The base user item recommender serves as the foundational model for various other recommendation modules. This core model focuses on analyzing user interactions with items to generate predictions about which products or content a user may find appealing. It utilizes data such as product views, cart activity, and past purchases to create a comprehensive understanding of user preferences.
Inspire's other recommender modules build upon this base model by leveraging its predictions and tailoring the recommendations to specific contexts and use cases.
This includes the following modules:
Key Requirements
Before setting up the Base User Item Recommender, ensure that:
-
The ENTERPRISE subscription is active.
-
User engagement data (product views, add-to-cart actions, and purchases) is available via web tracking.
-
A product feed is configured with product IDs that match the product IDs in your transaction/engagement data. This is essential for the model to correctly link user interactions to products.
-
If your catalog contains product variants (e.g., different sizes or colors of the same product), ensure these are identifiable via a variant ID, parent ID, or similar identifier so the model can correctly group variant-level interactions.
-
Sufficient engagement history is available for meaningful collaborative filtering (the model needs enough data to identify patterns between users and products).
How it Works
The recommender uses a technique called collaborative filtering. In simple terms:
-
It looks at what consumers with similar behaviour have viewed or purchased.
-
If people who bought product A also often buy product B, the model learns this connection.
-
Your consumers then get personalized suggestions based on both their own behaviour and the behaviour of others with similar interests.
Engagement Scoring
The model supports multiple ways of scoring user interactions with products:
-
Count — Each interaction (view, add-to-cart, purchase) is counted equally, regardless of when it happened.
-
Time-discounted — More recent interactions are weighted higher than older ones. This ensures the model adapts to changing customer preferences over time.
-
Seasonal — Interactions are weighted based on seasonal patterns, accounting for recurring trends (e.g., holiday shopping behavior).
The engagement scorer determines how the model interprets user behaviour, and can be configured to best fit the business context.
You can further shape the results with business rules in the Inspire cockpit, by enabling one of the above mentioned modules.
Limitations to Consider
-
Popularity bias: Popular products may be recommended more frequently than niche or new items.
-
Cold start: New products and new consumers may initially lack enough data for strong recommendations.
-
Data dependency: Recommendations depend on the accuracy and completeness of consumer engagement data.