An interactive Power BI dashboard analyzing employee attrition patterns at IBM, helping HR teams identify retention risks and develop targeted strategies to reduce turnover.
Employee attrition costs companies an average of $15,000 per employee in recruitment and training expenses. This dashboard helps organizations:
- Identify high-risk departments and roles
- Understand factors driving employee turnover
- Predict at-risk employees before they leave
- Make data-driven retention decisions
- Sales Department has the highest attrition rate at 20.6% (4.5% above company average)
- R&D shows moderate attrition at 13.8%
- HR has the lowest attrition at 19.0%
- 54% of attrition occurs within the first 5 years
- Critical period: 0-2 years (highest spike)
- Employees who survive past 5 years show strong retention
- Employees with low job satisfaction (1-2) are 2.8x more likely to leave
- Work-life balance rating of 1 correlates with 31% attrition rate
- Average income of attrited employees: $4,787
- Average income of active employees: $6,832
- 30% income gap between groups
- Employees working overtime: 30.5% attrition
- Employees without overtime: 10.4% attrition
- Overtime = 3x higher attrition risk
- Power BI Desktop - Data visualization and dashboard development
- Power Query - Data transformation and cleaning (removed 3 redundant columns)
- DAX (Data Analysis Expressions) - Created 10+ calculated measures
- Data Modeling - Single-table analysis with optimized relationships
- 5 KPI Cards: Total employees, active count, attrition count, rate %, avg income
- 7 Interactive Visuals:
- Attrition by Department (Bar Chart)
- Age Distribution (Donut Chart)
- Income vs Age Attrition Patterns (Scatter Plot)
- Attrition Trend Over Tenure (Area Chart)
- Job Satisfaction Impact (Column Chart)
- Work-Life Balance Analysis (Stacked Bar)
- 4 Dynamic Slicers: Department, Gender, Overtime, Job Role
- Cross-filtering: Click any visual to filter entire dashboard
- Detailed employee table with conditional formatting
- Matrix breakdown by Department → Job Role
- AI-powered Key Influencers analysis
Source: IBM HR Analytics Employee Attrition Dataset (Kaggle)
Size: 1,470 employees × 35 attributes
Key Fields:
- Demographics: Age, Gender, Marital Status
- Job Info: Department, Role, Level, Business Travel
- Compensation: Monthly Income, Salary Hike, Stock Options
- Satisfaction: Environment, Job, Relationship, Work-Life Balance ratings
- Performance: Performance Rating, Training Times
- Tenure: Years at Company, in Role, with Manager, since Promotion
- Target Variable: Attrition (Yes/No)
// Basic Metrics
Total Employees = COUNT('Employee Data'[EmployeeNumber])
Attrition Count =
CALCULATE(
COUNT('Employee Data'[EmployeeNumber]),
'Employee Data'[Attrition] = "Yes"
)
Attrition Rate % =
DIVIDE([Attrition Count], [Total Employees], 0) * 100
Active Employees =
CALCULATE(
COUNT('Employee Data'[EmployeeNumber]),
'Employee Data'[Attrition] = "No"
)
// Advanced Metrics
Avg Monthly Income = AVERAGE('Employee Data'[MonthlyIncome])
Avg Income (Attrited) =
CALCULATE(
[Avg Monthly Income],
'Employee Data'[Attrition] = "Yes"
)
Dept Attrition Rate =
VAR DeptAttrition = CALCULATE([Attrition Count], ALLSELECTED('Employee Data'[Department]))
VAR DeptTotal = CALCULATE([Total Employees], ALLSELECTED('Employee Data'[Department]))
RETURN DIVIDE(DeptAttrition, DeptTotal, 0) * 100
Based on the analysis, I recommend:
-
Sales Department Intervention
- Implement monthly 1-on-1s with Sales reps
- Review compensation structure (30% below company average)
- Reduce overtime requirements
-
Early Tenure Support Program
- Mandatory mentorship for 0-2 year employees
- 90-day check-in protocol
- Clear career progression roadmap
-
Work-Life Balance Initiatives
- Flexible work arrangements
- Overtime cap policies
- Wellness programs for high-stress roles
-
Satisfaction Monitoring System
- Quarterly pulse surveys
- Exit interview analysis
- Predictive attrition model (ML next phase)
- Download
IBM_HR_Analytics_Dashboard.pbixfrom/PowerBIfolder - Open with Power BI Desktop (free download: microsoft.com/power-bi)
- Click slicers to filter data
- Hover over visuals for detailed tooltips
- Navigate to "Employee Details" page for drill-down analysis
Browse the /Screenshots folder for static views of all visuals
Download IBM_HR_Analytics_Report.pdf from /Documentation folder
This project includes comprehensive documentation:
- 📊 Insights Summary - Detailed analysis findings and business recommendations
- 📖 Data Dictionary - Complete field definitions and data descriptions
- 🖼️ Screenshots - Dashboard visualizations and examples
- Data Preparation: Cleaned and transformed raw CSV data using Power Query
- DAX Proficiency: Created complex calculated measures with CALCULATE, DIVIDE, FILTER functions
- Data Visualization: Chose appropriate chart types for different data stories
- UX Design: Implemented interactive slicers, cross-filtering, and drill-through navigation
- Business Intelligence: Translated data patterns into actionable business recommendations
- Storytelling: Designed dashboard flow to guide users through key insights
Challenge 1: Attrition Rate % showing as decimal (0.161 instead of 16.1%)
- Solution: Multiplied by 100 in DAX and formatted as fixed decimal
Challenge 2: Too many job roles cluttering slicer
- Solution: Changed to dropdown style instead of tile layout
Challenge 3: Scatter plot hard to interpret with overlapping points
- Solution: Adjusted transparency (60%), added size variation, enabled zoom slider
- Add predictive attrition model using Python/R integration
- Include time series analysis with monthly attrition trends
- Create employee risk score calculator
- Build automated email alerts for high-risk departures
- Integrate with HR systems for real-time data refresh
Arjun Thorlikonda
- LinkedIn: [https://www.linkedin.com/in/arjun-thorlikonda-422b9a256/]
- Email: [arjunthorlikonda15@gmail.com]
This project is licensed under the MIT License - see LICENSE file for details.
- Dataset: IBM HR Analytics (via Kaggle)
- Inspiration: [Any Power BI community resources you used: ]
- Tools: Microsoft Power BI, Power Query, DAX
⭐ If you found this project helpful, please star this repository!
📧 Questions or feedback? Open an issue or reach out via LinkedIn!