This project takes a data-driven look at how Tech and Sports channels perform on YouTube. Using Python for cleaning and exploration, SQL for structured analysis, and Power BI for interactive visualization, it compares the two categories across subscribers, views, and video output — turning raw channel data into clear, actionable business insights.
Whether you're a data analyst exploring the pipeline or a recruiter skimming for skills, this repo demonstrates a complete end-to-end analytics workflow: raw data → cleaning → EDA → SQL → dashboard → insights.
- Objectives
- Dataset
- Tools & Technologies
- Data Cleaning
- Exploratory Data Analysis
- SQL Analysis
- Power BI Dashboard
- Key Business Insights
- Project Structure
- How to Run
- Future Improvements
- Author
- 🧹 Clean and preprocess multiple raw YouTube datasets
- 🔍 Perform Exploratory Data Analysis (EDA) to uncover patterns
- 🗄️ Analyze channel performance using structured SQL queries
- 📊 Build an interactive, decision-ready Power BI dashboard
- 💡 Translate findings into clear business insights
| Dataset | Description |
|---|---|
| Tech Channels | Channel-level metadata for top tech creators |
| Tech Videos | Video-level statistics for tech channels |
| Sports Channels | Channel-level metadata for top sports creators |
Raw data lives in
data/raw/, cleaned outputs are saved todata/cleaned/.
| Category | Stack |
|---|---|
| Language | Python |
| Data Handling | Pandas, NumPy |
| Visualization | Matplotlib, Power BI |
| Database | SQLite, SQL |
| Environment | Jupyter Notebook, VS Code |
Every dataset was put through a consistent preprocessing pipeline before analysis:
- ✅ Removed duplicate records
- ✅ Handled missing values
- ✅ Converted and standardized date columns
- ✅ Standardized column naming conventions
- ✅ Corrected inconsistent data types
- ✅ Exported cleaned, analysis-ready datasets
Key questions explored through Python-based EDA:
- 🏆 Which channels lead in subscribers?
- 👀 Which channels lead in total views?
- 📊 How are subscribers distributed across each category?
- 🎥 What do video-count statistics reveal?
- 🔗 How correlated are subscribers, views, and video count?
- ⚖️ How does Tech stack up against Sports overall?
17 SQL queries were written to dig into channel performance from every angle, including:
- Top-performing channels by subscribers and views
- Average subscriber counts by category
- Country-wise channel breakdowns
- Video count and upload statistics
- Channel rankings and aggregate performance metrics
All queries are available in the
sql/directory, executed against a local SQLite database.
An interactive dashboard was built to make the analysis explorable at a glance:
- 🔢 KPI Cards — subscribers, views, and video counts at a glance
- 🏅 Top 10 Tech Channels — ranked performance view
- ⚖️ Tech vs Sports Comparison — side-by-side category breakdown
- 🔬 Scatter Plot Analysis — subscribers vs. views correlation
- 🎚️ Interactive Slicers — filter by category, country, and more
- 💬 Business Insights Panel — key takeaways embedded in the dashboard
🏟️ Sports channels collectively command a far larger subscriber base than tech channels — reflecting sports' broader, mainstream appeal.
🎯 A small number of tech channels drive a disproportionate share of total tech subscribers — the category is more concentrated at the top.
📈 Higher subscriber counts generally track with higher total views, reinforcing subscriber count as a solid proxy for reach.
👥 Sports content attracts broader, more general audiences, while tech content serves a smaller but highly engaged niche community.
youtube_data_analysis_project/
│
├── dashboard/ # Power BI (.pbix) dashboard file
├── data/
│ ├── raw/ # Original, unprocessed datasets
│ └── cleaned/ # Cleaned, analysis-ready datasets
├── images/ # Dashboard screenshots & visual assets
├── notebooks/ # Jupyter notebooks for cleaning & EDA
├── sql/ # SQL scripts and queries
├── README.md
├── requirements.txt
└── LICENSE
# 1. Clone the repository
git clone https://github.com/<your-username>/youtube_data_analysis_project.git
cd youtube_data_analysis_project
# 2. Install dependencies
pip install -r requirements.txt
# 3. Open the notebooks
jupyter notebook notebooks/
# 4. Explore the dashboard
# Open dashboard/*.pbix in Power BI Desktop- 🔌 Integrate the live YouTube Data API for real-time metrics
- 🔄 Automate scheduled data refresh and updates
- 🤖 Build predictive models to forecast channel growth
- ☁️ Deploy the dashboard online for public access
Vaishnavi Wangalwar
If you found this project useful or interesting, consider ⭐ starring the repo!