Employee attrition poses significant challenges to organizations, affecting productivity, morale, and financial performance. High turnover rates can escalate recruitment costs, deplete institutional knowledge, and disrupt workflows. This case study explores the dynamics of employee attrition within an organization, identifying key contributing factors, and offer strategic recommendations to enhance employee retention.
The dataset provided by IBM encompasses data from three distinct departments, offering a comprehensive view of employee behavior, satisfaction, and organizational dynamics.
This analysis addresses three primary business objectives:
To navigate the complexities of employee attrition, the following key questions guide our analysis:
Our analysis reveals that 16.1% of employees are leaving their positions, while 83.9% are staying. This overall attrition rate serves as a baseline for understanding the turnover landscape within the organization.
Age is a significant factor influencing employee turnover. Employees in their early 20s exhibit the highest attrition rates, nearing 50%. As age increases, the likelihood of leaving diminishes sharply, stabilizing between 10-20% for those in their mid-20s to 30s, and dropping below 10% for employees aged 50 and above.
Attrition rates between males and females are relatively comparable, ranging between 15-20%. This parity suggests that gender is not a primary driver of employee turnover within the organization.
Different job roles exhibit varying attrition trends. Sales Representatives lead with the highest attrition rate at nearly 40%, followed by Healthcare Representatives at around 20%. In contrast, positions such as Managers, Research Directors, and Manufacturing Directors showcase the lowest turnover rates, all below 10%.
Departmental analysis reveals that the Human Resources and Sales departments share the highest attrition rates at approximately 20%, whereas Research & Development maintains a notably lower rate of around 10%.
The distribution of monthly income provides deeper insights into attrition trends:
Overall Income Distribution:
Employees Staying:
Employees Leaving:
Analyzing tenure reveals a clear distinction between those who stay and those who depart:
This disparity suggests that shorter tenure correlates with higher attrition rates.
Promotion frequency offers additional perspective:
While the difference is modest, it hints that career progression opportunities may influence retention.
Training engagement presents a subtle yet notable trend:
The minimal variance suggests that while training is consistent, it may not be the primary driver of attrition.
Age dynamics play a pivotal role in employee retention:
This indicates that younger employees are more susceptible to leaving the organization.
Educational attainment offers a nuanced view:
This marginal difference suggests that education level has a limited impact on attrition rates.
Marital status emerges as a significant determinant:
Employees Staying:
Employees Leaving:
The stark contrast, especially among single employees, underscores the influence of personal circumstances on employment decisions.
Through meticulous analysis, several key insights have surfaced:
Younger Employees Have Higher Attrition Rates:
Compensation is a Critical Factor:
Sales and HR Roles are High-Risk:
Short Tenure and Limited Career Progression Correlate with Attrition:
Single Employees are at Higher Risk:
Training Opportunities Matter:
Job Satisfaction Plays a Role:
Drawing from the insights gathered, the following strategic recommendations are proposed to mitigate employee attrition:
Flexible Career Pathways:
Mentorship Programs:
Salary Adjustments:
Performance-Based Bonuses:
Targeted Retention Initiatives:
Workload Balancing:
Internal Promotions:
Skill Development Programs:
Job Satisfaction Surveys:
Recognition Programs:
Custom Training Plans:
Leadership Training:
Social Engagement Programs:
Work-Life Integration:
Transparent Communication:
Exit Interviews:
You can view and interact with the Python code used to generate the graphs here.
You can view the dataset used for the analysis here.