Collection Analytics services help organizations understand their customer segments and scores them based on their propensity and ability to pay.

Market Equations helps organizations prioritize collection efforts, minimize defaults, maximize recoveries and reduce overhead costs by identifying customers who have a higher propensity to pay and targeting them with innovative and tailor made collection strategies.

Our Centre of Excellence in India combines advanced analytics expertise and decades of consulting experience to help organizations in the Banking, Financial Services and Telecom industry develop innovative and targeted collection strategies to increase collections and maximize customer profitability.

Market Equations helps a leading Bank develop a scorecard at the customer account level such that higher the score the more likely the account would default in the next month. This would enable the bank devise differential multimedia treatments at a segment level and optimize collection efforts to boost collection revenues.

Objective:

• Identify Good Accounts from Bad to prioritize collection efforts
• Build a scorecard to discriminate the potential good accounts (which have higher likelihood of regular payments) from the bad accounts for early delinquent portfolios (bucket 1 i.e. < 30 DPD and bucket 2 i.e. < 60 DPD)
• Scores to be used to maximize returns by prioritizing collection efforts
• Develop a targeted collections approach rather than a flat approach across all customer segments
• Develop targeted multimedia follow up treatments

Collection Analytics outsourcing services from Market Equations India helps organizations prioritize and target collection efforts to help them maximize collections, minimize defaults, maximize recoveries, reduce overhead costs and maximize customer profitability.

Approach:

Logistic regression technique was used on the historical customer profile and transactional data to first identify and select important features and then discriminate accounts at a segment level. Various statistical techniques like cross tab analysis, regression selection methods, information values etc were used to select the important fields.

The model was developed over 6 month data and validated over the next 6-month data. The final output (score) was created at the account level such that the higher the score, more likely the account would default in the next month. The model would evaluate all eligible accounts on the first day of the month and the prediction of the model (score) would be valid for the rest of the month.

After the model development, a detailed model summary document was shared that presented the details of the overall model fit statistics, forecasted figures, and model validation results.

Outcome:

Market Equations helped the bank devise differential follow up treatment strategies at the segment level to optimize collection efforts to boost collection revenues.