YouTube Comments Scraper extracts the public comments of any YouTube video: text, author, likes, reply count, date and pinned status, sorted by top or newest. It is made for marketers, researchers, data teams and creators who want to download YouTube comments and export them to Excel, CSV or JSON without a YouTube API key or quota.

What does YouTube Comments Scraper do?

You give it a list of YouTube videos and it returns one row per comment. For each video it opens the comment section, follows the "load more" pages and stops when it reaches your limit or when YouTube has no more comments.

It accepts:

• Watch URLs: https://www.youtube.com/watch?v=aircAruvnKk
• Short links: https://youtu.be/aircAruvnKk
• Shorts URLs: https://www.youtube.com/shorts/<id>
• Embed and live URLs: /embed/<id>, /live/<id>
• Plain 11-character video IDs: aircAruvnKk

You can mix all of these in the same run. If the same video appears twice in your list, it is scraped only once.

What comes out: the comment text, the author's handle and channel ID, whether the author is the video's creator or verified, the like count, the number of replies, the date as YouTube shows it ("5 years ago"), whether the comment is pinned, a direct link to the comment, and the time it was scraped.

You don't need a Google account, cookies or a YouTube Data API key.

What data can you extract?

Every comment is one item in the dataset with these fields:

Field	What it means	Example
videoId	ID of the video the comment belongs to	aircAruvnKk
commentId	YouTube's ID of the comment	UgwEXAMPLEcommentId4AaABAg
text	Full text of the comment	Bio teacher: what is a neuron?…
author	Display handle of the person who wrote it	@example_user
authorChannelId	Channel ID of the author	UCexampleChannelId000000
isCreator	true if the author is the video's creator	false
isVerified	true if the author has a verified badge	false
likes	Number of likes (YouTube rounds big numbers, e.g. "3K" becomes 3000)	3000
replyCount	Number of replies to the comment	35
publishedTime	Date as shown by YouTube, relative and in English	7 years ago
isPinned	true if this is the comment pinned by the creator	false
url	Direct link that opens the video at this comment	https://www.youtube.com/watch?v=aircAruvnKk&lc=…
scrapedAt	When the comment was collected (ISO 8601, UTC)	2026-10-08T20:09:08.128Z
error	Only on error items: why a video gave no comments	Comments are turned off or not available

The default table view in the Console shows videoId, author, text, likes, replyCount, publishedTime, isPinned and commentId. All fields are always in the full JSON, CSV and Excel exports.

How to use YouTube Comments Scraper

1. Open the robot in Apify Console and go to the Input tab.
2. In Video URLs or IDs, paste one video per line (any of the URL forms above, or IDs).
3. Set Max comments per video. The default is 100.
4. Choose Sort by: top (YouTube's "Top comments") or newest (newest first).
5. Leave Proxy on the default Apify Proxy.
6. Click Start and wait for the run to finish. You can watch the comments appear in the Output tab.
7. Click Export and download the results as JSON, CSV, Excel, HTML or XML, or read them through the API.

Input

Field	Type	Default	What it does
videos	array of strings	["https://www.youtube.com/watch?v=aircAruvnKk"]	Video URLs (watch, youtu.be, shorts, embed, live) or 11-character IDs.
maxComments	integer, minimum 1	100	Stop after this many comments for each video. There is no upper limit in the input; the robot also stops when YouTube has no more comments.
sort	top or newest	top	Order in which YouTube serves the comments.
proxy	object	{ "useApifyProxy": true }	Apify Proxy settings.

