Customer Retention

• Developed and implemented retention strategies/models saving the client over $100 million dollars annually



• Used a methodical approach which included:
  
  • Data Reduction and Transformation Techniques
    
  • Data Cleansing
    
  • Predictive Modeling
    
    
• Identified key triggers of attrition through statistical analysis techniques

Purchase Likelihood Model

• Improve selection of "most" valuable customers through:
  
  • Data Preparation Techniques
    
  • Data Reduction
    
  • Data Aggregation & Trending
    
  • Data Cleansing
    
  • Predictive Modeling Techniques
    
    
• Results includes:
  
  • Saving on DM cost by targeting the most profitable and most responsive customer to be included in sales promotional campaigns
    
  • 93% incremental lift in the number of customers responding to the campaign
    

Cohort Analysis

• Analysis seek to improved client's understanding of "preferred/valued" households based on income, investment assets and bank deposits
• Construct cohorts based on household income, value of household deposits with bank and value of household investments/assets
• Identify prospects households for which the client could target as a growth strategy by region and by product offering
• Provide insight into household income, investment assets and bank deposits
• Identify most profitable cohort segments for targeting

Sales Forecasting

Evaluate the impact of economic and financial data/information in influencing market share

This was accomplished through a two-stage process which include:

• Aggregation of monthly economic and the client sales data
  
• Variable reduction through factor analysis
  
• Utilizing results of factor analysis as input for regression modeling
  

The knowledge gain is a catalyst for:

• Determining the impact of various economic scenario on credit quality and performance of retail/lease portfolio
  
• Forecasting varoius credit quality measures
  
• Developing an "early warning system" for varoius measure of credit/portofolio performance by identifying the relationship among economic, demographics and financial factors that lead to, and contemporaneously related to credit delinquency, bankruptcy and financial losses


• Identifying the strength of the relationship as it relates to economic risk and the length of impact lead time

Penetration/Promotional Analysis

• Converted disaggregated panel data into value-add insight for client through:
  
  • Data Preparation Techniques
    
  • Data Transformation
    
  • Data Aggregation & Trending
    
  • Data Cleansing
    
• Results includes:
  
  • Identifying opportunities for Buy One Get One Free promotions verses multi-unit purchase promotions on clients brand