YouTube Search Scraper

Scrape YouTube search results by keyword — no API key, no login, no quota. Give it a list of search queries and get back clean, structured video data: title, channel, view count, duration, published date, thumbnail and a direct watch URL. Built for video discovery, keyword monitoring and trend tracking at scale.

This actor talks to YouTube's internal innertube endpoint (the same one the website uses), so it is fast and datacenter-proxy friendly — you don't need expensive residential proxies to run it.

Features

• 🔎 Search by keyword — pass any number of queries, each scraped independently.
• 📄 Deep pagination — up to 1,000 videos per query via continuation tokens.
• 🎯 Clean output — one row per video with parsed integer view counts.
• 🔑 No API key — nothing to register, no Google quota.
• 💸 Cheap — datacenter proxies are enough; ~$0.5 per 1,000 videos.
• 🧱 Robust — curl_cffi (Chrome TLS fingerprint) with a urllib fallback.

Use cases

• Video discovery — surface the top videos for any topic or niche.
• Competitor & keyword monitoring — track who ranks for your target keywords.
• Content research — find high-performing formats, titles and channels.
• Trend tracking — watch how results shift over time for a query.

Input

Field	Type	Description
mode	select	Scraping mode. Currently search (keywords → videos).
queries	array	List of keyword search queries. Each is scraped up to maxItems.
maxItems	integer	Max videos per query (default 100, max 1000).

Example:

{
  "mode": "search",
  "queries": ["lofi hip hop", "python tutorial"],
  "maxItems": 100
}

Output

Each dataset item is a single video:

{
  "query": "lofi hip hop",
  "video_id": "n61ULEU7CO0",
  "title": "lofi hip hop radio - beats to relax/study to",
  "channel": "Lofi Girl",
  "channel_url": "https://www.youtube.com/channel/UCSJ4gkVC6NrvII8umztf0Ow",
  "duration": "6:10:58",
  "views": 56023605,
  "views_text": "56,023,605 views",
  "published_time": "4 years ago",
  "thumbnail": "https://i.ytimg.com/vi/n61ULEU7CO0/hq720.jpg",
  "url": "https://www.youtube.com/watch?v=n61ULEU7CO0",
  "source": "youtube",
  "scraped_at": "2026-07-20T15:00:00+00:00"
}

views is a best-effort integer ("1.2M views" → 1200000); the raw string is always kept in views_text. All fields are nullable.

Pricing

Roughly $0.5 per 1,000 videos on Apify's platform, depending on proxy usage. Search runs cheaply on datacenter proxies.

Related actors

• YouTube Channel Scraper
• YouTube Comments Scraper
• YouTube Keyword Scraper
• YouTube Transcript Scraper

Keywords

youtube scraper, youtube search scraper, youtube video scraper, scrape youtube search results, youtube keyword scraper, video discovery, youtube api alternative, youtube data extraction, youtube monitoring, youtube trends, no api key.

Disclaimer

Only scrapes publicly available search results. Respect YouTube's Terms of Service and applicable laws when using scraped data.