Introduction
Kahn (2018), in his study on Psychological Conditions of Personal Engagement and Disengagement at Work, demonstrated a strong connection between high employee performance and positive business outcomes. To foster a productive workforce that maximizes potential, ABC Company conducts employee performance analysis to identify areas for improvement and growth.
Project Objectives
The goal of this project is to understand the factors influencing employee productivity, enabling the HR team to develop policies that enhance both performance and job satisfaction.
Problem Breakdown
- What is the distribution of employee performance scores across different ranges?
- How does performance vary across different departments?
- What is the relationship between employee satisfaction and performance scores?
- How does gender impact performance scores?
- How do work hours per week influence performance scores?
- How does salary range affect performance scores?
Dataset Overview
Sourced from Kaggle, the dataset contains comprehensive employee performance information. It includes several factors such as employee demographics, job satisfaction, salary range, and performance scores, essential for evaluating employee productivity.
Methodology
Input Data to PostgreSQL
Initial data storage in PostgreSQL for structured management.
Extract, Transform, Load with Airflow
Airflow DAGs handle data extraction, transformation, and loading processes.
Data Cleaning
Essential cleaning performed within Airflow with validation checks.
Load to Elasticsearch
Cleaned data is pushed to Elasticsearch for indexing.
Visualization with Kibana
Dashboards created in Kibana for data exploration and analysis.
Conclusion
Department
Engineering and Operations departments perform the best, while IT, HR, and Customer Support need improvements.
Employee Satisfaction
Higher satisfaction correlates with better performance, though not universally significant.
Salary Impact
Employees earning $6,000-$10,000/month perform better.
Work Hours
Optimal performance peaks at 30 hours/week; declines with excessive hours.
Performance Distribution
Most employees score between 1-3.
Employee Gender Analysis
No significant difference across genders.
Recommendations
Review Salaries and Incentives
Increase incentives for lower salary brackets to enhance productivity.
Support for Underperformers
Provide training and mentorship programs for underperformers.
Promote Work-Life Balance
Avoid excessive work hours to maintain peak productivity.
Apply Best Practices
Share successful strategies from Engineering/Operations with other departments.
Enhance Satisfaction
Introduce employee recognition programs and growth opportunities.
Libraries and Tools
Github Repository
Access the complete and documentation for the Employee Performance Analysis Pipeline project on GitHub: GitHub Repository