Recommendation Strategy

Recommendation Strategies are complex machine learning models that dictate the logic behind various approaches to product recommendations. Powered by several algorithmic decisions, these recommendation strategies analyze, learn, and discover patterns in user, product, and contextual data to present every user with a personalized experience. They determine which products are included in the experience.

Vue.ai supports the following recommendation strategies:

• Contextual: This strategy tailors recommendations based on product attributes such as color and style, the frequency at which different products are bought together, and how various products are viewed together in a particular session. Some examples of contextual recommendation strategies are:
  • Visually Similar Products
  • Similar Products
  • Bought Together
  • Viewed Together

• Popularity-based: This strategy delivers recommendations based on product popularity, product ratings, and best sellers. Some examples of popularity-based recommendation strategies include:
  • Best Sellers
  • Best Sellers in Category
  • Most Popular (Trending)
  • Most Popular in Category
  • Top Rated
  • Most Popular in Geo

• Journey-based: This strategy recommends products based on a customer’s past engagement with the website. It displays recently viewed, last purchased, recently purchased, and products similar to those recently viewed. Some examples of journey-based recommendation strategies are:
  • Recently Viewed
  • Recently Purchased
  • Inspired by browsing history
  • Last Purchased
• Bundling: This strategy recommends a set of products to form an ensemble based on a chosen product. For example, if your customer views a black dress, this strategy displays a black shoe, a handbag, and a belt that can be bought together with the black dress.  This strategy applies to the fashion and home decor industry. Some examples of bundling strategies include:
  • Complete the look
  • Style it With/How to wear it
  • Curations (by name)

• Profile-based (Collaborative Filtering): This strategy identifies similarities between users based on their website interactions, such as add to carts, product views, products clicked, products liked, etc.,  to serve relevant recommendations. The product recommendation engine analyses users’ information with similar tastes to assess the probability that a specific customer may like something. Some examples of profile-based strategies include:
  • Top Picks
  • Bought by Others
  • Viewed by Others
  • Liked by Others