Workforce Retention Analysis

An end-to-end HR analytics project: cleaning a 5,000-record employee dataset, analyzing turnover, absenteeism, and compensation patterns, and building an interactive Excel dashboard to support a workforce retention strategy.

Business Question

Meridian Financial Services was losing staff and seeing inconsistent performance across departments. This project answers:

1. Which department has the highest turnover, and does it correlate with salary or tenure?
2. Is there a relationship between absenteeism and performance?
3. Which employee demographics are most associated with resignation?
4. How does compensation compare across departments, and does it align with performance?

Key Findings

• Sales has the highest turnover (49%): driven by below-average pay and the lowest performance score of any department.
• Absenteeism and performance are linked: Information Technology has the highest absenteeism (10.4 days/year) alongside below-average performance.
• Employees aged 31–40 drive the largest share of resignations (~35%), the group with the most external options and highest flight risk.
• Human Resources is a hidden retention risk: its best-in-company performance score is paired with below-average pay.

Process

1. Data cleaning: removed 50 duplicate records, standardized missing values, documented data-quality issues.
2. Descriptive statistics: company-wide and department-level averages for salary, performance, and absenteeism.
3. Pivot table analysis: turnover rate, absenteeism vs. performance, demographic breakdowns, salary comparison.
4. Written summary: plain-language findings and recommendations for HR leadership.

Repo Structure

├── README.md
├── analysis/
│   └── MeridianHR_Analysis_FINAL.xlsx    # full workbook: data, pivots, dashboard
├── docs/
│   └── written-summary.md                # board-ready narrative summary
└── data/
    └── clean_data.csv                    # cleaned dataset (anonymized)

Tools Used

Excel (PivotTables, SUMPRODUCT/AVERAGEIFS formulas, native charts)

Note on Data

This dataset is a fictional project dataset. Employee names have been anonymized in the public version of this data (data/clean_data.csv) as standard practice for any HR dataset, real or simulated.