A Power BI project that explores employee attrition (turnover) to uncover key factors, high-risk departments, and actionable insights to improve employee retention.
This analysis provides a data-driven understanding of employee turnover patterns within the organization.
It combines Power Query data cleaning, interactive dashboards, and drill-down visualizations to help HR teams identify and act upon attrition causes.
This project uses the HR Analytics Dataset from Kaggle, containing details of 1,470 employees across departments.
The dataset includes information on demographics, job roles, satisfaction levels, income, performance ratings, and attrition status.
Key Columns:
Age,Gender,Department,JobRoleMonthlyIncome,DistanceFromHome,JobSatisfactionYearsAtCompany,Attrition(Yes/No)
Source: HR Analytics - Kaggle
Data Source → Power Query (Cleaning & Transformation) → Data Model (Relationships, Measures, DAX) → Power BI Dashboards (KPIs, Decomposition Tree, Insights)
Performed in Power Query (Power BI) to ensure accuracy and consistency.
Key Transformation Steps:
- Removed duplicates & handled missing/null values
- Corrected data types (e.g., ensuring numeric fields were not stored as text)
- Created meaningful categorical bins for:
- Age Groups → Young Professionals, Mid-Career, Experienced
- Income Ranges → Low, Medium, High
- Tenure Bins → Early Tenure, Mid Tenure, Veterans
- Job Satisfaction → Low, Medium, High, Very High
- Distance from Home → Near, Mid Range, Far
| Phase | Objective | Description |
|---|---|---|
| Identification Phase | Locate high-attrition departments | Identified which departments (Sales, HR, R&D) had the highest attrition rates using aggregated metrics and visuals. |
| Investigation Phase | Diagnose root causes | Explored job-related and demographic factors such as salary, tenure, job satisfaction, and distance from home. |
| Insight Phase | Translate findings into action | Derived HR-specific recommendations for improving retention and employee engagement. |
The Power BI report includes:
An Overall Company Dashboard
Department-level Dashboards (Sales, HR, R&D)
An Interactive Decomposition Tree for root-cause analysis
| Insight Category | Observation | Impact |
|---|---|---|
| Departments | Sales (20.63%) & HR (19.05%) have the highest attrition rates | Require targeted retention strategies |
| Demographics | Younger, single employees with shorter tenures and lower salaries are more likely to leave | Indicates early-career disengagement |
| Work-Life Balance | Employees living farther away and with low job satisfaction show higher turnover | Suggests need for hybrid/flexible options |
Provides an overview of 1,470 employees and the overall attrition rate (16.12%).
Includes KPIs, departmental treemap, and job role breakdown.
Interactive exploration of attrition rate by dimensions such as department, role, and satisfaction level.
The Sales → Sales Representative → Entry Level path shows 42.11% attrition, the highest risk group.
Department with the highest attrition (20.63%) among 446 employees.
Explores patterns by age, satisfaction, and monthly income.
Largest department (961 employees) but relatively lower attrition (13.84%).
Focuses on employee tenure and engagement levels.
A smaller team (63 employees) yet second-highest attrition (19.05%).
Explores satisfaction and promotion rates.
| Metric | Value | Description |
|---|---|---|
| Total Employees | 1,470 | Company size analyzed |
| Overall Attrition Rate | 16.12% | Percentage of employees who left |
| Highest Attrition Department | Sales (20.63%) | Critical turnover zone |
| Second Highest Department | Human Resources (19.05%) | Needs retention attention |
| Common Factor | Low income & low satisfaction | Primary driver of exits |
- Flexible Work Options — for employees living far from the office
- Engagement Programs — boost satisfaction among early-career staff
- Compensation Review — ensure pay equity and motivation
- Career Development Plans — improve retention through growth opportunities
Imran Akbar
If you found this project insightful, give it a star on GitHub!