Purpose – The purpose of this paper was to examine the analysis of human research data to identify drivers of employee promotion. These characteristics may enhance the identification of high potential employees, which in turn helps to build a pool of high potential employees. This allows organizations to fill key positions and improve organizational performance. 

Approach – Seven classification machine learning techniques were compared building predictive models. The best model was chosen on evaluation metrics and performance. 

Findings – Building on existing literature, this paper found that decision trees are the most accurate and most transparent model for this type of human resource classification problem. Along with the technical work, discussions and recommendations on assessing datasets for fairness and implementation were presented in response to many barriers to talent analytics. 

Originality/Value – This proof-of-concept case study provides a foundation for further work with real human resource data and contributes to the research in the area of human resource and talent development. 

Key Words – Human Resource Development, Talent Management, Talent Analytics, Machine Learning, People Analytics 

Paper Type – Proof of concept case study.