Method and medium for customer product recommendations

A non-transitory machine-readable medium storing instructions executable by a processing resource to:

• obtain, from a digital products data store, a product portfolio offering of products offered for purchase;
• identify a first product activity of a first customer from a first data store, the first product activity identifying a first set of products of the product portfolio offering in which the first customer has expressed interest, wherein the first product activity includes at least an in-store activity of the first customer;
• analyze interdependence between products of a second set of products of the product portfolio offering purchased by a second customer, the second set of products identified from a second data store, wherein interdependence comprises a dependence between two or more products to accomplish a task;
• compare the first set of products associated with the first product activity of the first customer to the second set of products;
• generate an optimized product portfolio offering for the first customer by predicting a set of related products from the product portfolio offering based on application of a predictive analytics layer to the first set of products, the second set of products, and the product portfolio offering, wherein the set of related products is based on similarity between the first set of products and the second set of products and interdependence between each of the first set of products and each of the second set of products, wherein a number of products in the optimized product portfolio offering is less than a number of products in the product portfolio offering;
• rank each product of the set of related products in the optimized product portfolio offering based on the product's ability to enable sale of one or more other products of the optimized product portfolio offering and an ability of the product to complete a task of the first customer;
• based on the rank of each product of the set of related products, generate a reduced number of display products to display to the first customer from the optimized product portfolio offering to a recommended subset of the set of related products in order to cause a product menu to load the recommended subset of the set of related products faster than reducing the number of display products to display to the first customer from the product portfolio offering;
• display the recommended subset of the set of related products to the first customer in the product menu based on the comparison of the first set of products and the second set of products, an affinity of the first set of products and the second set of products, a similarity of the first customer and the second customer purchasing the first set of products and the second set of products, respectively, and a digital customer intent, wherein the product portfolio offering comprises a plurality of products having interdependencies;
• and
• aggregate and store data of the first customer into a data lake based on the first product activity of the first customer and global purchases of the first customer.