Customer : E-Commerce (Italy)
Category : CRM
Tags : clustering, personalization

Customer base analysis

Target: personalise the homepage and the newsletters shown to each user in order to increase their propensity to purchase

Data: several years of navigation logs and site activity
▪Interests
▪Temporal dimension

Implemented algorithms
▪Clustering algorithm with respect to the activity on the website (product categories)
▪Clustering algorithm with respect to the activity on the website (temporal activity)
▪ Survival curve algorithm for conversion/channel analysis

Implemented interfaces
▪ Creation of APIs that can be called from external systems and that return results related to the selected user
▪ Dynamic graphical interface for data exploration