Salifort Motors

HR SUMMARY

HR Data Analysis and Employee Retention Strategy

https://rodders.me/projects/salifort_project/salifort-motors/

Project Introduction

Objective

Utilize HR data analysis to identify factors contributing to employee turnover and propose actionable strategies to enhance retention.

Methodology

Exploratory Data Analysis and Machine Learning Model (XGBoost) to identify key predictors of employee attrition and predict future risks

Key Findings from Data Analysis

Main Observations

  • High turnover among employees with low satisfaction scores, excessive hours (>200/month), or extreme project loads.
  • Significant departures linked to salary levels—59% of those who left were on low salaries.

Critical data points

  • 20% low last evaluation scores in leavers vs. 10% in stayers

Detailed Analysis Insights

Employee Performance and Workload

  • 45% of employees engaged in 2 or fewer projects left the company.
  • Employees working more than 300 hours monthly showed a 100% turnover rate.

Tenure and Management

  • High turnover noted within 3 to 5-year tenure bracket.
  • Current employee-to-manager ratio stands at 23:1, suggesting a need for more management positions.

Risk Prediction Using ML Model


Probability Model Accuracy
High Risk >90% 72%
Medium Risk > 70% 91%
General Risk >50% 92%

Current Risk Predictions

Model Accuracy

  • 91.98% of past leavers would have been identified using the model, demonstrating effectiveness.

Current Risk Levels

67 employees identified above low risk, with 12 at high risk and 33 at medium risk

Estimate Cost of lost employees

  • Assuming a nominal cost of 50% of total annual salary to replace a member of staff
# of Staff Cost Per Head Est Cost
High Salary Left Cost 48 £75,000 £3,600,000
Medium Salary Left Cost 769 £50,000 £38,450,000
Low Salary Left Cost 1,174 £25,000 £29,350,000
Est. replacement cost 1,991 £71,400,000

Assumptions on salary have been made.

Blind application of salary review

The costs (and savings) by applying a 30% pay increase to everyone who left

# of Staff Cost Per Head Est Cost
High Salary Left Cost 48 £45,000 £2,160,000
Medium Salary Left Cost 769 £30,000 £23,070,000
Low Salary Left Cost 1,174 £15,000 £17,610,000
Est. replacement cost 1,991 £42,840,000

Assumptions on salary have been made.

Application of Machine Learning

Using the machine learning model prediction, we can identify at risk employees before it's too late.

# of Staff Cost Per Head Est Cost
High Salary Left Cost 12 £75,000 £900,000
Medium Salary Left Cost 33 £50,000 £1,650,000
Low Salary Left Cost 67 £25,000 £1,675,000
Est. Salary increase cost 112 £4,255,000

Assumptions on salary have been made.

Strategic Recommendations

Compensation Adjustments

  • Review salary structures to match industry standards

Workload Management

  • Balance project allocations and working hours

Recognition Programs

  • Enhance recognition mechanisms

Career Development and Satisfaction

Career Growth Opportunities
  • Expand opportunities for career advancement
Job Satisfaction
  • Employee surveys to tailor interventions to enhance the workplace
Management Development
  • Aggressively expand management training to reduce the employee-to-manager ratio and improve support

Implementation of Flexible Working

Flexible Working Options

  • Implement more flexible working arrangements to cater to diverse employee needs, enhancing satisfaction and retention.

Accident Analysis

  • Investigate the high incidence of accidents and ensure workplace safety and well-being.

Conclusion and Next Steps

Action Plan

  • Immediate review of salary and benefits packages
  • Introduction of targeted career development and management training programs

Follow-Up

  • Schedule quarterly reviews to monitor the impact of implemented strategies and adjust as necessary