Predictive Analytics In Banking. Use Cases - Ciklum

Research Report:

Predictive Analytics In Banking. Use Cases

Learn in the Report:
  • How to predict customer gender and future expenses with data analytics
  • Pros and cons of different Predictive Modeling approaches
  • How to put predictive analytics to work for your organization
GET YOUR COPY
By submitting your details you authorize Ciklum to contact you by e-mail with information about Ciklum updates, and agree to our Terms and Privacy Notice.
It costs banks 6x more to attract new customers than to retain existing ones
In the banking, finance and insurance industries a perfect customer experience is what keeps clients coming back. Data Analytics can help you grasp the enormous potential of available data and tailor propositions for your customers.
DATA SCIENTISTS ANALYZED THE FOLLOWING DATA:

7 mln

individual transactions

15,000

unique customers

15

months

Model development for predicting:
  • customer behaviour,
  • aggregated daily turnover by expense category for the whole customer base,
  • individual expenses by category in the next month – Customer Lifetime Value (LTV).
Technologies:
  • Black-box modelling to predict customer gender
  • AUC-ROC and RMSLE metrics for predicting daily aggregated turnover,
  • Time-series prediction method ARIMA to improved error metrics,
  • XGBoost applied.
Data Science Team:
Download the report
By submitting your details you authorize Ciklum to contact you by e-mail with information about Ciklum updates, and agree to our Terms and Privacy Notice.