Web modules: Recommendation Strategies

This article provides a comprehensive overview of all web modules currently available in the platform. Each module plays a specific role in powering personalized recommendations, merchandising strategies, and user experiences across your website. The documentation includes detailed descriptions, technical names, and key functionalities for each module. For your convenience, a Glossary is included at the end of this article to explain technical terms and internal jargon.

🔍 Note: The modules are grouped by goal. To choose strategies by goal and see where to place them on your website, see Recommendation strategies: the right products for every visitor.

Item modules
Module	Rerank	Candidate set	Schema rules	Backfill	Boost	Audience check	OData	FullInfo
Inspire new visitors
GetPopularItemsWeb	✓	Trending and Popular	Category and Brand		✓	✓	✓	✓
GetPopularItemsInCategoryWeb	✓	Popular Items			✓	✓	✓	✓
GetPopularItemsInBrandWeb	✓	Top in Brand			✓	✓	✓	✓
GetPopularCategoriesWeb	✓	Popular and Trending						✓
GetPopularBrandsWeb	✓	Popular and Trending						✓
Make it personal
GetUserItemRecommendationsWeb		User History		Favorite Category & Brand	✓	✓	✓	✓
GetUserLookAlikeItemsWeb		User History, Similar Items		Favorite > Popular, Trend	✓	✓	✓	✓
GetUserBrandHistoryWeb		User History						✓
Help them decide
GetSimilarItemsWeb	✓	Similar Content	Category and Brand	Top Viewed	✓	✓	✓	✓
GetUserItemHistoryWeb		User History					✓	✓
Grow the basket
GetPIMRelatedItemsWeb	✓	Products Bought Together	Target Group, Category, Brand, etc.	Trending (visit)	✓	✓	✓	✓
GetPIMRelatedItemsForBasketWeb	✓	Products Bought Together	Target Group, Category, Brand, etc.	Trending (visit)	✓	✓	✓	✓
GetUserCrossSellingItemsWeb		User History		Favorite > Popular, Trend	✓	✓	✓	✓
Promote campaigns
GetMerchandisingItemsWeb	✓	Merchandising Products	Popularity & Discount			✓		✓
GetNumOfUsersRightNowWeb		Get Number of Users

🔍 Note: All strategies also support the Content-, Product-, Category- and Brand-filter and split testing, except GetNumOfUsersRightNowWeb.

Content modules
Module	Rerank	Candidate set	Schema rules	Backfill	Boost	Audience check	OData
Promote campaigns
GetMerchandisingContentWeb	✓	Merchandising Products	Popularity			✓
Turn content into sales
GetItemsBasedOnContentWeb	✓	Content to Products	Category and Brand	Top Viewed	✓	✓	✓
GetContentBasedOnItemWeb	✓	Product to Content			✓	✓	✓
GetContentBasedOnProductCategoryWeb	✓	Product-Category to Content		Category to Content	✓	✓	✓
GetContentBasedOnProductBrandWeb	✓	Brand to Content			✓	✓	✓
Keep readers engaged
GetSimilarContentWeb	✓	Similar Content	Category	Top Viewed	✓	✓	✓
GetPopularContentWeb*		Trending and Popular			✓	✓	✓
GetUserContentRecommendationsWeb		Similar Content		Top in Favorite Content	✓	✓	✓
GetUserContentHistoryWeb		Recent History					✓

* Trending: Reflects the change in popularity over the selected time period, indicating whether a product has become more popular.

🔍 Note: All 'content' strategies listed above also support the Content-, Product-, Category- and Brand-filter, Fullinfo and split testing.

Glossary

This glossary provides definitions for technical terms used in this article.

Backfilling Definition: Products to be added at the end of the prioritized list of products when personalized recommendations aren't available.

Candidate set Definition: Defines the logic of the module - how the product selection is made, which data sources and filters are applied, and which attributes influence the final candidate list shown to the user. In Merchandising, this list (typically 50 products) is dynamic and behavior-based, forming the limited set within which boosting actions can take place.

Content- / Product- / Category- / Brand-filter Definition: Use this filter to remove items from the output based on a comma-separated list.

Explicit Schema Rules Definition: Module specific filtering. The rules are set up via a customized schema.

FullInfo Definition: All mapped parameters from the item Schema are included in the API output. This enriches the recommendation data with additional item attributes.

Integration to Merchandising and CDP (Audience Check) Definition: Connect the Recommendation Engine with the CDP.

Merchandising Boost Definition: Boosting products with a given value from the product Schema via Merchandising Boosted Campaigns.

ODatafilter Definition: Use this filter to remove groups of products from the output. Can be used on all mapped parameters from the product schema. Read more.

Personalized Rerank Opportunities Definition: Reranking the products based on User & Cookie; buy, basket and visit + Items similar or related to the user interactions. Read more.

Split testing Definition: Split testing (A/B testing) compares two setups (A and B) by assigning users based on their cookie id to see which performs better.

Next step

Ready to adjust a strategy? See Module adjustment and tuning for how to change the settings.