Youtube Channel Videos Scraper

Extract a channel's videos at scale — latest uploads, popular videos, filtered ranges. Uses YouTube Data API v3 for fast, reliable extraction with pagination up to thousands of videos per channel. No browser or proxy required.

Key Features

• Channel resolution — Accepts @handle, channel URL, /channel/UC..., /c/custom, /user/legacy
• Deep pagination — Extract up to 5,000 videos per channel via uploads playlist
• Statistics enrichment — Views, likes, comments via batched videos.list API calls
• Content details — Duration, HD/SD, caption flags, licensed content, topic categories
• Shorts & Live detection — Automatic classification based on duration and broadcast content
• Date filtering — Filter by publish date range (startDate / endDate)
• Sorting — Newest, oldest, or most popular (by view count)
• Multi-channel — Process multiple channels in a single run with cross-channel deduplication
• Channel profile — Subscriber count, total views/videos, description, country, joined date, avatar & banner per record (no extra API quota)
• Full enrichment — Descriptions, tags, thumbnails, topic categories, extracted description links, and hashtags always included — nothing to configure
• Lightweight — No browser needed — runs on apify/actor-node:22
• Reliable — Graceful per-channel error handling, partial result output, and detailed run summary

Why Choose This Actor?

• Official API, not scraping — built on the official YouTube Data API v3. No residential proxies, no unofficial endpoints, no browser automation to break when YouTube changes its HTML. That means fewer failed runs and no dependency on a paid proxy plan.
• Clean, flat, spreadsheet-ready output — a single flat record per video imports straight into Google Sheets, Excel, or a BI tool. No nested unwrapping required.
• Deep by default — up to 5,000 videos per channel, multiple channels per run, with cross-channel deduplication.
• Rich per-video metadata — statistics, duration, Shorts/Live flags, tags, topic categories, caption/definition/licensed flags, thumbnails, channel profile, description links, and hashtags — full enrichment on every record, no setup required.
• Cost control built in — maxVideosPerChannel and maxTotalResults cap how many records a run produces (and bills), even across many channels.
• Transparent & auditable — every record carries source, runId, sourceChannelRef, and retrievedAt, and each run writes a RUN_SUMMARY.json.

How It Works

1. Resolve channel — Convert @handle, URL, or channelId to a canonical channelId via channels.list
2. Get uploads playlist — Fetch the channel's uploads playlist ID via contentDetails
3. Paginate playlist — Extract video IDs via playlistItems.list (50 per page, up to 5,000)
4. Enrich videos — Batch enrich with videos.list (50 per call) for stats, duration, descriptions, tags, and topics
5. Filter & sort — Apply date range, Shorts/Live toggles, and sort order
6. Output — Push clean per-video records to dataset + write RUN_SUMMARY.json

---

Input

Channel Settings

Field	Type	Default	Description
channelRefs	string[]	required	Channel references: @handle, channel URL, /channel/UC..., /c/customURL
maxVideosPerChannel	integer	200	Max videos per channel (1–5,000)
maxTotalResults	integer	0	Global cap on total videos across all channels (0 = no limit)
includeShorts	boolean	true	Include YouTube Shorts (videos ≤60 seconds)
includeLive	boolean	false	Include live streams and premieres

Filtering & Sorting

Field	Type	Default	Description
startDate	string	""	Include videos published on/after this date (YYYY-MM-DD)
endDate	string	""	Include videos published on/before this date (YYYY-MM-DD)
publishedWithinDays	integer	0	Only include videos from the last N days (0 = disabled; applied on top of start/end date)
sortBy	string	"newest"	newest / oldest / popular (popular sorts by view count)
dedupeAcrossChannels	boolean	true	Skip duplicate videos across channels

Enrichment — always on

Every run returns the complete record. Channel profile (subscribers, total views/videos, description, country, joined date, avatar, banner), statistics, content details, thumbnails, full description, description links, hashtags, tags, and topic categories are all included automatically — there are no enrichment toggles to set.

---

Input Examples

Example 1 — Latest 200 videos from a single channel

Fetch the 200 most recent videos from one channel, newest first. Every record comes fully enriched.

{
  "channelRefs": [
    "@mkbhd"
  ],
  "maxVideosPerChannel": 200,
  "includeShorts": true,
  "includeLive": false,
  "sortBy": "newest",
  "dedupeAcrossChannels": true
}

Example 2 — Popular videos from the last 90 days, across channels, with a global cap

Pull each channel's most-viewed uploads from the last 90 days across three channels, capped at 1,000 total billed records.

{
  "channelRefs": [
    "@mkbhd",
    "https://www.youtube.com/@veritasium",
    "UC_x5XG1OV2P6uZZ5FSM9Ttw"
  ],
  "maxVideosPerChannel": 500,
  "maxTotalResults": 1000,
  "publishedWithinDays": 90,
  "includeShorts": true,
  "includeLive": true,
  "sortBy": "popular",
  "dedupeAcrossChannels": true
}

---

Output

Each video is a flat JSON record with up to 39 fields — full channel and video enrichment is included on every record.

Output Fields

Field	Type	Description
retrievedAt	string	Fetch timestamp (ISO 8601)
channelId	string	Channel ID
channelTitle	string	Channel name
channelHandle	string	@handle
channelUrl	string	Canonical channel URL
sourceChannelRef	string	The original input reference that produced this record
channelSubscriberCount	integer	Channel subscribers (null if hidden)
channelVideoCount	integer	Total public videos on the channel
channelViewCount	integer	Total channel views
channelCountry	string	Channel country code
channelPublishedAt	string	Channel creation/joined date (ISO 8601)
channelDescription	string	Channel "About" text
channelAvatarUrl	string	Channel avatar image URL
channelBannerUrl	string	Channel banner image URL
videoId	string	Video ID
videoUrl	string	Watch URL or /shorts/ URL
title	string	Video title
publishedAt	string	Publish timestamp (ISO 8601)
publishDate	string	YYYY-MM-DD
order	integer	1-based result position within the channel (after sort)
videoType	string	Normalized type: regular / short / live
duration	string	ISO 8601 duration (e.g., PT12M36S)
durationSeconds	integer	Duration in seconds
isShorts	boolean	Shorts flag (duration ≤60s)
isLive	boolean	Live/premiere indicator
viewCount	integer	Views
likeCount	integer	Likes (may be hidden)
commentCount	integer	Comments (may be disabled)
thumbnails	object	Thumbnail URLs (default/medium/high/standard/maxres)
description	string	Full video description
descriptionLinks	object[]	Links extracted from the description ({ url, text })
hashtags	string[]	Hashtags extracted from the title + description
tags	string[]	Video tags
definition	string	hd / sd
caption	string	true / false
licensedContent	boolean	Licensing flag
topicCategories	string[]	Wikipedia topic category URLs
source	string	api or api_partial
runId	string	Apify run ID

Output Example

---

Export Formats

Results are stored in an Apify dataset and can be exported in one click to JSON, CSV, Excel, XML, or HTML, or pulled programmatically via the Apify API and the Python / JavaScript clients. The flat record structure maps cleanly to spreadsheet columns.

---

Pricing

This Actor uses pay-per-event billing — you are charged per video record emitted. Failed channels emit an error item and are not charged.

Example costs at $0.75 / 1,000 videos (see the Actor's Pricing tab on Apify for the current rate):

Scenario	Videos	Est. cost
1 channel × 200 videos	200	~$0.15
5 channels × 200 videos	1,000	~$0.75
1 channel × 1,000 videos	1,000	~$0.75
10 channels × 500 videos	5,000	~$3.75

Use maxTotalResults to hard-cap the number of billed records on large multi-channel runs.

---

Performance

Scenario	Typical Time
1 channel x 200 videos	~30–90 seconds
10 channels x 200 videos	~5–15 minutes
1 channel x 5,000 videos	~10–40 minutes

---

Run Summary

Each run saves RUN_SUMMARY.json to the Key-Value Store with:

• Total channels processed / failed
• Total videos emitted
• Per-channel video counts
• Errors by channel
• API call count and quota usage
• Start and completion timestamps

Error Items

If a channel can't be resolved or processed, the Actor pushes a distinct error item to the dataset instead of failing the whole run (and does not charge for it):

{
  "isError": true,
  "sourceChannelRef": "@this-channel-does-not-exist",
  "error": "Could not resolve channel: \"@this-channel-does-not-exist\"",
  "retrievedAt": "2026-07-09T05:00:00.000Z",
  "runId": "JxebN0GRa9nkLuhDa"
}

Filter these out with isError (or keep them to audit which inputs failed).

---

Use Cases

• Competitor analysis — Export and compare video catalogs across channels
• Content audits — Analyze upload frequency, Shorts vs long-form ratio, and engagement
• Dataset generation — Build research datasets of niche channels
• Marketing research — Identify trending topics and high-performing content
• Monitoring & alerts — Track new uploads from partner/influencer channels
• Watchlist building — Create video lists for further scraping (comments, transcripts, etc.)

---

Limitations

• Uploads playlist only — Private and unlisted videos are excluded (YouTube API limitation)
• Shorts detection — Based on duration heuristic (≤60s); may not catch all edge cases
• Live detection — Only detects currently live or upcoming; past live streams show as false
• Hidden stats — Like count and comment count may be null if the creator has hidden them; channelSubscriberCount is null when the channel hides its subscriber count
• No verification flag — The official YouTube Data API does not expose a channel "verified" status, so it is not included
• Sequential processing — Channels are processed one at a time (concurrent processing planned for a future release)

---

FAQ

Is this legal? The Actor reads only publicly available channel and video data through the official YouTube Data API v3. It does not access private, deleted, or unlisted videos, and it collects no private personal data.

Do I need a proxy? No. Because it uses the official API, there is no proxy or residential-IP requirement and no proxy plan to pay for.

Do I need my own API key? No. The Actor is ready to run out of the box — you don't need to provide, configure, or manage anything.

Is there a limit on how many videos I can pull? Up to 5,000 per channel per run, across as many channels as you like. Very large runs may be subject to YouTube's daily API limits; if a job reaches that ceiling it can be continued the next day.

Why is likeCount or commentCount sometimes empty? The creator has hidden likes, or disabled comments — YouTube then returns no value.

How do I control cost? Use maxVideosPerChannel and maxTotalResults to cap how many videos (and therefore billed records) a run produces.

Which export formats are supported? JSON, CSV, Excel, XML, and HTML — plus the Apify API and Python/JS clients.

---

Support & Feedback

If you encounter any issues or have feature requests, please open an issue on GitHub.