Employee Attrition Analysis

The Scenario

Leadership wants to understand why attrition has been climbing and which employee segments are most at risk of leaving in the next 90 days. Exit interviews provide anecdotes; this project provides a model.

Using IBM's HR Analytics Employee Attrition dataset — 1,470 employees across 35 attributes — the goal was to identify the strongest predictors of attrition and surface them in an actionable dashboard that HR business partners can filter by department, tenure band, and job role.

The Approach

Cleaned and profiled the raw dataset, validated data types and nulls, then built a star-schema data model in Power BI Desktop. DAX measures were written for attrition rate, average tenure by segment, satisfaction score distributions, and a composite risk index for each department.

The dashboard includes a summary page with KPI cards and a drill-through page that surfaces the top individual risk factors for any selected department or role level. Key finding: overtime status was the single strongest predictor — employees working overtime were 2.4× more likely to appear in the attrition group.

Tools & Skills Used

Dataset

IBM HR Analytics Employee Attrition & Performance — 1,470 employee records, 35 features including job satisfaction, overtime flag, distance from home, monthly income, and years at company. Available on Kaggle.