Business / Analytics

Employee Turnover Prediction

Data-driven exploration of employee retention dynamics to predict turnover risk and identify key organizational drivers.

🏢 Business Context

High employee turnover leads to significant costs in recruitment, training, and lost productivity. Understanding why employees leave is crucial for HR departments to implement effective retention strategies.

📈 Analytical Approach

We analysed a dataset of employee demographics, performance metrics, and survey results. Using statistical correlation and regression models (and Machine Learning classifiers), we identified the strongest predictors of attrition.

  • Exploratory Data Analysis (EDA) to visualise trends.
  • Feature Engineering to create meaningful metrics.
  • Predictive modelling using Random Forest / Gradient Boosting.

Key Insights

The analysis revealed that lack of career progression and satisfaction with management were higher predictors of turnover than salary alone. The predictive model achieved 85% accuracy, allowing HR to proactively engage at-risk employees.

Tech Stack

R / PythonSQLTableauExcel
Turnover Chart