The organization was facing a significant challenge with employee turnover being much higher than expected. This trend resulted in the loss of trained staff and negatively affected productivity while the process of finding replacements was slow and costly. Although internal feedback and exit discussions suggested that issues like workload pressure and compensation were potential causes the leadership team did not have a clear understanding of the exact reasons behind the resignations.
The primary objective of this project was to move beyond gut feelings and use hard data to determine exactly why employees were leaving. Management requested a detailed review of employee records to identify which groups were most at risk and to understand the specific patterns driving this behavior such as overtime pay differences and job satisfaction levels. The ultimate goal was to provide the leadership team with actionable insights to reduce unnecessary turnover protect skilled employees and build a more supportive work environment
The organization is facing higher employee turnover than expected, leading to increased hiring costs, productivity loss, and pressure on remaining staff. While exit feedback suggests issues related to workload, compensation, travel demands, and career growth, the exact drivers of attrition are not clearly understood.
Without a data driven view, HR is unable to identify which employee groups are most at risk or when intervention is needed. The challenge is to analyze employee records to uncover the key factors behind attrition and provide actionable insights that support timely and effective retention strategies.
The dataset contains records for 1000 employees, where each row represents one employee and their employment status. It brings together personal attributes, job related details, compensation, work environment factors, and career progression metrics. This structure allows attrition to be analyzed not as a single event but as an outcome influenced by multiple workplace conditions over time.
The data supports analysis across departments, age groups, income levels, travel requirements, satisfaction scores, and tenure. By connecting these attributes with attrition status, the dataset helps identify consistent patterns that explain why certain employees leave earlier than others.
| Column Name | Short Explanation |
|---|---|
| Age | Employee age in years |
| Attrition | Indicates whether the employee left the organization |
| BusinessTravel | Frequency of work related travel |
| Department | Business unit where the employee works |
| DistanceFromHome | Distance between home and workplace |
| Education | Highest education level |
| EducationField | Field of academic study |
| EmployeeNumber | Unique identifier for each employee |
| EnvironmentSatisfaction | Satisfaction with work environment |
| Gender | Employee gender |
| JobLevel | Seniority level within the role |
| JobRole | Specific job position |
| JobSatisfaction | Satisfaction with job responsibilities |
| MaritalStatus | Employee marital status |
| MonthlyIncome | Monthly salary earned |
| OverTime | Indicates whether overtime is required |
| PerformanceRating | Employee performance score |
| TotalWorkingYears | Total professional experience |
| WorkLifeBalance | Rating of work life balance |
| YearsAtCompany | Total years spent at the company |
| YearsInCurrentRole | Time spent in the current role |
| YearsSinceLastPromotion | Time since last promotion |
| YearsWithCurrManager | Time working with current manager |
Data Preparation in SQL I established a dedicated SQL database to house the raw employee records and ensure a structured environment for analysis I wrote specific scripts to clean the data by removing five redundant columns that offered no analytical value such as the standard hours flag To enable deeper insights I engineered new features by grouping continuous variables like age and tenure into categorical ranges for better segmentation
Data Transformation I established a direct connection between Power BI and the SQL database to maintain a live and up to date data feed for the report Using Power Query I rigorously checked all data types and removed any inconsistencies to ensure the model was optimized for performance This bridge between the database and the visualization tool guaranteed that the final metrics would be both accurate and reliable
Dashboard Development I built an interactive Employee Attrition and Retention Report that translates raw numbers into a visual story for the leadership team The dashboard features dynamic filters that allow users to drill down into specific departments and demographics to pinpoint high risk areas
The analysis reveals that while the majority of the workforce is stable there are critical segments where turnover is dangerously high causing disruptions to business continuity. The overall attrition rate currently stands at 16 percent which translates to 237 employees leaving the organization out of a workforce of 1000. This trend is not widespread but is concentrated among specific groups driven primarily by age compensation and work conditions.
Younger employees particularly those under 25 in the Sales department are the most vulnerable with attrition rates spiking to 75 percent in some areas. Compensation is a clear deciding factor as departing employees earned an average of 4110 dollars compared to 6630 dollars for those who remained. Furthermore 66 to 75 percent of the employees in these high risk segments were frequent travelers indicating that burnout is a significant contributor to the decision to leave.
To reverse this trend the leadership team needs to focus on specific interventions rather than broad policies. Immediate actions should include mentoring programs for young sales staff and reviewing travel schedules to prevent exhaustion. Adjusting salary bands for high performers in the lower income bracket and improving the onboarding experience for new hires will also be essential to retaining top talent and reducing replacement costs.