#Ambition
Ambition
of the mission

To be able to animate its community, our client relied on a database of qualified customers on all of their past purchase history. In order to better anticipate the future behavior of its customers and to target them more specifically, the CRM team wanted to invest in all the available data in order to predict the future value of the customer. Our team was responsible for setting up a project to estimate the evolution of customer purchasing behavior in order to be able to offer ever more relevant and interesting products for each type of customer.
#Method
Our
approach

Think
Predicting purchases
To be able to interact more effectively with its customer base and focus efforts on profiles with the highest potential, we focused our attention on Customer Lifetime Value by proposing the establishment of indicators on customers and by estimating their future value.
Make
Advanced Predictive Models
The success of the project is based on a close tripartite collaboration between the teams of Eulidia, those of the brand and those of the group. Eulidia played a facilitating role in this organization. The provision of relevant features for machine learning model training allowed the CRM team to benefit from a predictive segment for each customer over the N+1 year, accompanied by a score reflecting their predisposition to make a purchase in the coming months.
Scale
Autonomy and Decision Making
The Dataiku platform and the underlying models are now in production. Eulidia has also implemented good working and organizational practices around Dataiku, and has trained the team to give them autonomy in the general functioning of the platform and the monitoring of tasks. The business also benefits from predictive customer indicators in order to animate customers directly from their usual tools in a more precise manner.
#Benefits
Indicators
of success
