Product Recommendations

Product Recommendations model can be used for building upsells, cross-sells, and personalized recommendations for your customers. You can use the model to recommend products based on the user's behavior, preferences, and other factors.

The Product Recommendations model utilizes multiple data points to generate relevant product offers:

• previous actions of the user: product views, purchases, etc.
• previous actions of other users
• products previously purchased together
• cart contents and selected product
• user profile data (when available)

Getting product recommendations

1. Create a product recommendations model in the Almeta Cloud console to get a model ID.
2. Query a web tag calculation or API endpoint to get the recommendations (see query format below).

Query format

_almeta.t('calculate', {
    model_id: MODEL_ID, // replace with your model ID
    data: {
        number_of_recommendations: 12,
        recommendation_filters: {
            exclude_collections: [],
            exclude_products: [ 'H71660', 'H71413' ],
        },
        cart: [
            { product_id: 'H71140', quantity: 1, price: 27.00 },
            { product_id: 'H72226', quantity: 1, price: 27.00 }
        ],
        selected_product: {
            product_id: 'H72226', quantity: 1, price: 27.00
        },
    }
});

Data Point	Description
model_id (required)	Unique identifier for the product recommendations model (obtained from Almeta Cloud console).
number_of_recommendations	Specifies the number of product recommendations to return. Default: 12
recommendation_filters	Filters to exclude certain collections or products from the recommendations.
cart	Current contents of the user's cart, including product ID (required), quantity, and price.
selected_product	Details of the product that has been selected by the user, including product ID (required), quantity, and price.

Common use cases

Product Recommendations can be used in various scenarios to provide personalized product suggestions to your customers.

Personalized Recommendations

Recommend products based on the user's behavior and preferences on your website or email campaign. In this case, you only need to provide the model_id.

Recommendations for the product page

Recommend products to upsell or cross-sell on the product page. Provide the selected_product data point to get recommendations based on the selected product. The selected product is automatically excluded from the recommendations.

Add to cart recommendations

Recommend products to add to the cart based on the selected product and cart. Provide the selected_product and cart data points to get recommendations based on the selected product and the current cart contents. The selected product and products in the cart are automatically excluded from the recommendations.

In-cart and thank you page recommendations

Recommend products to upsell or cross-sell on the cart or thank you page. Provide the cart data point to get recommendations based on the current cart contents. The products in the cart are automatically excluded from the recommendations.