YouTube Channel Scraper

Scrape YouTube channel metadata and recent videos for content research, affiliate research, competitor analysis, and AI training datasets. Built with TypeScript and Crawlee CheerioCrawler (HTTP-first, no browser) by parsing embedded ytInitialData JSON from channel About and Videos pages.

Run it on the Apify platform for scheduling, API access, proxy rotation, monitoring, and dataset exports (JSON, CSV, Excel).

Legal / ToS notice: This Actor is intended for publicly available data only. Respect YouTube Terms of Service, robots rules, and applicable privacy laws (GDPR/CCPA). Do not use this tool to bypass authentication, scrape private content, or violate platform policies. You are responsible for how you use the data.

What does YouTube Channel Scraper do?

Given one or more channel URLs (@handle, /channel/UC…, /c/, /user/), this Actor:

1. Normalizes and deduplicates channel URLs
2. Optionally loads each channel About tab for metadata (title, description, subscribers, total views, video count, country)
3. Loads the Videos tab and extracts recent video listings up to maxVideos
4. Parses window["ytInitialData"] / var ytInitialData embedded in the HTML
5. Handles consent walls, captchas / unusual traffic, and not-found pages gracefully (dataset row with error)
6. Pushes structured dataset items (type: "channel" and type: "video")
7. Writes SUMMARY.md to the key-value store

Why use YouTube Channel Scraper?

• Content research — discover what top channels post and how often
• Affiliate / niche research — collect public stats before outreach
• AI / ML datasets — bulk channel + video metadata for training or ranking features
• Competitor monitoring — track public channel growth signals and new uploads
• Automation-ready — JSON dataset + Apify API, webhooks, and schedules

How to scrape YouTube channels

1. Open the Actor in Apify Console
2. Paste channel URLs (e.g. https://www.youtube.com/@MrBeast)
3. Set max videos per channel (e.g. 20)
4. Keep include channel info enabled if you need About-tab stats
5. Enable Apify Proxy if you hit consent or rate-limit pages
6. Click Start
7. Download results from the Dataset tab (JSON / CSV / Excel) or read SUMMARY.md from the key-value store

Input

Field	Type	Default	Description
channelUrls	string[]	MrBeast + Google Developers	YouTube channel URLs
maxVideos	integer	20	Max recent videos per channel (0 = channel info only)
includeChannelInfo	boolean	true	Scrape About-tab channel metadata
maxConcurrency	integer	3	Concurrent HTTP requests
maxRequestRetries	integer	3	Retries on failure / soft blocks
requestDelayMs	integer	1000	Used for max requests/minute rate limiting
proxyConfiguration	object	Apify Proxy on	Proxy settings

Example:

{
    "channelUrls": [
        "https://www.youtube.com/@MrBeast",
        "https://www.youtube.com/@GoogleDevelopers"
    ],
    "maxVideos": 10,
    "includeChannelInfo": true,
    "maxConcurrency": 3,
    "maxRequestRetries": 3,
    "requestDelayMs": 1000,
    "proxyConfiguration": {
        "useApifyProxy": true
    }
}

See examples/input.json.

Output

Channel record

{
    "type": "channel",
    "channelId": "UCX6OQ3DkcsbYNE6H8uQQuVA",
    "title": "MrBeast",
    "description": "SUBSCRIBE FOR A COOKIE! …",
    "subscriberCount": 391000000,
    "videoCount": 841,
    "viewCount": 80123456789,
    "channelUrl": "https://www.youtube.com/@MrBeast",
    "country": "United States",
    "scrapedAt": "2026-07-18T12:00:00.000Z",
    "error": null
}

Video record

{
    "type": "video",
    "videoId": "dQw4w9WgXcQ",
    "title": "I Spent 50 Hours Buried Alive",
    "description": "I spent 50 hours buried alive...",
    "viewCount": 312456789,
    "likeCount": null,
    "publishedAt": "2 years ago",
    "duration": "15:42",
    "url": "https://www.youtube.com/watch?v=dQw4w9WgXcQ",
    "channelTitle": "MrBeast",
    "thumbnail": "https://i.ytimg.com/vi/dQw4w9WgXcQ/hqdefault.jpg",
    "channelId": "UCX6OQ3DkcsbYNE6H8uQQuVA",
    "channelUrl": "https://www.youtube.com/@MrBeast",
    "scrapedAt": "2026-07-18T12:00:00.000Z",
    "error": null
}

You can download the dataset in various formats such as JSON, HTML, CSV, or Excel.

Data fields

Field	Channel	Video	Description
type	yes	yes	"channel" or "video"
channelId	yes	yes*	UC… channel id
title	yes	yes	Channel or video title
description	yes	yes	About text / snippet
subscriberCount	yes	—	Subscribers
videoCount	yes	—	Total videos on channel
viewCount	yes	yes	Lifetime channel views or video views
channelUrl	yes	yes*	Canonical channel URL
country	yes	—	Country when present on About
videoId	—	yes	11-char video id
likeCount	—	yes	Often null on list pages
publishedAt	—	yes	Relative time (e.g. "2 years ago")
duration	—	yes	e.g. 15:42
url	—	yes	Watch URL
channelTitle	—	yes	Owning channel name
thumbnail	—	yes	Thumbnail image URL
scrapedAt	yes	yes	ISO scrape time
error	yes	yes	Set on failures / blocks

* optional enrichment on video rows when channel context is known.

How it works (technical)

• CheerioCrawler fetches About and Videos HTML over HTTP
• ytInitialData is extracted via regex + brace-balanced JSON parse
• Channel stats come from channelMetadataRenderer, c4TabbedHeaderRenderer, aboutChannelViewModel, and channelAboutFullMetadataRenderer
• Videos come from videoRenderer / gridVideoRenderer / richItemRenderer trees
• Fallbacks: Open Graph meta tags, DOM subscriber text, and a[href*="watch?v="] anchors
• Graceful handling for consent walls, captchas, 403/429, and missing channels

Pricing / cost estimation

This Actor uses Pay-per-event (PPE) pricing at about $0.003 per dataset result (channel or video row), plus a tiny actor-start fee. How much does it cost to scrape YouTube channels?

Scenario	Approx. results	Est. cost
10 channels, info only	~10	~$0.03
10 channels × 20 videos + info	~210	~$0.63
100 channels × 10 videos + info	~1,100	~$3.30

Platform free-tier and compute unit usage still apply when running on Apify. Enable proxy only when needed to control proxy costs.

Tips

• Prefer Apify Proxy (residential if you see consent / unusual traffic pages)
• Keep maxConcurrency at 2–5 for stable extractions
• Set maxVideos only as high as you need — list pages return first-screen results without infinite scroll pagination in this HTTP version
• likeCount is often null on Videos-tab listings (YouTube does not always embed likes there)
• Use @handle URLs for stability when custom /c/ paths change

FAQ, disclaimers, and support

Is scraping YouTube legal?
Scraping publicly available pages for legitimate research is a gray area that depends on jurisdiction, purpose, and ToS. This tool does not bypass logins or access private data. Consult legal counsel for your use case.

Why do I get consent or empty results?
Some regions show a cookie consent wall without ytInitialData. Use proxies (often US residential) and retries.

Does this use the official YouTube Data API?
No. It parses public HTML / embedded JSON. For API-quota-based access, use Google’s YouTube Data API instead.

Need a custom solution?
Open an issue on the Actor’s Issues tab in Apify Console, or request a custom scraper for deeper pagination, Shorts-only mode, or comment extraction.

Output example file

See examples/output.json for a full sample payload.