YouTube Outlier Detection

Goal

Find high-performing videos in your niche for competitive intelligence and content inspiration.

Scripts

• ./scripts/scrape_youtube_outliers.py - Scrape and score outliers
• ./scripts/update_transcripts.py - Fetch transcripts for existing outliers

Usage

# Default run
python3 ./scripts/scrape_youtube_outliers.py

# Custom parameters
python3 ./scripts/scrape_youtube_outliers.py \
  --keywords "AI automation,AI agents" \
  --days 30 \
  --min_score 1.5 \
  --limit 50

How It Works

1. Searches YouTube for keywords
2. Calculates outlier score (views / channel average)
3. Applies recency boost
4. Fetches transcripts
5. Generates Claude summaries
6. Saves to Google Sheet

Output

Google Sheet with:

• Outlier Score, Views, Duration
• Title, Video Link, Channel
• Thumbnail, Summary, Transcript
• Publish Date

vs Cross-Niche Outliers

• youtube_outliers: Your core niche (daily monitoring)
• cross-niche-outliers: Adjacent niches (weekly inspiration)

Environment

APIFY_API_TOKEN=your_token
ANTHROPIC_API_KEY=your_key

---

Schema

Inputs

Name	Type	Required	Description
keywords	string	No	Comma-separated keywords (default: niche keywords)
days	integer	No	Lookback period in days (default: 30)
min_score	float	No	Minimum outlier score (default: 1.5)
limit	integer	No	Max results (default: 50)

Outputs

Name	Type	Description
sheet_url	string	Google Sheet with outlier videos

Credentials

Name	Source
APIFY_API_TOKEN	.env
ANTHROPIC_API_KEY	.env

Composable With

Skills that chain well with this one: title-variants, recreate-thumbnails, cross-niche-outliers

Cost

Apify credits + Claude API