YouTube Search Scraper - Videos, Channels & Views

Search public YouTube videos by keyword and export structured titles, watch URLs, video IDs, channels, displayed views, publication age, duration, and thumbnail URLs. Run one query or a bounded batch without a YouTube Data API key.

What data can I extract?

• Video title, ID, and canonical watch URL
• Channel name and channel URL when available
• Displayed view count and relative publication age
• Video duration and thumbnail URL

Input example

{
  "search_query": "small business marketing",
  "max_items": 25
}

Use search_queries for up to 25 keyword searches. max_items is a hard run-wide result ceiling.

Output example

{
  "title": "Small Business Marketing Strategy",
  "video_url": "https://www.youtube.com/watch?v=dQw4w9WgXcQ",
  "video_id": "dQw4w9WgXcQ",
  "channel_name": "Example Channel",
  "channel_url": "https://www.youtube.com/@example",
  "views": "1.2M views",
  "published": "2 years ago",
  "duration": "12:34",
  "thumbnail_url": "https://i.ytimg.com/vi/dQw4w9WgXcQ/hqdefault.jpg"
}

Common use cases

• Discover videos and channels for a keyword
• Research content, competitors, creators, and search-result coverage
• Build seed lists for media monitoring or later enrichment
• Give an AI agent current public YouTube search context

Use with AI agents and MCP

Example agent intent:

Find 50 public YouTube videos about warehouse automation and return title, URL, channel, views, publication age, and duration.

{
  "search_query": "warehouse automation",
  "max_items": 50
}

The strict input and dataset schemas let Apify MCP clients inspect the callable contract before execution.

Pricing and cost control

The suggested launch price is $0.003 per result. Output charges are approximately $0.30 for 100 videos or $3.00 for 1,000, plus any platform charges shown by Apify. The live Store pricing is authoritative. Use max_items to bound spend.

Reliability

The Actor scrolls YouTube search results, deduplicates by video ID, validates required fields, and is enrolled in the fleet's selector monitoring and staged self-healing process.

Limitations and responsible use

• Extracts public search-result metadata, not transcripts or comments.
• View counts and publication times are display strings, not normalized historical metrics.
• YouTube can vary results by locale, session, and upstream experiments.
• Use the data in accordance with applicable law and platform terms.

FAQ

Does this Actor need a YouTube API key?

No.

Can I search multiple keywords?

Yes. Supply search_queries; duplicates are removed and max_items limits total output.

Does it download videos or audio?

No. It returns public search metadata and links only.

See CHANGELOG.md for maintained schema changes.