YouTube-Channel-Performance-Analysis

This project analyzes the performance of YouTube videos based on various metrics, such as views, watch time, revenue, subscriber growth, and audience engagement. The goal is to uncover trends and actionable insights that help optimize content strategies for YouTube creators.

Project Overview

This repository contains an in-depth analysis of YouTube video performance, focusing on metrics such as:

• Views and Revenue: Understanding the correlation between views and estimated revenue.
• Watch Time and Subscribers: Exploring how watch time impacts subscriber growth.
• Day of the Week Performance: Identifying the best days to publish videos for maximum engagement and revenue.
• Audience Engagement: Visualizing metrics like likes, dislikes, comments, shares, and subscriber activity.

Key insights include:

1. Higher views lead to higher revenue.
2. Longer watch times result in higher subscriber growth.
3. Thursdays tend to perform best in terms of average revenue generated.
4. Detailed clustering reveals high and low-performing video groups based on key metrics.

Features and Visualizations

Correlation Analysis:

• Scatter plots showing:
• Views vs. Revenue
• Watch Time vs. Subscribers Insights into how key metrics relate to each other.

Day of the Week Analysis:

Bar Charts:

• Total revenue and engagement metrics (likes, comments, shares) by day of the week. Box Plots:
• Distribution of subscribers gained and lost for each day.

Clustering of Video Performance:

Clustering videos based on:

• Views
• Watch Time (hours)
• Estimated Revenue
• Subscribers Gained

Results show distinct clusters of high, moderate, and low-performing videos.

Engagement Metrics:

• Separate graphs for each day of the week showing:
• Total New Subscribers and Unsubscribes.

Technologies Used

• Python: For data analysis and visualization. Libraries:
• pandas for data manipulation.
• matplotlib and seaborn for visualizations.
• scikit-learn for clustering analysis.

Key Visualizations

Scatter Plots:

• Visualizing relationships between views, revenue, watch time, and subscribers.

Box Plots:

• Subscriber activity by day of the week.

Bar Charts:

• Engagement metrics (likes, dislikes, comments) across days of the week.

Clustering Graphs:

• Insights into video performance clusters.