Targeted Marketing Based On
Predictive Customer Behavior

Challenges and Industry

A national cosmetics retailer was not using the customer data it collected to its fullest potential. Company leadership was aware that they did not have a clear understanding of their customers - which hindered the ability to produce effective marketing campaigns, launch desired products and generally grow revenue. When approaching GFAIVE, the retailer had already collected over 1 million data entries on its customers, but only 20% of the entries were connected to transactions.

The challenge for GFAIVE was to build a customer intelligence tool that performed accurate customer segmentation and provided insights and predictions about each segment's behavior.



To create the customer intelligence tool GFAIVE had to first prepare and clean the current dataset of the client. This step was imperative for successful implementation into any system and for further building of historical and predictive models.

Next, to perform an accurate customer segmentation, machine learning algorithms were applied to the 20% of customers that the client had full data on. This helped identify hidden correlations and factors that influenced customer behavior that would otherwise have been difficult to recognize for a human.

Finally, the results of the analysis could then be applied to the remaining customer base – allowing it to identify duplicate entries, and take focused actions on previously “invisible” customers. After the customer intelligence tool was applied to the client's entire dataset, it was integrated into their existing marketing systems and set up to perform without constant supervision.



As a result, the cosmetics retailer gained a deeper understanding of their customers and an ability to quickly identify which were most valuable customers, which were at risk of churn and what actions were needed for each of them.

Following the integration of the tool and leveraging it to create marketing campaigns, the client saw a 12% reduction in marketing costs as they were now launching more effective campaigns. Moreover, an 11% sales growth further illustrated the success of the tool as the retailer could now customize marketing messages and cater effectively to every customer segment.

Technologies Used

© 2021 by GFAIVE - ML demand forecasting for fashion retail. Privacy Policy

  • Instagram - White Circle
  • GFAIVE LinkedIn page
  • GFAIVE Facebook page
  • Twitter - White Circle
  • YouTube - White Circle