Releva has completed an $870,000 financial round led by New Vision Fund 3 with participation by HR Capital AD, Verto Invest and the investment arm of private investors.
Releva завърши финансов рунд от $870 000, воден от Фонд Ню Вижън 3 с участието на Ейч Ар Капитал АД, Верто Инвест и подкрепата на частни инвеститори.
Search the Documentations

Examples of Recommendations

Examples of Recommendations on Home page:

In this section, we will describe the recommender use cases on the Home Page.

Recommended for You

Recommended for You collects information from the last viewed products and current cart of the user and recommends back products that are frequently bought together and viewed one after another.

Here is the config: 

In this case, Recommended for You will be used for all users, since we configured the recommender with the Everyonesegment. We also boosted it up with products from the user’s favorite category (automatically collected).

Popular products

Popular products recommender is tracking popular products, based on the users’ product views and purchase frequencies. Here again we boosted the user’s favourite category, but this time we set a higher weight for this component.

Below we can see the top 3 products:

This case has the following configuration: 


Promo recommends promotional (discounted) products, returned in the order, defined by the Recommended for you algorithm. We added a filter to explicitly remove products that are not discounted (as an alternative to boost). The configuration is the following:

Examples of Recommendations on Category page

Now we will describe the recommender use cases for the Category page.

New from this category

New from this category recommender brings to the user’s attention new products, ordered by the date of publishing. The configuration is the following:

We are boosting products from the same category, as users may not expect products from the Outdoor category while they are looking at the Indoor category. We also have increased the weight of the category booster. Note that we are boosting products with categories that are the same as the categories of the products on the current page. Thus, we got the same three products ordered in a slightly different way:

You looked at

You looked at recommender brings back the browsing history of the user. We again boosted products which have same categories as the products on the page.

We have looked at two products from the home category, namely Mud Soap and The Field Report and they are returned on top of our list. But since we boosted, we also got Cydney Plaid shirt from a different category that we have looked at just before the soup and the report. The configuration of this case is: 

Recommended for you

This is the same recommender that we described on the Home page. The only difference is that here we are boosting products from the same category. Below are the products we got:

I was in the Men category when I got this recommendation back. Since I boosted by Men category I don’t expect only products from this category. Indeed, have the two bags I bought before and the Boots I’ve looked at today (and bought before).

Examples of Recommendations on Product page:

In what follows we will describe the recommendation use cases on the Product page.

Frequently bought together

Recommends products that regularly co-occur in the same basket. Here we don’t filter by category in order to help users find complementary products. For example, users that look for a jacket might as well choose to buy new boots or a backpack.

For this, we need the following configuration: 

You may also like

Recommends products that other people view and then buy. Again no boost by any category as we want complimentary products from other categories.

And the config: 

People who viewed this also viewed

Recommends products that other users of the shop regularly view one after each other.


For demonstration purposes, we also boosted with high weight, prices above 100 USD. This means that there might still be products with lower prices but we shall expect products with prices above 100 USD to score higher on the list.

For the sake of demonstrating features we also included only products from the men category. Note that we have only three products in the Men category and I got Ayres Chambray that costs less than 100 USD as I boosted by price. If I’ve used include there would be only two recommended products – the Duckworth Jacket and the Ranger boots.

You looked at

This is the same recommender as the one on the Category page. It returns back the browsing history of the user but we haven’t boosted it by any category in this case.

To demonstrate custom fields, e.g., fields that each partner can add to their products, we are showing (including) only products that have a custom field named “tags“ with the value “Shirts“. Here is how to configure:

And here is the full set of configurations for this recommender: The products I got:

Indeed I got only products with tag Shirts.

Examples of Recommendations on Cart page:

In what follow we will describe the recommender use case on the Cart page.

Forget something?

This is our special within-basket recommendation system which generates real-time personalized product recommendations to supplement the user’s current product basket.

An important characteristic of online e-commerce shopping is that it is highly personal. Customers show both regularity in purchase types and purchase frequency, as well as exhibit specific preferences for product characteristics, such as brand affinity for wine or price sensitivity for milk.

Our recommender captures personalization affinities of the user combined with co-occurrence to support alternative products based on similar users behaviours.

We also used a price boost to score higher products that are above 100 USD.

I added to the basket the Moon Cycle shirt and got back two complementary products the Dawson Trolley and Duckworth Jacket based on other user baskets and the Red Ranger Boots that I bought before.

Previous Recommendations
Next Algorithms
Table of Contents