We present below the different types of Product Recommendations that can be implemented thanks to the algorithms of Blendee’s artificial intelligence engine.

Trending Recommendations

List of the most popular items in a catalog based on the amount of views and clicks. This analysis is conducted within the last 73 days. The products/content ( items ) that are shown are drawn from a basket that contains not only the most popular but also other items with the aim of helping to highlight less popular products/content in order to give them relevance.

Are the same products always shown?
The algorithm uses a multiplier (x4) in order not to show the same elements over and over again. For example: if you have chosen 5 elements (maximum number of elements to show) Blendee will chooseà 5×4 elements that will therefore allow you not to always show the same elements (which would otherwise be even “more popular”). It is also possible to enable random “elements” so as to give greater visibility even to less popular elements.

Personalized Recommendations

A list of items based on the user’s recent history. This list is based on the user’s recent browsing history and considers sales and views. From this profile, the system proposes elements that other users similar to ours have liked. The function scales towards elements with less similar characteristics, in order to satisfy the number of products/contents requested (obviously within the limit of availability of the catalog).

What happens if it’s the First Time Visit?
The algorithm selects the required set of elements by selecting from the most popular items.

Personalized Trending Recommendations

Mix of the previous two: among the elements to be suggested to the user, more emphasis is given to those that are currently more popular.

Browsing History Recommendations

List of items that the user has seen in the last few days (the data is historicized and not in real time, in case you want to show the items seen in the current session use the Session Remarketing type). If the number of elements seen is low, the function scales to the most popular ones in order to satisfy the number of elements required.

What if you’ve never sailed?
The algorithm selects the required set of items by selecting from the most popular products.

Browsing History Trending

List of products/content that the user has seen in the last few days (the data is historicized and not in real time, in case you want to show the products/content seen in the current session use the Session Remarketing type). In case the products/contents seen are a low number, the function scales to the most popular ones.

What if you’ve never sailed?
The algorithm selects the required set of elements by selecting from the most popular items.

Personalized Recommendations by Browsing History

A list of suggested items based on the experiences of users with a recent browsing history similar to that of the user who is browsing.

What if you’ve never sailed?
The algorithm selects the required set of elements by selecting from the most popular items.

Personalized Recommendations by Sales

A list of suggested items based on the experiences of users with a recent purchase history similar to that of the user who is browsing. Only confirmed purchases are taken into account in order to produce the choice of output products.

What if you have never bought it?
The algorithm selects the required set of elements by selecting from the most popular items.

Shopping Cart Recommendations

List of products to recommend based on the list of products currently in your cart. Driven by the behavior of users who have seen or purchased such products.

What happens if I don’t have any products in my cart at that time?
The platform ignores any recommendations, so the container target area will not be customized.

Remarketing

List of products that the user has seen in the last few days and that he has not bought (the data is historicized and not in real time, in case you want to show the products seen in the current session use the Session Remarketing type). Consider products seen in the last 180 days but not purchased. If the number of these products is less than the required number, the rest of the products will be those that other users similar to our user have liked.

What if you’ve never sailed?
The algorithm selects the required set of items by selecting from the most popular products.

Session Remarketing

A list of items that the user has seen in the current session. If the items seen in the session are less than those requested, this algorithm also scales as in the case of Remarketing.

What if you’ve never sailed?
The algorithm selects the required set of elements by selecting from the most popular items.

Session Browsing History

A list of items that are related to what the user has seen in the current session based on the experiences of other users who have seen the same items. If the items seen in the session are less than those requested, this algorithm also scales as in the case of Remarketing.

What if you’ve never sailed?
The algorithm selects the required set of items by selecting from the most popular products/content.

Frequently Bought Togheter

Only available on product detail pages. A list of products that are often bought along with the product that the current page refers to. In the event that the product is not often purchased with others, the recommendation will not be shown.

What happens if there aren’t enough purchases associated with the current product?
The platform ignores any recommendations, so the container target area will not be customized.