Recommendations

As a part of managing your customers' Product discovery experience, with Bloomreach Recommendations, you can display recommended products powered by our AI-driven algorithms. Recommendations allow your customers to explore and discover more products resulting in more engagement, ATC, and conversions.

You can configure and manage the Recommendation widgets directly from the Bloomreach Dashboard without the need to make complex technical API-level changes. Using simple Dashboard operations, you can manage widgets end-to-end and use API calls only to override the default configuration if required.  

You can power your pages with easy merchandising options listed below: 

  • Configuration and Management: Recommendation widgets can be configured and managed end-to-end via the Bloomreach Dashboard without using the APIs. However, if you wish to override the default configuration, you must send an API request. 
  • Switching context: The widget configuration is kept lean so you can switch from one type of use case to another as and when required. For example, use the Frequently Viewed Together and Frequently Bought Together widgets interchangeably by making a simple modification via the Dashboard. 

Recommendations Widget Types


Recommendation widgets are classified into four family types — Global, Item, Personalization, and Visual. The API requirements are distinct for each widget type.

Refer to the list below to learn about each widget type family and the component algorithms:

  1. Global Widgets
    These widgets fetch the most popular product recommendations for a given time period or products that generate the most revenue.
AlgorithmsDescription
BestsellerDisplays products that are likely to generate the most revenue.
Trending ProductsDisplays products based on the most popular or trending products for a given time period.

Refer to this article for API requirements.

  1. Item Widgets
    These widgets use the unique identifier of the item (item_ids) on your site to show relevant recommendations.
AlgorithmsDescription
Frequently bought togetherDisplays other products that are commonly bought together with a given product.
Frequently viewed togetherDisplays other products that are commonly viewed along with a given product.
Similar ProductsDisplays other products that are similar to a given product.

Refer to this article for API requirements.

  1. Personalization widgets

These widgets use real-time user data from user_id (or _br_uid_2) to fetch personalized recommendations.

AlgorithmsDescription
Experience-Driven RecommendationsDisplays highly targeted product recommendations by creating real-time user affinity profiles and understanding user intent by leveraging Bloomreach Search understanding and site browsing behavior.
Past PurchasesDisplays previously purchased products to a logged-in customer to enable quick add-to-cart (ATC) for repeat purchases.
Recently Viewed ProductsDisplay personalized product recommendations based on the user's past purchasing history, browsing behavior, and customer preferences.

Refer to this article for API requirements.

  1. Visual widget

This widget uses the image input to show relevant recommendations.

AlgorithmsDescription
Visual SearchDisplays visually similar shoppable recommendations once the customer uploads an image.

Refer to this article for API requirements.

NOTES:

  1. Pathways widgets differ from Recommendation widgets. Pathways family includes Keyword-based and Category-based widgets.
  2. Global, Item, Pathways, and Personalization widget types have their respective Email Recommendations versions.

Where Can Recommendations Add Value?


Recommendations is a customized, unified solution that allows you to add widgets to the pages on your site as per your business needs. Some of them are listed below:

Bestseller Products

You can promote the sale of products that are most commonly purchased by your customers. The algorithm uses a combination of RPV, ATCs, and Views in order to maximize revenue. Below is an example of a recommendation widget used for Bestseller.

1156

Experience-Driven Recommendations

The Experience-Driven Recommendations widget enables you to personalize the shopping experiences for your customers by leveraging the automatically generated real-time user profiles to deliver highly targeted product recommendations.

In addition to curating the recommendations based on historical data, the Experience-Driven Recommendation engine delivers a unique and optimal digital experience in real-time by rapid data aggregation and analysis of ongoing shopper's site visit.

It matches the visitor’s search and browsing activities with the attributes of the products they interact with, leveraging the data to build a unique affinity profile for each individual in real time and recommend products that have the highest chances of conversions for the shoppers. 

Each user’s affinity profile is constantly refreshed with every interaction, providing unique insights into their preferences for specific colors, brands, sizes, categories, price ranges, or any such attribute present in the product.

Frequently Bought Together

You can show products that are bought along with a specific product by adding a Frequently Bought Together widget on the Product page. Below is an example of a recommendation widget used for Frequently Bought Together. Visit this article to learn more.

Frequently Viewed Together

You can build a recommendation experience for your users on your product details page to show products that are viewed along with the selected product by other customers. Below is an example of a recommendation widget used for Frequently Viewed Together. Visit this article to learn more.

1920

Past Purchases

The best way to allow your logged-in customers to repeat a purchase is by using the Past Purchases widgets that display products based on past purchase data. Below is an example of a recommendation widget used for Past Purchases.

Recently Viewed Products

Recently Viewed Products Widget allows you to display personalized product recommendations based on the user's past purchasing history, browsing behavior, and customer preferences. This user-based widget enables you to re-engage and re-target shoppers who are actively browsing and searching for products on your site. Visit this article to learn more.

Similar Products

You can increase the visibility of other products that are similar using the Similar Products widget. Given below is an example of a recommendation widget used for Similar Products. Visit this article to learn more.

1489

Trending Products Recommendations

In order to improve the shopping experience, and increase end-user engagement and product conversions, you can improve product discovery (outside search) by recommending popular products at a given time. This widget can be displayed on the Home page or globally. An example is given below.

1920

❗️

A newly added algorithm to be displayed is cached in the client's browser local storage for 24 hours. If an algorithm is enabled, the user will either need to use incognito mode or clean their browser history for the last 24 hours. Else, the algorithm will be visible the next day (after 24 hours when the local storage will expire).

📘

How to Create these Experiences

To create the above experience, follow these steps:

  1. Integrate Recommendations and Pathways: To learn how to integrate Recommendations and Pathways widgets, go to the Integration Overview page. 

  2. Configure widgets: To learn how to create and manage widgets, go to Widget Configurator page. 

  3. Merchandise Widgets as per business goals: To learn how to set merchandising rules, go to Merchandising Capabilities.