🎬 YouTube Data Collector

Paste a YouTube video, channel, playlist, search URL, hashtag, video ID, channel ID, playlist ID, @handle, or keyword. Choose the result package you want and a size. The Actor detects every input automatically and separates the output into clean datasets.

🚀 Quick Start

1. Paste one or more items into 📥 Paste what you have.
2. Keep 🎯 Choose result package on 🧲 Complete if you simply want everything from each input.
3. Pick 📦 Choose size.
4. Run the Actor.

Most users only need those three fields. You never need to choose video, channel, playlist, or search manually: source detection is automatic.

🎯 Result Packages

Package	Use when	Collects
🧲 Complete	You want the full dataset from whatever you pasted	Videos, channels, playlists, Shorts, streams, posts, transcripts, comments, replies, contacts, insights, media, diagnostics
⚡ Quick	You want a fast clean table first	Lightweight metadata only, without comments, transcripts, media expansion, or contact crawling
🧠 Research	You compare topics, keywords, and engagement	Search across videos, channels, and playlists with content insights and enrichment
💬 Comments	You monitor audience feedback	Flat comments and replies, newest/top sorting, continuation tokens, dedupe across scheduled runs
📧 Leads	You build creator prospect lists	Public emails, websites, social links, contact links, external-page enrichment, and validation
📝 Transcripts	You build AI/RAG datasets	Transcript text, timestamped segments, language fallback, match terms, and compact transcript output
🎞️ Media	You audit technical/video extras	Media formats, chapters, SponsorBlock segments, related videos, live chat/replay when public

📥 Accepted Inputs

You can mix input types in the same run.

Input	Example
Video URL	https://www.youtube.com/watch?v=wwSzpaTHyS8
Shorts URL	https://www.youtube.com/shorts/...
Video ID	dQw4w9WgXcQ
Channel handle	@NASA
Channel ID	UC...
Playlist URL	https://www.youtube.com/playlist?list=...
Search URL	https://www.youtube.com/results?search_query=...
Hashtag	https://www.youtube.com/hashtag/apify
Keyword	web scraping tutorial

📦 Size Presets

Size	Best for	Default limits
test	Fast validation	Tiny sample, minimal comments and replies
small	First real run	Recommended first production run
medium	Research dataset	More items, comments, replies, and enrichment
large	Big export	Large but capped export
unlimited	Maximum collection	No global result-row cap

Advanced API users can still override exact limits with hidden fields such as maxResults, maxItems, maxComments, and maxRepliesPerComment.

📤 Output

The Actor writes separate datasets so the Apify Output tab stays readable. The first output is Videos, not a mixed table, so channel, playlist, post, comment, and diagnostic rows do not appear as odd rows inside a video table.

Dataset	Contains
Videos	Video, Shorts, stream, and movie rows only
Channels	Channel profiles from channel inputs and channel search results
Playlists / search	Playlist, show, and other non-video search entity rows
Posts	Community post rows from channel Community/Posts tabs
Transcripts	Rows with transcript text, subtitle status, or transcript keyword matches
Contacts	Rows with public emails, websites, socials, or contact links
Media	Rows with media formats, SponsorBlock, chapters, live chat, or related videos
Comments	One flat row per comment or reply
Diagnostics	Invalid inputs, unavailable videos, empty sources, and source failures
All records	Compatibility dataset with every non-comment, non-diagnostic main row

The same clean row can appear in more than one focused dataset. For example, a video with transcript and media data appears in Videos, Transcripts, and Media; it is still one billed main result.

🧾 Canonical Fields

Meaning	Field
Video/channel/search URL	url
Source page that produced the item	sourceUrl
Views	views
Likes	likes
Subscribers	subscribers
Video description	description
Comment text	text in the Comments dataset
Transcript text	transcriptText
Creator emails	emails

Legacy raw aliases can still be returned with the hidden includeRawFields API option.

🧩 Example Inputs

Complete mixed run

{
  "targets": [
    "https://www.youtube.com/watch?v=wwSzpaTHyS8",
    "@NASA",
    "web scraping tutorial"
  ],
  "scenario": "complete",
  "runSize": "test"
}

Everything from a video

{
  "targets": ["https://www.youtube.com/watch?v=wwSzpaTHyS8"],
  "scenario": "complete",
  "runSize": "small"
}

Everything from a channel

{
  "targets": ["@NASA"],
  "scenario": "complete",
  "runSize": "small"
}

Fast metadata only

{
  "targets": ["@NASA", "web scraping tutorial"],
  "scenario": "quick_metadata",
  "runSize": "small"
}

Research keywords and engagement

{
  "targets": ["web scraping tutorial"],
  "scenario": "research_insights",
  "runSize": "small"
}

Comment monitoring

{
  "targets": ["https://www.youtube.com/watch?v=wwSzpaTHyS8"],
  "scenario": "comments_monitoring",
  "runSize": "small",
  "stateKey": "my-video-comments"
}

AI transcripts

{
  "targets": ["https://www.youtube.com/watch?v=wwSzpaTHyS8"],
  "scenario": "transcripts_for_ai",
  "runSize": "small",
  "subtitlesLanguages": ["en", "any"],
  "subtitlesFormat": "plaintext"
}

Creator leads

{
  "targets": ["@NASA", "ai automation agency"],
  "scenario": "creator_leads",
  "runSize": "small",
  "emailValidation": "syntax"
}

Media audit

{
  "targets": ["https://www.youtube.com/watch?v=wwSzpaTHyS8"],
  "scenario": "media_audit",
  "runSize": "small"
}

🧠 Feature Coverage

Feature	Status
Video metadata	Included
Channel about data	Included
Channel profile output	Included
Playlists	Included
Search results	Included
Shorts	Included
Streams	Included
Community posts	Included
Comments and replies	Included
Continuation tokens	Included
Dedupe across scheduled runs	Included
Transcripts and captions	Included
Language fallback	Included
Transcript search	Included
SponsorBlock segments	Included
Related videos	Included
Chapters	Included
Most-replayed heatmap	Included
Music credits	Included
Creator contacts	Included
Content insights	Included
Live chat and replay chat	Best effort when YouTube exposes public continuation
Media format summaries	Included
Entity-separated output datasets	Included
Large payload KVS fallback	Included
Diagnostics dataset	Included

🛠️ API Compatibility

The old jobPreset field and legacy scenario values such as auto, video_full, channel_full, playlist_full, and search_full are still accepted for existing integrations. New Console runs should use the visible result packages in scenario.

Legacy inputs such as startUrls, searchQueries, youtubeHandles, videoIds, channelIds, playlistIds, and extraData still work.

✅ Reliability

The Actor uses Chrome-like TLS requests through curl_cffi, Apify Residential Proxy on Apify Cloud, per-request proxy sessions, retry handling for transient YouTube responses, KVS fallback for large payloads, and separate diagnostics output for invalid or unavailable sources.

💾 Export

Download Videos, Channels, Playlists / search, Posts, Transcripts, Contacts, Media, Comments, Diagnostics, or All records from Apify in JSON, CSV, Excel, XML, RSS, or HTML, or consume them through Dataset API endpoints.