YouTube AI Niche Analyzer

Automated intelligence pipeline that scrapes, analyzes, and delivers daily reports on the AI/automation YouTube niche.

Built on the WAT Framework (Workflows → Agents → Tools) — where AI handles reasoning and deterministic scripts handle execution.

---

What It Does

Every day at 7:00 AM, this pipeline automatically:

1. Scrapes 500+ videos across 7 AI keywords from the YouTube Data API
2. Analyzes trending topics, engagement benchmarks, content gaps, and optimal posting strategies
3. Generates a branded 10-page PDF report with charts and dynamic insights
4. Pushes structured data to Google Sheets for further exploration
5. Emails the finished PDF report to your inbox via Gmail

All five steps run end-to-end in under 2 minutes with zero manual intervention.

---

Pipeline Overview

┌─────────────────┐     ┌─────────────────┐     ┌─────────────────┐
│   SCRAPE        │────▶│   ANALYZE       │────▶│   REPORT        │
│                 │     │                 │     │                 │
│ 7 keywords      │     │ Trending topics │     │ 10-page PDF     │
│ 14 API searches │     │ Top channels    │     │ Google Sheets   │
│ 500+ videos     │     │ Benchmarks      │     │ Gmail delivery  │
│ 450+ channels   │     │ Opportunities   │     │                 │
└─────────────────┘     └─────────────────┘     └─────────────────┘

---

Report Insights

Analysis	What You Get
Trending Topics	Top 15 rising topics ranked by view velocity and volume
Top Videos	10 highest-performing videos with engagement breakdowns
Top Channels	10 channels ranked by engagement rate (likes + comments / views)
Duration Sweet Spot	Which video length gets the most views
Best Publish Day	Day-of-week breakdown by average views
Best Publish Hour	Optimal upload time (UTC)
Title Patterns	Performance by format: how-to, listicle, question, comparison, news, tutorial
Content Opportunities	Rising topics with low competition — your next video ideas

---

Quick Start

Prerequisites

• Python 3.10+
• YouTube Data API key or Google OAuth credentials
• Google Cloud project with YouTube, Sheets, and Gmail APIs enabled

1. Clone & Install

git clone https://github.com/CodeWithArhum/Youtube-Analysis.git
cd Youtube-Analysis
pip install -r requirements.txt

2. Configure Environment

Create a .env file in the project root:

YOUTUBE_API_KEY=your_youtube_api_key_here
GOOGLE_SHEETS_ID=your_google_sheets_id_here

Place your Google OAuth client_secret*.json file in the project root for Sheets and Gmail access.

3. Run the Pipeline

# Full pipeline (scrape → analyze → PDF → sheets → email)
python tools/run_pipeline.py

# Or run individual steps
python tools/youtube_scraper.py
python tools/analyze_data.py
python tools/build_pdf.py
python tools/push_to_sheets.py
python tools/send_email.py

4. Schedule Daily Runs (Windows)

# Creates a Windows Task Scheduler job for 7:00 AM daily
setup_schedule.bat

---

Project Structure

Youtube-Analysis/
│
├── tools/                  # Deterministic Python scripts (execution layer)
│   ├── youtube_scraper.py  # YouTube Data API collector
│   ├── analyze_data.py     # Data processor & trend analyzer
│   ├── build_pdf.py        # Branded PDF report generator
│   ├── build_slides.py     # Google Slides report builder
│   ├── push_to_sheets.py   # Google Sheets uploader
│   ├── send_email.py       # Gmail sender with PDF attachment
│   └── run_pipeline.py     # Full pipeline orchestrator
│
├── workflows/              # Markdown SOPs (instruction layer)
│   ├── scrape_youtube.md   # Scraping procedure & error handling
│   ├── analyze_data.md     # Analysis methodology & computations
│   └── build_report.md     # Report generation & delivery steps
│
├── .tmp/                   # Intermediate files (gitignored)
│   ├── raw_youtube_data.json
│   ├── analysis_results.json
│   └── ai_youtube_report.pdf
│
├── setup_schedule.bat      # Windows Task Scheduler setup
├── requirements.txt        # Python dependencies
├── .env                    # API keys (gitignored)
└── CLAUDE.md               # AI agent operating instructions

---

The WAT Framework

This project is built on the Workflows, Agents, Tools architecture:

Layer	Role	Location
Workflows	Human-readable SOPs that define what to do	workflows/
Agents	AI reasoning layer that orchestrates execution	Claude / LLM
Tools	Deterministic scripts that do the work	tools/

Why? When AI handles every step directly, accuracy compounds downward — 90% per step means 59% after five. By offloading execution to deterministic scripts, the AI focuses on orchestration where it excels, and the scripts deliver consistent, testable results.

---

API Usage

API	Units Per Run	Daily Limit
YouTube Data API v3	~1,430	10,000
Google Sheets API	~20 requests	300/min
Gmail API	1 email	100/min

Safe to run 5-7x per day with overhead for other usage.

---

License

This project is for personal and educational use.

---

Built by @CodeWithArhum