Banking and Decision Science:Overview

Banks and Financial Institutes are leveraging Data Science and Insights for delivering values to their customers, innovating new product and propositions, minimizing risks to their shareholders, improving operation efficiency & excellence and increasing employee engagement.


DnI Consulting provides broad range of Data Science & Insights services across products and portfolios of the retail, commercial and investment banks. We have extensive experience and knowledge of working on broad range of business problems, analytics tools & technologies and statistical & machine learning techniques.


Data Science and Insights for Banking and Financial Services:-
Customer Analytic Product & Portfolio Analytics Sales and Marketing Analytics Risk Analytics Operations Analytics
Customer Segmentation Pricing & Scenario Analytics Campaign Analytics Portfolio Monitoring & Management Collection Analytics
Cross Sell Management Profitability and Revenue Modeling Market Mix Modeling Scorecard Development: Acquisition & Behavior Call Center Analytics
Retention and Attrition Management Offer and Proposition Analytics Sales Driver Analytics Regulatory Reporting IVR Analytics
Web Analytics Sales Opportunity Analytics Loss Forecasting Fraud Strategy and Analytics
Channel Analytics Leads Identification and Modeling Model Validation and Monitoring
Next Best Action Modeling
CLTV Modeling
Data Science & Insights deployment:-
Offering Deployment Scenario
Customer Analytics Developing a cross sell model for targeting customers who are more likely to respond to a Personal Loan/Credit Card product.
Product & Portfolio Analytics Credit Card Providers run different credit card acquisition offers such as 3.9% BT for 6 month, 5.9% BT for 9 months and similarly retail offers. Running simulations and scenarios for estimating profitability of different offers and scenarios.
Sales and Marketing Analytics Market Mix Model to understand the drivers of current account sales and finding optimal allocation of the marketing budget in driving current account sales.
Risk Analytics Monitoring portfolio risk monitoring and building strategies for minimizing portfolio risk including Risk Based Pricing.
Operations Analytics Operations Analytics Collection Model development for improving operational effectiveness and efficiency.
Point of Views:-
Average Product holding: Number of banks are looking to increase cross product holding of the customers or average product held by a customer Considering high acquisition cost, the banks and financial institutions leverage customers’ relationship to grow their share of wallet (SoW) from the existing customers and increase overall product penetration. One of the banking clients had very high customer satisfaction but very low product penetration and wanted to understand the ways to augment product penetration. Product penetration could be low due to multiple reasons but two prominent scenarios could be

Product Proposition mismatch: The financial institution have high credit card base and most of the customers are high spend Transactor customers. The other products being offered are Personal Loan and Insurance products. The customers were regular payers since they had good financial position. The financial institution build different credit card proposition for these high value Transactors to increase share of wallet from the customers.

Sales Forces In-effectiveness: The banking client has good branch presence and most of the customer acquisitions are driven from the branch network. Due to lack of sales target and education programs, the financial advisors and branch personnel were not proactive in discussing with customers on their financial needs and product requirements. Most of the current product sales were initiated by the customers.

Increase Fraud Losses:A credit card portfolio loss has seen an increasing trend in last couple of quarters. Fraud has serious repercussions not only financial losses but also customer dissatisfaction. The financial client has seen an increasing trend on fraud losses on their credit card portfolio. Following approach has been used to reduce the fraud losses.
  • Exploratory Data Analysis to understand the aggregate credit card usage – domestic vs international, Card Present(CP) vs Card Not Present (CNP)
  • Defining Predictive Modeling framework and prepare data
  • Build predictive model classifying transactions as fraudulent
  • Validation and deployment of new predictive model
Wealth Management Portfolio Churn:An exodus of fund & customers for a wealth management portfolio.

Managing existing customer base is no longer an easy task and financial organizations have realized it in a hard way. Ease of switching to another financial institution has increased the challenges.

One of the wealth management clients has seen erosion of Asset Under Management (AUM) for its portfolio.

3 pronged strategy :-
  • Identifying recipient competitor products
  • Customer Survey & Market Research to understand reasons along with internal data analytics
  • Predictive Model to identifying customers likely to move the fund out