YouTube Search Scraper

Search YouTube by keyword and stream back structured results — videos, channels, and playlists — with built-in filters for upload date, view count, content type, and live/shorts.

Why use this actor

• No account / no API key required — just give it a query and the actor returns the same results YouTube shows on the web.
• Filter the way YouTube does — sort by relevance, upload date, view count, or rating; restrict to videos / channels / playlists; or narrow to today's or this week's uploads, live streams, or shorts.
• Mixed result types in one stream — videos, channels, and playlists are returned together, each tagged with a type discriminator so you can split them downstream.
• No quota limits — unlike the official YouTube Data API, runs are not capped by a daily quota; pay only for the results you receive.
• Bulk queries in one run — pass a list of search terms; every record is tagged with _input and _filter so you can join results back to your input list.
• Stable JSON output — every row carries _input, _filter, _source, _scrapedAt envelope fields, ready for pipelines, spreadsheets, and databases.

How it works

1. You provide one or more search queries (e.g. twice, python tutorial) and pick a filter.
2. The actor fetches search results the same way YouTube's web app does, then paginates with continuation tokens until it has gathered maxResults items per query.
3. Results stream into your dataset as flat JSON records, ready to download as JSON, CSV, or Excel.

You do not need to manage scrapers, browsers, or rotating IPs — all handled internally.

Input

{
  "queries": [
    "lofi hip hop",
    "python tutorial"
  ],
  "maxResults": 20,
  "filter": "relevance",
  "proxyConfiguration": {
    "useApifyProxy": true,
    "apifyProxyGroups": ["DATACENTER"]
  }
}

Field	Type	Description
queries	array	One or more search queries to run. Each query is processed independently.
maxResults	integer	Maximum results returned per query. Pagination is automatic. Default: 20. Set to 0 for unlimited.
filter	string	Filter / sort mode. See table below. Default: relevance.
proxyConfiguration	object	Apify Proxy settings. Datacenter is the default and works well for search.

filter options

Value	What it does
relevance	Default YouTube ranking (best match for the query).
upload_date	Sort by newest upload first.
view_count	Sort by most viewed first.
rating	Sort by highest-rated first.
video_only	Restrict results to videos.
channel_only	Restrict results to channels — useful for influencer discovery.
playlist_only	Restrict results to playlists.
live	Currently live streams only.
short	Videos under 4 minutes (catches most Shorts).
today	Uploaded in the last 24 hours.
this_week	Uploaded in the last 7 days.

Output

Input: queries: ["twice"], maxResults: 5, filter: "relevance" — the actor returns a mixed stream of videos and channels:

{
  "type": "video",
  "videoId": "i0p1bmr0EmE",
  "title": "TWICE \"What is Love?\" M/V",
  "channelTitle": "JYP Entertainment",
  "channelId": "UCaO6TYtlC8U5ttz62hTrZgg",
  "publishedTimeText": "8 years ago",
  "lengthText": "3:44",
  "viewCountText": "918,997,637 views",
  "descriptionSnippet": "TWICE(트와이스) \"What is Love?\" M/V Spotify https://goo.gl/jVLYYY iTunes & Apple Music https://goo.gl/DKyKZf Google Music ...",
  "thumbnails": [
    {
      "url": "https://i.ytimg.com/vi/i0p1bmr0EmE/hqdefault.jpg?sqp=-oaymwEcCOADEI4CSFXyq4qpAw4IARUAAIhCGAFwAcABBg==&rs=AOn4CLA8eBN8cI0jbqpUfPBbCVH7jph99A",
      "width": 480,
      "height": 270
    }
  ],
  "_input": "twice",
  "_filter": "relevance",
  "_source": "S1-primary",
  "_scrapedAt": "2026-05-18T10:42:57.499074+00:00"
}

{
  "type": "channel",
  "channelId": "UCzgxx_DM2Dcb9Y1spb9mUJA",
  "title": "TWICE",
  "descriptionSnippet": "TWICE Official YouTube Channel.",
  "subscriberCountText": "19.3M subscribers",
  "thumbnails": [
    {
      "url": "https://yt3.googleusercontent.com/rj-m7CQIV8hIHx_lB8cjs0NxrDFJaUkVwZ5tCnZTNSVEr2IqXGF-2e7nr7og1IGeRXtHikce=s176-c-k-c0x00ffffff-no-rj-mo",
      "width": 176,
      "height": 176
    }
  ],
  "_input": "twice",
  "_filter": "relevance",
  "_source": "S1-primary",
  "_scrapedAt": "2026-05-18T10:42:57.499074+00:00"
}

