Skip to main content

The Problem

One of our clients has a large stable of products to offer their consumers via their website. They wanted to increase revenue by exploiting cross-selling products via an automated system of product recommendations. In other words, what should be recommended to each specific consumer or type of consumer?

The Solution

FESSEX Consulting implemented for them a recommendation system which predicts a consumer’s utility for an item based on other consumers’ previous utility with the same item and the consumer’s purchase behavior (history.)

For example: A consumer purchases product (a), and afterwards it purchases product (b). If another consumer purchased product (a) there is a probability that s/he will like product (b). The recommendation algorithm finds such probability in order to recommend the appropriate product. Additionally, the recommendation will depend on previous behavioral data of each consumer, and finding other consumers with similar behaviors and what they purchase.

recommended Items chart
  • To visualize the results we created a dashboard that updates daily showing the products the company should recommend to consumers.
  • Furthermore, the recommendation results are connected to the website adding a layer of customization to the consumer experience.

The Results

The recommendation algorithm assisted in adding 15K additional product purchases per month via cross-selling on the website and additional personalized email marketing campaigns based on the recommendation results, leading to an additional $750K revenue per month during the first 3 months.

The Importance of Product Recommendation

Product recommendations and the results of actions based on them have the following benefits:

  1. Better inventory management: if a company knows what their consumers will order they can keep more of those items on hand rather than those that are not.
  2. Better understanding of how products affect each other: It may not be obvious that two products potentiate each other. With a product recommendation that relationship becomes testable.
  3. Better consumer understanding: If it’s known what a consumer will purchase based on a data driven system, a company can understand better what needs are driving their consumers providing additional value to their consumers.

In summary, cross-selling based on product recommendations can generate significant additional revenue, higher customer satisfaction and larger margins.

Contact us today for a FREE Discovery Call, to learn how we can help you with your next project.


Contact Us