Identifying Key Factors Driving Employee Turnover
This project analyzes employee data to identify key factors driving turnover and develops a predictive model that can accurately forecast which employees are at risk of leaving the organization.
Developed a Random Forest classification model that predicts employee turnover with 98.4% precision and 91.5% recall, identifying workload and satisfaction as the primary drivers of retention issues.
Accuracy in predicting turnover
Ability to identify at-risk employees
Primary drivers of turnover
The analysis revealed a strong correlation between excessive workload and employee turnover:
Relationship between number of projects and employee turnover
No employee who had been at the company for more than 6 years left, suggesting longer tenure correlates strongly with higher retention. Focus retention strategies on employees in their first 5 years.
Relationship between working hours, satisfaction, and turnover
Employees who received low evaluation scores despite working fewer hours were at risk of leaving, suggesting recognition/career growth issues. Workload management alone isn’t sufficient—feeling valued matters.
This project followed a structured analytical approach to develop a reliable predictive model:
Performance comparison of different classification models
This predictive model offers valuable applications for HR and management teams:
Identify at-risk employees before they decide to leave for targeted interventions.
Optimize project assignments and hours to prevent burnout and sustain satisfaction.
Forecast turnover to plan recruitment needs and maintain operational continuity.
Addressing work-life balance and recognition is more effective for retention than focusing solely on compensation.