{
  "type": "video",
  "videoId": "FxkAkEaalB8",
  "title": "TWICE (트와이스) PLAYLIST 2025 [UPDATED]",
  "channelTitle": "FEARLIX",
  "channelId": "UCqNQp_VOtVjQBOR_OhrizMA",
  "publishedTimeText": "10 months ago",
  "lengthText": "2:52:07",
  "viewCountText": "765,898 views",
  "descriptionSnippet": "TWICE (트와이스) is a nine-member girl group under JYP Entertainment, Formed through the reality show SIXTEEN. Group ...",
  "_input": "twice",
  "_filter": "relevance",
  "_source": "S1-primary",
  "_scrapedAt": "2026-05-18T10:42:57.499074+00:00"
}

… 2 more results in the same run.

Common fields (every row)

Field	Type	Description
_input	string	The query exactly as you supplied it. Use this to join results back to your input list.
_filter	string	The filter value used for this run (e.g. relevance, upload_date).
_source	string	Internal tag for the fetch path. S1-primary means the fastest, richest path; S2-* would indicate a fallback.
_scrapedAt	string	ISO-8601 UTC timestamp when the record was scraped.
type	string	Result type discriminator: video, channel, or playlist.

Video fields (type: "video")

Field	Type	Description
videoId	string	YouTube video ID (the part after watch?v=).
title	string	Video title.
channelTitle	string	Display name of the channel that uploaded the video.
channelId	string	Channel ID (starts with UC...).
publishedTimeText	string	YouTube's localized upload time (e.g. "8 years ago", "3 days ago").
lengthText	string	Video duration as displayed (e.g. "3:44", "2:52:07").
viewCountText	string	View count as displayed (e.g. "918,997,637 views").
descriptionSnippet	string	Short description snippet shown under the result.
thumbnails	array	Thumbnail URLs with width and height.

Channel fields (type: "channel")

Field	Type	Description
channelId	string	Channel ID (starts with UC...).
title	string	Channel display name.
descriptionSnippet	string	Short channel description.
subscriberCountText	string	Subscriber count as displayed (e.g. "19.3M subscribers").
thumbnails	array	Channel avatar URLs with width and height.

Playlist fields (type: "playlist")

Field	Type	Description
playlistId	string	YouTube playlist ID.
title	string	Playlist title.
videoCount	string	Number of videos in the playlist.
channelTitle	string	Display name of the channel that owns the playlist.

Error envelope

Queries that fail to fetch return a structured error instead of crashing the run:

{
  "_input": "some query",
  "_error": "fetch_failed",
  "_errorDetail": "network error or upstream rejection",
  "_source": "S1-primary",
  "_scrapedAt": "2026-05-18T10:42:57.499074+00:00"
}

Filter on _error to triage failed rows.

Pricing

This actor is billed per result: $3.50 per 1,000 results. Each video, channel, or playlist record = 1 result. Errors are not billed.

Other Sosmed Actors

Platform	Actor	Best for
YouTube	YouTube Channel Scraper	Pull a channel's full upload list + metadata
YouTube	YouTube Video Detail Scraper	Full metadata, description, and stats for any video
Bluesky	Bluesky Search Scraper	Keyword search across the Bluesky timeline
Reddit	Reddit Search Scraper	Search posts across all subreddits
X / Twitter	X Search Scraper	Keyword search across X posts
Tumblr	Tumblr Blog Scraper	Pull posts from any Tumblr blog

Browse the full catalog at apify.com/xtracto.

Notes

• Result freshness — YouTube ranks search results dynamically; the same query rerun a few minutes later may return a slightly different order. Use _scrapedAt to track when each row was captured.
• Pagination — the actor chains continuation tokens automatically until maxResults is reached or YouTube returns no more pages. Very large maxResults values for narrow queries may end early simply because YouTube has no more results to give.
• Why filters help — relevance is broad and noisy for popular keywords. For trend monitoring, today or this_week give you only fresh uploads; for influencer discovery, channel_only strips out the videos; for view-driven research, view_count puts the biggest hits first.
• Counts as text — viewCountText, subscriberCountText, and publishedTimeText are returned as YouTube's localized strings ("918,997,637 views", "19.3M subscribers", "3 days ago"). Parse these client-side if you need numeric values.
• No deduplication — the actor does not deduplicate by videoId / channelId / playlistId across pages; downstream pipelines should dedupe if strict uniqueness is required.