Youtube Video Scraper

Search YouTube and extract video metadata including titles, full descriptions, view counts, channel info, and more. Supports advanced filters for upload date, duration, sort order, and country targeting.

Key Features

• Advanced Search Filters — Filter by upload date (hour/today/week/month/year), duration (short/medium/long), and sort order (relevance/date/views/rating)
• Full Video Descriptions — Fetches complete descriptions from each video page, not just search snippets
• Multi-Keyword Search — Process multiple search keywords in a single run
• Country Targeting — Get localized results for 30+ countries
• YouTube Shorts Support — Include or exclude Shorts from results
• Automatic Deduplication — Removes duplicate videos across keyword searches
• Progress Tracking — Real-time status updates in the Apify Console

Quick Start

1. Click Try for free on the YouTube Video Scraper page
2. Enter your search keywords
3. Configure filters (optional)
4. Click Start
5. Download results in JSON, CSV, Excel, or other formats

Input Configuration

{
    "searches": ["JavaScript tutorial", "React hooks"],
    "maxResults": 50,
    "maxPerSearch": 20,
    "uploadDate": "week",
    "duration": "medium",
    "sortBy": "viewCount",
    "country": "US",
    "includeShorts": true
}

Parameter	Type	Default	Description
searches	string[]	required	Keywords to search on YouTube
maxResults	integer	50	Total maximum videos to scrape (1-500)
maxPerSearch	integer	20	Maximum videos per keyword (1-100)
uploadDate	string	"any"	Filter: any, hour, today, week, month, year
duration	string	"any"	Filter: any, short (<4min), medium (4-20min), long (>20min)
sortBy	string	"relevance"	Sort: relevance, date, viewCount, rating
country	string	"US"	Two-letter country code for localized results
includeShorts	boolean	true	Include YouTube Shorts in results

Output Data

Each video result includes:

{
    "videoId": "tDGiWn0flK8",
    "title": "From Zero to Your First Agentic AI Workflow in 26 Minutes",
    "url": "https://www.youtube.com/watch?v=tDGiWn0flK8",
    "thumbnailUrl": "https://i.ytimg.com/vi/tDGiWn0flK8/maxresdefault.jpg",
    "duration": "26:23",
    "durationSeconds": 1583,
    "channelName": "Nate Herk | AI Automation",
    "channelUrl": "https://www.youtube.com/@nateherk",
    "viewCount": 34000,
    "uploadTime": "3 days ago",
    "description": "Full video description text...",
    "isShort": false,
    "searchKeyword": "AI automation",
    "extractedAt": "2026-02-27T10:30:00.000Z"
}

Field	Type	Description
videoId	string	YouTube video ID
title	string	Video title
url	string	Clean video URL (no tracking parameters)
thumbnailUrl	string	Thumbnail image URL
duration	string	Duration in MM:SS or HH:MM:SS format
durationSeconds	integer	Duration in seconds
channelName	string	Channel name
channelUrl	string	Channel page URL
viewCount	integer	Approximate view count
uploadTime	string	Relative upload time (e.g., "3 days ago")
description	string	Full video description
isShort	boolean	Whether the video is a YouTube Short
searchKeyword	string	The keyword that found this video
extractedAt	string	ISO 8601 timestamp of extraction

Use Cases

• Market Research — Analyze trending topics and competitor content
• Content Strategy — Find inspiration and track performance metrics
• Academic Research — Collect data for social media studies
• Brand Monitoring — Track brand mentions and engagement
• SEO Analysis — Research keywords and video optimization strategies

How It Works

1. The scraper searches YouTube with your keywords and filters
2. For each search, it scrolls through results to collect video cards
3. It visits each video page to fetch the full description
4. Results are deduplicated and saved progressively to the dataset
5. The scraper survives platform migrations and can resume from where it left off

Limitations

• View counts are approximate (e.g., "1.2M" → 1200000)
• Upload times are relative ("3 days ago"), not exact dates
• YouTube may rate-limit very large scraping jobs
• Maximum 500 videos per run, 100 per keyword

FAQ

Q: Why are view counts approximate? A: YouTube search results display abbreviated view counts (e.g., "1.2M views"). The scraper converts these to numbers, which may differ slightly from exact counts.

Q: Can I search in languages other than English? A: Yes! Enter keywords in any language. The scraper interface uses English for consistency, but search results are based on your keyword language and selected country.

Q: What happens if the scraper is interrupted? A: Results are saved progressively. If the scraper is interrupted, all previously saved videos are kept in your dataset. The scraper also supports migration resume.

Q: How do I get results for a specific country? A: Use the country parameter with a two-letter country code (e.g., "TW" for Taiwan, "JP" for Japan).

---

YouTube Video Scraper — built for reliable, scalable video data extraction.

Keywords: youtube scraper, video scraper, youtube search, video metadata, youtube api alternative, apify actor, Claude Code, Gemini, Codex, OpenClaw