Machine learning pilot for customer retention

A behavioral predictive model for customer churn in the segment of consumer loans.

Erste Group



With around 47,000 employees, serving over 16,2 million clients in more than 2,700 branches in 7 countries, Erste Group is one of the largest financial services providers in Central and Eastern Europe. Erste Group IT (EGIT) is the group’s trusted and innovative delivery partner, responsible for building and maintenance of its core IT systems.

As the worldwide banking sector is engaging in far reaching digital transformation, Erste Group IT also identified an opportunity to accelerate their business using machine learning methods. They gave trust to Cognexa and a partner company ISDD plus to carry out a machine learning Proof of Concept (PoC) project in the field of customer predictive modelling.


The basic idea of the project was to make use of the richness of data collected in one of the group’s countries. We focused on generating actionable business intelligence from available data and we created a behavioral predictive model for customer churn in the segment of consumer loans. 

Our team delivered a Python based model, as well as the corresponding data processing pipelines. 


After successful acceptance testing, the software components were passed to the customer’s in house teams for integration.


Customer churn modelling for a multinational banking institution


14 weeks to delivery


Python, SQL


Successful pilot project was important in shaping the way how Erste uses machine learning

“Cognexa went far beyond traditional data analytics and piloted the usage of the powerful technology in a high quality delivery.“
Branislav Šebo
CEO, Erste Group IT
Inspired by our work? Or want more information? Don’t hesitate and write us a message.
Subscribe to newsletter
*Required fields
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.