Get Started

Product Recommendation

Push the right product to the right consumer by building a recommendation system using collaborative filtering and machine learning.

This project offers an adapt-and-apply solution allowing retail analytics teams to build a recommendation system in order to push the right product to the right customers. The solution showcases how to use the new Recommendation System Plugin to solve a real-world use case. More details on the specifics of the solution can be found on the knowledge base.

Business Overview

95% of companies that implement personalization see 3x ROI in the year after the investment. And this is no surprise: indeed, 91% of consumers are more likely to shop with brands that recognize, remember, and provide relevant offers and recommendations. Recommending the right product to consumers has thus become a must-do to secure market shares and grow loyalty. This can notably be done by implementing a recommendation engine based on a collaborative filtering approach that aims at answering a simple question: what are the items that users with interests similar to yours like?

By doing so, brands can recommend products that have not yet been purchased by a user and that a group similar to this user has purchased. What the outcome is: encourage customers to discover new products they simply would not have actively looked for, and keep them engaged longer. Several use cases can then be run, many of them online: personalization of website homepage in a very tailored way (for logged-in users), sending of personalized promotional emails based on past transactions, … Impact of built recommendation engines will be maximized by the quality of data:  the richer the history is, the more performant the recommendation engine will be.


  • Input transaction data, split your customers into established vs growth depending on the number of past purchases or interactions with the brand
  • Leverage our recommendation plugin to calculate user < > item affinity
  • Define the solution settings by easily changing the settings of our recommendation plugin
  • Leverage Dataiku’s auto ML features to train a model that will predict whether a user will buy an item
  • Consume the output and understand which products should be pushed to which consumers, and for each product what are the top ten most similar ones
  • Automate the pipeline to rebuild with new/fresh data and to send the results to a CRM system