YouTube Scraper (Legacy)

Legacy actor. For new runs, use YouTube Research & Transcript Scraper: lentic_clockss/youtube-research-scraper.

This older actor stays public only because Apify does not allow paid actors to be unpublished until monetization is cancelled. New development and new pricing live on the replacement actor.

Search YouTube, scrape channel video lists, and optionally enrich results with detailed metadata or transcripts. No YouTube Data API key required.

This Actor is designed for three common jobs:

• Search YouTube by keyword and export a clean dataset
• Scrape videos from one or more channels
• Pull transcripts for specific videos when captions are available

It supports three input modes from the current input schema:

• searchQueries
• channelUrls
• videoUrls

Optional heavier modes:

• scrapeDetails: fetches richer per-video metadata
• includeTranscript: attempts transcript extraction with timestamps

What this Actor is good for

1. Fast topic research

Use searchQueries when you want a spreadsheet of videos for a topic such as python automation, real estate investing, or home workout.

Typical output includes:

• videoId
• title
• url
• channelName
• channelId
• viewCount
• duration
• durationSeconds
• publishedText
• description
• thumbnailUrl
• source
• scrapedAt

This is the lightest mode and the best starting point if you want broad coverage quickly.

{
  "searchQueries": ["python automation"],
  "maxResults": 100
}

2. Channel monitoring and library export

Use channelUrls to pull videos from one or more channels. This is useful for competitor tracking, content audits, and scheduled snapshots.

{
  "channelUrls": ["https://www.youtube.com/@freecodecamp"],
  "maxResults": 200
}

If you need fields such as exact publish date, category, or fuller video metadata, turn on scrapeDetails.

3. Targeted transcript extraction

Use videoUrls when you already know which videos you care about and want transcripts or richer metadata for those specific items.

{
  "videoUrls": ["https://www.youtube.com/watch?v=XVv6mJpFOb0"],
  "includeTranscript": true,
  "transcriptLanguage": "en"
}

Transcript extraction depends on captions being available for the video. Videos without captions will return empty transcript fields.

Important runtime notes

This Actor has clear feature tiers in practice:

• Base search and channel listing are the lightest runs
• scrapeDetails is a heavier enrichment step
• includeTranscript is the heaviest mode and is best used selectively

If you need transcripts, the most reliable and economical workflow is usually:

1. run a lighter search or channel scrape first
2. identify the subset of videos you actually need
3. run transcript extraction on those specific videoUrls

That keeps runtime more predictable and avoids using transcript mode on large batches that do not need it.

Pricing and expectations

Pricing is controlled by the live Apify Store configuration, not by this README.

Two practical points matter:

• base result collection is lighter than detail or transcript enrichment
• transcript-heavy runs are materially more expensive than plain metadata collection

So this README does not promise that transcripts are free, and it does not give fixed cost examples for large transcript runs. If you are planning a bigger workflow, start with a small sample and validate runtime, output shape, and cost on your own use case first.

Input schema

The current input fields are:

Field	Type	Default	Notes
searchQueries	array	["web scraping tutorial"]	Keyword searches. Each query returns up to maxResults videos.
channelUrls	array	—	Supports @handle, /channel/ID, and /c/name formats.
videoUrls	array	—	Best for targeted metadata or transcript extraction.
maxResults	integer	50	Applies to searchQueries and channelUrls. Max allowed by schema is 500.
scrapeDetails	boolean	false	Fetches richer metadata and increases runtime.
includeTranscript	boolean	false	Attempts transcript extraction and increases runtime further.
transcriptLanguage	string	"en"	Preferred transcript language. Falls back if unavailable.

Example outputs

Base result

{
  "videoId": "XVv6mJpFOb0",
  "title": "Automate Everything with Python",
  "url": "https://www.youtube.com/watch?v=XVv6mJpFOb0",
  "channelName": "Tech With Tim",
  "channelId": "UC4JX40jDee_tINbkjycV4Sg",
  "viewCount": 1200000,
  "duration": "45:12",
  "durationSeconds": 2712,
  "publishedText": "3 months ago",
  "description": "In this video we'll automate...",
  "thumbnailUrl": "https://i.ytimg.com/vi/XVv6mJpFOb0/maxresdefault.jpg",
  "source": "search",
  "scrapedAt": "2026-04-17T10:00:00Z"
}

With details enabled

{
  "videoId": "dQw4w9WgXcQ",
  "title": "How Large Language Models Work",
  "channelName": "3Blue1Brown",
  "viewCount": 8420000,
  "publishedAt": "2026-01-15T18:00:00Z",
  "category": "Education",
  "durationSeconds": 1591,
  "description": "A visual walkthrough of the transformer architecture..."
}

With transcript enabled

{
  "videoId": "XVv6mJpFOb0",
  "title": "How Large Language Models Work",
  "transcript": [
    {
      "start": 0.0,
      "duration": 3.2,
      "text": "The transformer architecture has fundamentally changed"
    },
    {
      "start": 3.2,
      "duration": 2.1,
      "text": "how we think about sequence modeling."
    }
  ],
  "transcriptLanguage": "en",
  "transcriptText": "The transformer architecture has fundamentally changed how we think about sequence modeling."
}

Common use cases

• Topic research for content, SEO, and market scanning
• Competitor channel monitoring
• Building video datasets for downstream analysis
• Pulling captions for a selected set of videos
• Sending YouTube metadata into spreadsheets, automations, or LLM workflows

What this Actor does not promise

• It does not guarantee transcripts for every video
• It does not guarantee one fixed price across light and heavy run modes
• It does not replace YouTube's official API for every production scenario

This Actor is best used when you want flexible scraping without managing YouTube API quota, and when you understand that transcript and detail enrichment are heavier modes than plain metadata collection.

FAQ

Q: Can I get transcripts for any video?
A: Only when captions are available for that video. Auto-generated and manual captions may both work, but some videos will return empty transcript fields.

Q: What language codes can I use?
A: Set transcriptLanguage to the preferred code such as en, es, ja, de, or fr. If that language is unavailable, the Actor falls back to another available language.

Q: Should I turn on transcripts for a large search?
A: Usually no. The better workflow is to search first, shortlist videos, then run transcript extraction on that subset.

Q: Can I export the results?
A: Yes. Output is available in Apify datasets and can be consumed as JSON, CSV, Excel-compatible exports, or through downstream automation tools.

Developer notes

This Actor is MCP-friendly and can also be used from Apify integrations such as Make, Zapier, or n8n.

Example payload:

{
  "searchQueries": ["home workout"],
  "maxResults": 20,
  "scrapeDetails": true,
  "includeTranscript": false
}

For transcript-heavy workflows, prefer a second targeted run over a single large all-in-one run.