Skip to content

khairulanum91-source/workforce-intelligence

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Workforce Intelligence Dashboard

HR Analytics Case Study — Medika Nusantara (Fictional Client)

🔗 View Live Dashboard


Project Overview

This project simulates a real-world Business Analyst engagement where HR leadership needs to identify which employees are at risk of leaving — before it's too late to act.

The client, Medika Nusantara (a fictional healthcare organization), was experiencing rising attrition but had no systematic way to prioritize retention efforts. This dashboard was built to give HR managers a data-driven view of workforce risk, structured around actionable insights rather than raw numbers.


Business Problem

"We know some employees are unhappy, but we don't know who to focus on — or why."

HR teams often rely on gut feel or exit interviews that come too late. The goal of this project was to translate messy employee data into a clear, prioritized risk signal that managers can act on immediately.


What This Dashboard Does

  • Attrition Risk Scoring — Each employee is assigned a risk level (High / Medium / Low) based on performance, satisfaction, and tenure signals
  • Confidence Score — Shows how reliable each risk assessment is, so HR can prioritize high-confidence cases first
  • Messy vs. Clean Data Toggle — Demonstrates the impact of data quality on analysis reliability; a before/after view of the same dataset
  • Employee Detail Panel — Click any employee to see their full profile, risk drivers, and suggested retention action
  • Search & Filter — HR managers can search by name, department, or risk level

Key Analytical Findings

Risk Level Employee Count Primary Driver
High Risk 8 employees Low satisfaction + poor performance
Medium Risk 12 employees Stagnant tenure, no promotion
Low Risk 30 employees Stable engagement scores

⚠️ Data quality gap identified: The messy dataset contained 23% missing values and inconsistent formatting — enough to misclassify 4 employees from High to Medium risk. This finding was flagged as a recommendation to improve HR data collection processes.


Tools & Methods

Area Detail
Data Preparation Manual data cleaning (CSV), inconsistency flagging
Analysis Logic Rule-based attrition scoring, confidence weighting
Visualization Interactive HTML dashboard (vanilla JS, no frameworks)
Deliverable Format Live web app + dataset files

Files in This Repository

File Description
index.html Interactive dashboard (open directly in browser or via GitHub Pages)
dataset_A_messy.csv Raw HR data with quality issues — simulates real-world input
dataset_B_clean.csv Cleaned version used for final analysis

Context

Context

This is an independent BA case study designed to simulate a real-world analyst engagement from end to end:

  1. Defining the business problem from a stakeholder perspective
  2. Structuring the analytical approach
  3. Cleaning and interpreting messy data
  4. Building a deliverable that non-technical stakeholders can use immediately

The fictional client, department names, and employee data are entirely synthetic — designed to reflect realistic HR analytics scenarios in the healthcare sector.

About the Author

Khairul Anum — Business Analyst & Market Research Specialist

Currently transitioning into BA/BI roles with a background in Business Development and market research. This project reflects my interest in turning messy organizational data into clear decisions.

📎 LinkedIn | 📂 Other Projects

About

HR attrition risk dashboard for a fictional healthcare client — built as a BA case study using employee data analysis and interactive visualization.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages