An implementation and analysis of the popular IBM Employee Attrition dataset using Power BI. This repository highlights data restructuring, exploratory analysis, and business intelligence techniques to uncover employee turnover patterns.
The core objective of this project is to analyze workforce metrics and identify the primary operational drivers behind employee attrition.
- What is the average age and demographic profile of departing employees?
- Which business departments and job levels suffer from the highest turnover?
- How do compensation tiers and tenure correlate with workforce exit rates?
- Data Preparation: Microsoft Excel (Sanity checks & dataset familiarization)
- Data Modeling & DAX: Microsoft Power BI
- Data Source: Kaggle - IBM HR Analytics Attrition Dataset
The link to the dataset is as follows: https://www.kaggle.com/datasets/pavansubhasht/ibm-hr-analytics-attrition-dataset
To prepare the dataset for clean dashboarding, the data structure was optimized with custom columns and calculated metrics:
- Attrition (Binary Column): Converted text values ("Yes"/"No") into integer data (1/0) to enable seamless aggregation calculations.
- Age Group (Conditional Column): Segmented employee ages into two strategic groups: "Adult" and "Senior".
- Salary Slab (Conditional Column): Categorized monthly earnings into clear bands (Upto $5K, $5K-$10K, $10K-$20K, and $20K+).
- Attrition Rate (Calculated Measure):
- Demographic Profile: The average age of employees who left the organization is 37 years. Male employees show a higher attrition rate (17%) compared to female counterparts (14.8%).
- Department Vulnerabilities: The Research & Development sector experienced severe turnover, accounting for 133 total exits (56% of total attrition). The Sales department followed closely at 39%.
- The Compensation Factor: Low salary is the strongest driver of turnover. 69% of all departed employees belonged to the salary bracket below $5,000 per month (163 individuals).
- Tenure & Experience Risks: Transition periods are critical; 25% of all total departures happen during an employee's first year with the company. Staff at Job Level 1 experienced the highest volume of attrition (143 employees).
- Educational Backgrounds: Attrition numbers were highest among staff members hold degrees in Life Sciences (89) and Medical disciplines (63).
- Compensation Adjustment: Review and adjust compensation structures for roles falling into the under $5,000/month bracket to match market competitive rates.
- R&D Department Focus: Introduce targeted engagement strategies, workload audits, and team support programs within the high-stress Research & Development pipeline.
- Onboarding Retention Initiatives: Develop structured mentorship and milestone checks specifically targeted at workers during their crucial first year of employment.
- Career Development Pathways: Establish clear lateral and upward growth opportunities for entry-level professionals (Job Level 1) to curb early attrition.
- Data Engineering: Writing specialized DAX statements to build metrics and clean data schemas.
- Interactive Design: Setting up cross-filtering components and multi-range dynamic slicers.
- Data Synthesis: Synthesizing complex categorical variables into easy-to-read matrix summaries.
- Dashboard UI/UX: Designing clean data layouts using KPIs, clustered bar charts, and trend lines.