Bank

A banking institution applied advanced statistical analysis to uncover patterns in customer behavior, risk assessment, and operational performance. Quantitative methods drove data-informed decision-making across key business units, helping leadership identify high-value segments and reduce portfolio risk.

The decision behind the question

A retail bank wanted to know which customer segments were quietly carrying most of its credit risk while also generating most of its fee revenue, because a single blended view of the portfolio hid both.

Data and methods used

Account-level transaction histories, product holdings, and delinquency records were combined into one customer table. Group analysis separated the portfolio into behavioural segments, and prediction models ranked the account features that moved default risk the most.

What the analysis found

• •Fee revenue was concentrated in a small share of accounts that also showed the steepest balance volatility month to month.
• •Delinquency risk tracked payment timing patterns far more closely than it tracked income band.
• •Two segments that looked identical on demographics behaved very differently once product mix was taken into account.

What changed as a result

Segment definitions were rewritten around behaviour rather than demographics, and the risk team began reviewing payment-timing signals monthly instead of quarterly.

Related work is grouped under Market Research, and the methods used here are described in more detail under analytical capabilities.