Current Situation: HR and managers lack a systematic tool to assess employee churn risk. They rely on intuition and limited data points, leading to:
- Reactive retention strategies: Responding to resignations after the fact, missing opportunities to proactively engage and retain valuable talent.
- Potential talent loss: Losing "A-player" employees with high institutional knowledge and productivity, leading to significant costs and disruptions.
Cost Comparison:
- Finding new talent: Studies show replacing an employee can cost 1.5-2 times their annual salary, including recruitment, onboarding, and lost productivity.
- Retaining an A-player: Investing in employee engagement, career development, and competitive compensation is often significantly cheaper than replacing them.
Desired Outcome: Develop a machine learning model that predicts the probability of employee resignation and identifies key factors contributing to this risk.
Key Deliverables:
- A user-friendly web interface allowing HR/managers to input employee data.
- The interface displays the:
- Probability of the employee leaving the company.
- Top factors influencing this prediction, providing actionable insights.
Impact:
- Proactive talent retention strategies: Identify and address employee concerns before they resign, focusing on high-risk individuals.
- Improved employee engagement and satisfaction: Creating a positive work environment reduces churn and attracts top talent.
- Reduced hiring and training costs: Saving money associated with replacing lost employees.
Additional Considerations:
- Data access and quality.
- Model fairness and interpretability.
- Implementation within existing HR workflows.
By predicting and addressing potential churn, this project helps companies avoid the significant costs associated with losing valuable employees.