Dynamic contextual variables

Dynamic contextual variables are the real-time inputs that power:

• Personalization (based on viewed, bought, or carted products)

• Dynamic filtering and rule application (brand, category, stock, etc.)

• Algorithm scoring (via recency, frequency, score)

• User segmentation and localization (country, language, device_type)

The table below lists the Dynamic contextual variables:

Variable	Type	Description	Example
page_type	string	Identifies the type of page currently viewed. Used to adapt recommendations or merchandising logic.	"homepage", "category", "product", "cart", "checkout", "search"
viewed_items	array	List of product IDs viewed by the user during the current session or historically.	["SKU123", "SKU456"]
bought_items	array	List of product IDs previously purchased by the user.	["SKU987"]
cart_items	array	List of product IDs currently in the user’s cart.	["SKU111", "SKU222"]
array_item_id	array	Input list of item IDs dynamically passed into a strategy (for example, “items viewed” or “items bought”).	["SKU123", "SKU456"]
referrer	string	URL of the previous page; used to infer user intent or navigation path.	"https://www.google.com"
recency	number	Time elapsed since the user’s last interaction (view, add_to_cart, or purchase). Used to weigh freshness.	24h, 3d
frequency	number	Number of user interactions (views, clicks, purchases) over a given period.	5
monetary	number	Total monetary value associated with the user’s purchases.	250.00
segment	string / array	Dynamic segment assigned to the user (e.g., VIP, returning, new).	["returning_user", "FR"]
device_type	string	Device detected in the current session.	"mobile", "desktop"
country	string	User’s detected country based on IP or site configuration.	"FR"
currency	string	Currency currently displayed on the site.	"EUR"
language	string	Language of the current page or session.	"en", "fr"
session_id	string	Unique identifier for the current browsing session.	"session-abc123"
anonymous_id	string	Persistent anonymous identifier used for non-logged users (cookie, localStorage, or cookieless ID).	"anon-9876"
score	number	Relevance score computed by the recommendation algorithm for a given item.	0.87
rule_id	string	ID of a dynamic filter or business rule applied during recommendation computation.	"filter-brand-exclude-Nike"