TABLE OF CONTENTS
This model recommends products with attributes similar to the product currently in view. It enables your customers to discover new products from the catalog and quickens the purchasing decision. Based on various criteria, the model assigns, similarity score for each product, and the products with the highest similarity score are recommended to the customers.
You can configure this model to create various types of recommendation strategies by assigning varied importance to product attributes such as brand, color, and category.
Configure this model to create a strategy with the following values:
Assign maximum importance to color
Assign negative importance to brand
Assign medium importance to pattern
Choose yes for 1:1 affinity-based personalization
Recommend products similar to those viewed by the customer with similar colored products appearing first, followed by products of similar pattern, and not showing products of the same brand. The recommendation results will be personalized, i.e., adapted based on the data collected about the customer's individual preferences.
For example, if a customer views a red-colored, floral dress the customer has displayed a liking for long-sleeved dresses in the past, then the configured strategy will recommend more similar long-sleeved dresses in the following order.
In the above example, if you choose yes to 1:1 affinity-based personalization. The result is displayed irrespective of the customer's past likings.
Configure Similar Products
- Choose Assets > Strategies from the top navigation bar.
- Click New Strategy.
Choose Contextual > Similar Products. The Manage Strategy page is displayed.
Click to configure model parameters.
Click + to add a Product Attribute and choose an attribute from the drop-down. The default product attributes are brand, color, and pattern.
Drag the slider to indicate how important the product attribute is to the model. The values range from -1 to 1.
Note: Click Reset to Default to restore model parameters to default values.
Indicate whether you want to apply 1:1 affinity-based personalization on the model.
Note: If you choose yes, the configured model parameters will be used as a starting point to deliver product recommendations. Thereafter the recommendations will be based on individual customer's affinities.