Employee Performance Dashboard

On This Page

• C2.1  Project Overview
• C2.2  Business Problem
• C2.3  Dataset Description
• C2.4  Objectives
• C2.5  Step-by-Step Solution
• C2.6  Complete Python Code
• C2.7  Expected Output
• C2.8  Visualizations
• C2.9  Key Insights
• C2.10  Business Recommendations

Capstone Project 2: Employee Performance Dashboard

C2.1  Project Overview

This project builds a performance analytics dashboard for a company's HR team, analyzing employee performance scores across departments and identifying the factors linked to strong performance.

C2.2  Business Problem

Management wants to understand which departments are performing best, whether tenure affects performance, and where to focus coaching and development resources.

C2.3  Dataset Description

Sample size: 150 employee records from the latest annual review cycle.

C2.4  Objectives

• Compare performance score distributions across departments
• Identify the department with the highest average performance
• Explore whether tenure is associated with better performance
• Recommend where HR should focus training and development efforts

C2.5  Step-by-Step Solution

• Step 1: Load the employee performance CSV into Pandas
• Step 2: Check data quality — missing values, duplicate employee IDs
• Step 3: Create a box plot to compare score distributions by department
• Step 4: Calculate average performance score per department
• Step 5: Create a scatter plot of tenure vs performance score
• Step 6: Summarize findings into HR-focused recommendations

C2.6  Complete Python Code

import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt

# 1. Load data
df = pd.read_csv("employee_performance.csv")

# 2. Clean data
df = df.drop_duplicates(subset="employee_id")
df = df.dropna(subset=["performance_score"])

# 3. Department-wise summary
avg_by_dept = df.groupby("department")["performance_score"].mean().sort_values(ascending=False)
print("Average Performance by Department:\n", avg_by_dept.round(1))

# 4. Correlation: tenure vs performance
correlation = df[["tenure_years","performance_score"]].corr()
print("Correlation:\n", correlation)

# 5. Visualize
sns.boxplot(data=df, x="department", y="performance_score")
plt.title("Performance Score by Department")
plt.savefig("performance_boxplot.png")

C2.7  Expected Output

C2.8  Visualizations

C2.9  Key Insights

• Engineering has the highest average performance score (79.8), while Support trails behind at 71.6 — an 8+ point gap.
• The Support department also shows the widest score spread in the box plot, suggesting inconsistent performance rather than a uniformly weaker team.
• Tenure and performance show only a weak positive relationship — experience alone doesn't guarantee stronger scores.
• A handful of low outlier scores appear across every department, indicating individual coaching needs rather than a department-wide issue.

C2.10  Business Recommendations

• Launch a targeted coaching program for the Support department to reduce score variability and lift the department average.
• Since tenure alone doesn't predict performance, invest in structured onboarding and training programs rather than relying on time-served.
• Identify and support the specific low-scoring outlier employees individually across all departments, rather than applying blanket departmental policies.
• Study Engineering's practices (management style, tools, workload) as a potential internal best-practice model for other departments.