is a widget on an eCommerce site that recommends products to visitors.

When there are product recommendations, it is an either-or situation: either the same products are recommended for all site visitors or they recommend a specific product for each visitor to the site.

Increasing engagement with their global audience is a key task for marketers, and there are two different strategies they can use to do this.

The first ranking is always based on the product’s previous performance. This means that to recommend purchases categorically, the current popularity rating or trend of all available products is estimated and considered.

Another strategy includes suggesting contextual recommendations. These are a bit different. Contextual recommendations are suggestions based not on the consumer’s previous purchases, but rather their browsing needs at that time (i.e. a category they’re looking in). For example, such recommendations are “similar products in the category”, “bought together”, and “most popular among other buyers”.

What are Content Recommendation Engines?

Content recommendation engines are product suggestion engine that focuses on content hosted on a site. They work by attempting to engage readers, get viewers into new content and deeper engagement.

An affinity profile is generated by digital interaction and activity. If you browse a lot of pop music pages on your online browsing, the site can recommend more articles that are similar to what you’ve been looking at in the past.

Example with Amazon’s engine

Amazon’s recommendation engine has endless use cases and is effective at tailoring its recommendations for different kinds of customers.

Amazon’s product recommendations are based on what users engaged with and how often they purchased that product. Their engine can easily track user behaviour by collecting important data points, that can later be stored and analyzed to identify consumer patterns over time. These recommendations are individually tailored but also depend somewhat on the information collected about other consumers in an effort to predict what their customers may want next.

For known users, the experience is personalized and employs a form of content-based filtering to recommend products based on your previous interests.

Navigate between the letters, to explore the glossary terms