Product Recommendations

Lytics provides tools to marketers to understand both your users as well as the content they interact with. Just as Lytics automatically ingests your content and makes it available for content recommendations, Lytics automatically ingests your product catalog to make it available for product recommendations.

Recommendations can be customized to include any subset of products in your catalog and ranked by things like product similarity, brand affinity, global trends, etc.

Product recommendations share a lot of the tooling available within Lytics content recommendations that you may already be familiar with. Recommendations in Lytics are always performed against a collection of content or products that are relevant for your users. You may be looking to make complementary recommendations, such as in adjacent product categories, or there may be a particular campaign or initiative to promote a particular product line that might change the types of products you want to recommend.

Most customizations can be performed by creating relevant product collections. Consider the following recommendation use cases:

• Adjacent products: If a customer already has a dress in their shopping cart, you more likely want to recommend things that pair well with dresses — like a bag — than recommend similar dresses to what they already have. In this case, you’d likely want to recommend from a product collection of just bags, rather than from your global catalog.

• Promotions: Seasonal promotions can often inform the types of products you want to recommend — coats in Winter, grills in Summer, etc. In these cases, you might want to create a collection of Summer Products to ensure relevant recommendations that are consistent with your marketing objectives.

While Lytics ingests product data automatically, sometimes you’ll want to override things in your catalog, remove specific items, or even import your catalog directly from a custom data source.

You can update your product catalog in the same way you update user data. For more information and examples, see the articles on user fields and custom data sources.