Churn Detection, Verbit 

Lucas Bensaid

Data Science Fellows June 2020 Cohort 



The challenge here is to predict churn as much as possible 

Challenges (at least two) 

  1. Produced a clear definition of churn user behavior in order to detect it as soon as possible 
  2. Created, analyzed, and interpreted large datasets 
  3. Conducted reports that enabled a new business strategy 

Achievements (according to KPIs) 

Detection of different churn user profiles 

Future project development 

Generalize the churn analysis on the full database

Share this post

Share on facebook
Share on twitter
Share on linkedin
Share on email