YouTube Performance 📊

This dbt project transforms raw YouTube API data from a central analytics data source into clean, tested, and documented data models for analyzing video performance, content types, and viewer engagement across multiple channels.

---

📦 Project Overview

This warehouse powers reporting, dashboards, and analysis that answer key questions like:

• Which videos or content types drive the most engagement?
• How do tags, promos, or highlights perform across channels?
• What videos are performing best by upload day, time, or topic?

---

🔧 Data Sources

This project is built on top of a single data source:

• raw_youtube (via stage_youtube_api_metrics)
  • Powered by the YouTube Data API
  • Contains raw video, channel, and metric-level fields

---

🧱 Model Layers

The project follows modern dimensional modeling and dbt best practices:

models/ ├── staging/ ├── dimensions/ ├── bridge/ ├── facts/ └── reports/

---

✅ Staging Models

• stage_youtube_api_metrics: Prepares raw YouTube API data

✅ Dimension Models

• dim_video: Video metadata and flags like is_short, is_full_match
• dim_channel: Channel-level info (name, subscribers)
• dim_date: Standard date dimension for joining with fact tables
• dim_tag: Unique video tags extracted from metadata

✅ Bridge Table

• bridge_video_tag: Many-to-many relationship between videos and tags

✅ Fact Table

• fact_video_daily_performance: Core metrics like views, likes, and engagement rate

✅ Reporting Models

• report_video_daily_performance: Flattened report-friendly view of daily video metrics
• report_tag_daily_performance: Tag-level attribution of engagement metrics

---

🧪 Data Testing

Data quality is validated using dbt tests, including:

• Unique & not null constraints on surrogate keys
• Relationship tests between fact & dimension models
• Optional accepted values or derived content-type classifications

---

📄 Documentation

Generate interactive documentation using:

---

Preview details about each model, column, tests, and relationships via the dbt Docs UI.

---

🚀 Usage

Typical workflow:

dbt run # Build all models dbt test # Run all data quality tests dbt docs generate # Build documentation site dbt docs serve # Launch doc site at http://localhost:8080/

---

🏷️ Tags & Materialization

Models are organized using tags like:

• dimension, fact, bridge, report, stage

Materializations are configured in dbt_project.yml:

• Dimensions: table
• Fact/Report Tables: table
• Staging Models: view

---

🤝 Contributing

To contribute:

1. Fork the repo or work in a feature branch
2. Add your model or update existing logic
3. Run dbt test and dbt docs generate before submitting a PR

---

📬 Questions?

Feel free to reach out to Grapple Insights.

---

Made with ❤️ and dbt. Powered by the YouTube Data API.

---

Resources:

• Learn more about dbt in the docs
• Check out Discourse for commonly asked questions and answers
• Join the dbt community to learn from other analytics engineers
• Find dbt events near you
• Check out the blog for the latest news on dbt's development and best practices