Detecting and Explaining Outlier Customer Behaviors in Cohorts

This project aims to help Numr Inc. better understand when and why customers behave in unexpected ways—like suddenly leaving a service or becoming dissatisfied—even when they seem similar to happy or loyal customers. The intern will build a prototype tool that groups similar customers together and then finds the ones acting differently in terms of key business outcomes, such as churn or satisfaction. By spotting and explaining these “outliers,” the tool will allow Numr and its clients to act earlier—fixing issues before customers leave. This will help businesses improve customer experience, increase loyalty, and reduce revenue loss.

Faculty Supervisor:

Aijun An

Student:

Partner:

Numr

Discipline:

Computer science

Sector:

Information and cultural industries

University:

York University

Program: