YouTube Scraper

Collect YouTube videos, channel information, comments, and available transcripts as rows you can export for research.

• Discover videos by keyword, channel, direct video link, or popular topic.
• Compare titles, publishing dates, engagement, and channel statistics.
• Add captions and metadata to shortlisted videos.
• Export comments and replies as separate rows.
• Pay $1.50 per 1,000 delivered rows plus a small start fee.

What does YouTube Scraper do?

YouTube Scraper collects public data from youtube.com for content research, channel monitoring, and transcript analysis. You can combine several inputs in one run. Each row represents one video, channel, or comment; replies are comment rows too.

• Search: use searchQueries with optional date, duration, and result-type filters.
• Channels: use channelUrls to list videos, shorts, streams, or all three tabs.
• Known videos: use videoUrls for specific videos, optionally adding details or captions.
• Popular topics: enable includeTrending for topic-based discovery in the market selected by gl.
• Extra rows: includeComments, includeRelated, and includeChannelAbout are off by default. Every extra row costs the same as a video row and uses the originating input's allowance. Replies are enabled when comments are enabled, unless you turn off includeCommentReplies.

Transcripts are fields on video rows, not extra billed rows. Use discovery first, review the titles, then request details or transcripts for a shortlist.

What data can you extract from YouTube?

Every row has rowType and a stable id. The same rowType and id pair is deduplicated. Optional fields depend on what YouTube exposes and which options you enable.

Video rows

Field	Description	Example
videoId	YouTube video identifier.	XVv6mJpFOb0
title	Video title.	Example tutorial
url	Video watch URL.	https://www.youtube.com/watch?v=XVv6mJpFOb0
channelName	Public channel name.	Example Channel
viewCount	Parsed count; zero can mean unavailable.	12500
durationSeconds	Video length in seconds.	480
publishedAt	Publishing date or estimated date.	2026-09-12
publishedAtApproximate	Whether the date came from relative text.	true
description	Search snippet or fuller description.	Example lesson description
chapters	Chapter titles and starting times, when available.	[{"title":"Intro","startMs":0}]
transcript	Caption segments; at most 50 are retained.	[{"text":"Example caption","start":0,"duration":2}]
transcriptText	Joined caption text, subject to size limits.	Example caption
transcriptLanguage	Language actually returned.	en
enrichmentLevel	Enrichment applied to the video.	transcript

Channel rows

Field	Description	Example
channelId	Channel identifier.	UCXXXXXXXXXXXXXXXXXXXXXX
channelName	Public channel name.	Example Channel
channelUrl	Public channel link.	https://www.youtube.com/@Example
handle	Channel handle, when available.	@Example
subscriberCount	Parsed subscriber count.	15000
totalViews	Channel lifetime views, when shown.	800000
videoCount	Channel video count, when shown.	120
links	Public links listed by the channel.	[{"title":"Website","url":"https://example.com"}]

Comment and reply rows

Field	Description	Example
commentId	Comment or reply identifier.	example-comment-id
videoId	Parent video identifier.	XVv6mJpFOb0
authorName	Public author display name.	Jane Doe
text	Comment text.	Example comment text
isReply	Whether this row is a reply.	true
parentCommentId	Parent comment for a reply.	example-parent-id
rootCommentId	Top-level comment in the thread.	example-parent-id
likeCount	Likes when exposed.	12

How to scrape YouTube

1. Open YouTube Scraper in Apify Console.
2. Fill one target field, such as searchQueries.
3. Set maxResults to 3 for a small first run.
4. Click Start and wait for completion.
5. Open the dataset and download the rows or read them through the API.

With no target filled, the Actor reads the sample channel https://www.youtube.com/@freecodecamp (videos tab), limited to three rows. This example uses normal pricing, including the start fee. Filling a target selects the matching collection path automatically; no separate mode selection is needed. Mixed inputs are supported.

Search by keyword

Enter topics in searchQueries. Use gl for the market and hl for the interface language. For example, gl: "JP" selects Japan; when hl is omitted, the Actor selects a language for the market. These settings are separate from caption language.

List channel videos

Enter public channel links, channel IDs, or handles in channelUrls. Select channelTab: videos, shorts, streams, or all. All tabs share one allowance per channel. Enable includeChannelAbout to add a channel row before its videos.

Collect known videos and transcripts

Enter watch, shorts, live, embed, or youtu.be links, or 11-character IDs, in videoUrls. Enable scrapeDetails for richer metadata and includeTranscript for captions. Caption extraction can fall back to another available language; check the returned transcriptLanguage.

Discover popular topics

Set includeTrending to true and choose gl. Results come from popular topic hubs, not a complete ranked list of everything trending. Rows may include trendingTopic. Trending counts as one input.

What to fill in

Field	What to enter	Default
searchQueries	Keywords; up to 50 entries.	Empty; sample prefilled
channelUrls	Channel links, IDs, or handles; up to 50.	Empty
videoUrls	Video links or IDs; up to 500.	Empty
maxResults	Rows per distinct input, from 1 to 1,000.	70; form prefill 3
includeTrending	Add popular-topic discovery.	Off
trendingMaxResults	Smaller trending candidate cap, up to 200.	Smaller of maxResults and 200
gl	Two-letter country code, such as US or JP.	US
hl	Interface language, such as en or ja.	Selected from gl
uploadDate	any, hour, today, week, month, year.	any
resultType	any, video, channel, playlist, movie, short.	any
duration	any, short, medium, long.	any
channelTab	videos, shorts, streams, all.	videos
includeChannelAbout	Add a public channel-information row.	false
scrapeDetails	Request richer video metadata.	false
includeTranscript	Request available captions.	false
transcriptLanguage	Preferred caption language.	en
includeComments	Add comment rows.	false
includeCommentReplies	Add replies when comments are enabled.	true
maxComments	Top-level comment cap per video; 0 uses safety limits.	0
maxRepliesPerComment	Reply cap per comment; 0 uses safety limits.	0
commentSort	top or newest.	top
includeRelated	Add related videos as separate rows.	false
maxRelated	Related videos considered per source video.	20
includeEngagement	Include available engagement counts with details.	true
includeStructure	Include available chapters, topics, and hashtags with details.	true
includePlaybackMeta	Include available playback metadata with details.	true
includeFormats	Include available format information with playback metadata.	Follows includePlaybackMeta
includeCardsEndscreen	Include cards and end-screen information with details.	true
includeUpdatedMetadata	Refresh available video metrics.	false
includeResolveHandle	Resolve channel identifiers; use includeChannelAbout for channel rows.	false
includeGuide	Add category-navigation information to the run summary, not rows.	false
concurrency	Concurrent video enrichment operations, 1–16.	6
proxyCountry	Optional two-letter country override.	Selected market
useProxyForSearch	Apply the country setting to search and channel results.	false

Search example

{
  "searchQueries": ["ai workflow automation"],
  "maxResults": 3
}

Channel example

{
  "channelUrls": ["https://www.youtube.com/@freecodecamp"],
  "channelTab": "videos",
  "maxResults": 3
}

Video and transcript example

{
  "videoUrls": ["https://www.youtube.com/watch?v=XVv6mJpFOb0"],
  "scrapeDetails": true,
  "includeTranscript": true,
  "transcriptLanguage": "en",
  "maxResults": 1
}

Popular-topic example

{
  "includeTrending": true,
  "gl": "US",
  "maxResults": 3
}

maxItems and limit are accepted aliases for maxResults. Target aliases include query and keywords for searchQueries, channels for channelUrls, and videoIds for videoUrls. Comma- or newline-separated target text is split into entries.

Close field-name misspellings can be corrected; other unknown fields are ignored with a note. Unsupported filter values are ignored with a warning. Every run writes its input interpretation to INPUT_REPORT in the default key-value store. Invalid target links stop collection with a message such as INVALID_INPUT: videoUrls — use a YouTube video link ... or the 11-character video ID, before any row is charged. The start fee still applies. uploadDate is a relative filter; the Actor does not automatically widen it.

Output example

Illustrative video row; values below are examples, not a captured result:

{
  "rowType": "video",
  "id": "XVv6mJpFOb0",
  "videoId": "XVv6mJpFOb0",
  "title": "Example tutorial",
  "url": "https://www.youtube.com/watch?v=XVv6mJpFOb0",
  "channelName": "Example Channel",
  "viewCount": 12500,
  "durationSeconds": 480,
  "publishedAt": "2026-09-12",
  "description": "Example lesson description",
  "gl": "US",
  "hl": "en",
  "source": "search",
  "scrapedAt": "2026-10-24T00:00:00Z"
}

Export the dataset as JSON, CSV, Excel, XML, RSS, or an HTML table. JSON preserves nested caption and chapter lists. RUN_SUMMARY records delivered counts and warnings; ERROR_SUMMARY provides failure details when written.

Why use this YouTube scraper?

• Research in stages. Discover first, then enrich a shortlist rather than requesting every optional field immediately.
• Several inputs together. Combine keywords, channels, and known videos in one run.
• Consistent billing. Each delivered row uses the same price, including comments and replies.
• Visible input handling. Inspect INPUT_REPORT for corrections and ignored fields.
• No proxy setup. You do not need to set up proxies.

How much does it cost to scrape YouTube?

The Actor uses pay-per-event pricing: $0.005 per start event, plus the row price below. Platform usage is included in these prices.

Apify plan	Price per row	Price per 1,000 rows
Free	$0.0015	$1.50
Starter	$0.0015	$1.50
Scale	$0.0015	$1.50
Business	$0.0015	$1.50

Start billing is one event per GB of allocated memory, with a minimum of one. At the default 512 MB, one start costs $0.005. Video details and transcripts do not add a separate row charge; channel, comment, reply, and related-video rows each count separately.

• maxResults applies to each distinct input, defaults to 70, and accepts up to 1,000. Larger collections are processed in parts.
• All row types from one input share its allowance. A video can consume the only available row when maxResults is 1, leaving no room for comments.
• Free Apify plan runs are capped at 200 rows across all inputs.
• The run also stops at your Maximum cost per run (maxTotalChargeUsd in the API). Fewer available results can mean fewer rows.

Example: three keywords delivering 50 rows each produce 150 rows. At 512 MB with one start, the cost is 150 × $0.0015 + $0.005 = $0.230 on the Free plan.

What happens when the run timeout is too short

Some runs take longer than the run timeout you chose, for example many keywords or channels with comments and transcripts turned on. In that case the run does not fail. It ends early with the status message "Queued on the collection backend; this run will resume automatically when results are ready (about N s)." and shows as ABORTED for a while. When the results are ready, the same run starts again by itself, reports "Resumed: delivering N rows.", writes the rows to the same dataset and finishes as SUCCEEDED. Rows already delivered are skipped, so no row is delivered or charged twice, and you do not need to start another run.

• This can happen at most 3 times per run. For larger jobs, choose a longer run timeout rather than a very short one. A manual abort or an unrelated failure is not the same as this automatic pause.
• The synchronous API endpoints (run-sync, run-sync-get-dataset-items) wait at most 300 seconds and can return an empty dataset while the run is paused. For larger jobs, start the run asynchronously and read the dataset after the run reaches SUCCEEDED, or use a webhook on the ACTOR.RUN.SUCCEEDED event.

Use with AI agents and the API

Apify MCP server (Claude, ChatGPT, Cursor, VS Code)

Add the Actor to an MCP client that supports remote servers:

https://mcp.apify.com?tools=lentic_clockss/youtube-research-scraper

Sign in with Apify when prompted through OAuth, or send Authorization: Bearer <YOUR_APIFY_TOKEN>. The tool starts a run and returns run and dataset IDs; read rows with get-dataset-items. Clients without remote support can use npx @apify/actors-mcp-server --tools lentic_clockss/youtube-research-scraper with APIFY_TOKEN set.

Example prompt: "Search YouTube for ai workflow automation and return three video rows with titles, links, and channel names."

Apify API

from apify_client import ApifyClient

client = ApifyClient("<YOUR_APIFY_TOKEN>")
run = client.actor("lentic_clockss/youtube-research-scraper").call(
    run_input={"searchQueries": ["ai workflow automation"], "maxResults": 3}
)
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item)

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: '<YOUR_APIFY_TOKEN>' });
const run = await client.actor('lentic_clockss/youtube-research-scraper').call({
  searchQueries: ['ai workflow automation'], maxResults: 3
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items);

Over HTTP, send the input JSON to POST https://api.apify.com/v2/acts/lentic_clockss~youtube-research-scraper/runs with your token in the Authorization: Bearer header. After completion, read GET https://api.apify.com/v2/datasets/<datasetId>/items?format=json.

Integrations

Use the Actor's Integrations tab to connect results to Google Sheets, Make, Zapier, n8n, Slack, Airtable, or your own webhook. Use Apify Schedules for recurring collection.

FAQ

Is it legal to scrape YouTube?

This Actor collects public data. You are responsible for complying with YouTube's terms and applicable data-protection law. Processing personal data, including comment author information, requires a lawful basis.

Why did my run end as ABORTED?

The run was stopped before normal completion. Check its status message and log for the reason; do not assume it will resume automatically. Inspect the dataset before starting again. Delivery of unfinished collection is not guaranteed.

How many results can I get?

maxResults allows 1–1,000 rows per input. Available data, shared allowances for extra rows, duplicate removal, the Free plan cap, and your cost limit can reduce the total.

Why are some fields empty?

YouTube may omit captions, counts, dates, or playback information. Optional enrichment can also fail while leaving a video row available. Missing values can be absent, empty, null, or zero; zero does not always prove a measured count of zero.

Can I get every comment and reply?

Not necessarily. maxComments: 0 and maxRepliesPerComment: 0 use safety limits of 20,000 top-level comments and 2,000 replies per comment, but the input's row allowance takes precedence. Disabled comments or unavailable threads cannot be collected.

Are transcripts always complete?

No. Captions may be unavailable. The segment list retains at most 50 entries; transcriptSegmentCount records the count before that limit. transcriptText is capped at 100,000 characters and may be shortened further by the row-size limit.

Do I need a YouTube Data API key?

No YouTube Data API key is required. Programmatic access through Apify uses your Apify token.

Limitations

• Search filters do not guarantee matching rows. The playlist filter is not a playlist export; delivered row types remain video, channel, and comment.
• Topic discovery is not an exhaustive popularity ranking, and results vary by market and language.
• Private, deleted, restricted, or unavailable content may yield limited data or no rows. An entirely empty collection is reported as a failure.
• Selected attribute lists retain at most 50 entries. Size reduction may remove fields; inspect rowTruncatedFields and rowOversizeReduced.
• Derived publishing dates are marked by publishedAtApproximate. Do not treat them as exact timestamps.

Related scrapers

More social media and video scrapers from the same publisher:

• TikTok Scraper
• Reddit Scraper
• YouTube Shorts Scraper
• Hacker News Scraper

Browse all of our scrapers at apify.com/lentic_clockss.

Support

Found a problem or need another field? Open an issue in the Issues tab of this Actor's page and include the run ID and the input you used.