Example input:

{
  "videos": [
    "https://www.youtube.com/watch?v=aircAruvnKk",
    "https://youtu.be/dQw4w9WgXcQ"
  ],
  "maxComments": 500,
  "sort": "newest",
  "proxy": { "useApifyProxy": true }
}

Output example

One real item from a recent run on the 3Blue1Brown video "But what is a neural network?" (the author's handle and IDs are replaced because the commenter is a private person):

{
  "videoId": "aircAruvnKk",
  "commentId": "UgwEXAMPLEcommentId4AaABAg",
  "text": "Bio teacher: what is a neuron?\nMe: a thing that holds a number between 0 and 1",
  "author": "@example_user",
  "authorChannelId": "UCexampleChannelId000000",
  "isCreator": false,
  "isVerified": false,
  "likes": 3000,
  "replyCount": 35,
  "publishedTime": "7 years ago",
  "url": "https://www.youtube.com/watch?v=aircAruvnKk&lc=UgwEXAMPLEcommentId4AaABAg",
  "isPinned": false,
  "scrapedAt": "2026-10-08T20:09:08.128Z"
}

When a video gives no comments, you get one error item instead:

{
  "videoId": "aircAruvnKk",
  "error": "Comments are turned off or not available"
}

How much does it cost?

The robot uses pay per event pricing. You pay only for the comments you get, and Apify platform usage is already included in the price.

Event	Price (Free plan)
Comment scraped	$0.0008 per comment, which is $0.80 per 1,000 comments

Paid Apify plans get a discount on this price:

Plan	Price per 1,000 comments
Free	$0.80
Bronze (-10%)	$0.72
Silver (-20%)	$0.64
Gold and higher (-30%)	$0.56

Worked examples on the Free plan:

• One video with the default limit of 100 comments: 100 × $0.0008 = $0.08.
• 10,000 comments from one big video: $8.00 ($5.60 on Gold).
• 50 videos × 500 comments = 25,000 comments: $20.00 ($18.00 on Bronze, $16.00 on Silver, $14.00 on Gold).

What is NOT charged:

• Error items (the video could not be loaded, or comments are turned off).
• Comments the robot already returned for that video (duplicates are dropped before saving).
• Lines in your input that are not a YouTube video URL or ID (they are skipped).

If a video has fewer comments than your limit, you pay only for the comments it really has.

Use cases

• Sentiment analysis: download every comment on a product launch, trailer or ad and score it as positive, negative or neutral.
• Audience research: read what viewers ask in the comments to find content ideas, FAQ topics and pain points.
• AI and NLP datasets: build labelled text datasets for classification, topic modeling or training chatbots.
• Brand and campaign monitoring: run it on a schedule with sort set to newest to follow new reactions to your videos or your competitors' videos.
• Influencer vetting: check whether a creator's audience writes real comments, and how often the creator answers (isCreator).
• Community management: find the most liked comments and those with many replies (replyCount) to see which discussions matter.
• Competitor research: compare what people praise and complain about under competitors' videos.

Tips and limits

• Top-level comments only. The robot reads the main comment list. It does not open reply threads, so replies are not returned as items. Use replyCount to see how many replies a comment has.
• Dates are relative. publishedTime is the text YouTube shows, such as 7 years ago or 5 years ago (edited). YouTube does not show an exact date for comments in this list, so the robot does not invent one. Use scrapedAt as your reference point.
• Like counts are rounded the way YouTube displays them. A comment with "13K" likes is saved as 13000.
• Newest first uses YouTube's own "Newest first" menu. If YouTube does not offer that menu for a video, the comments come in the default order.
• You may get fewer comments than maxComments. The robot stops when YouTube returns no more pages, and it counts top-level comments only.
• Videos are processed one after another. Large lists take longer, but each video's comments are saved as soon as each page is read.
• Error items. You get an item with an error field when the video could not be loaded (Could not load video) or when the comment section is turned off or missing (Comments are turned off or not available). These items are free.
• Proxy. Requests go through Apify Proxy. When a request fails, the robot retries with new IPs, and uses the Residential proxy group as a last try when your account has access to it.
• Text is in English where YouTube adds its own words (for example years ago).

Use it from the API, Python or JavaScript

You can start the robot and read its results with the official Apify clients.

JavaScript (Node.js):

const { ApifyClient } = require('apify-client');

const client = new ApifyClient({ token: 'YOUR_TOKEN' });
const input = {
  videos: ['https://www.youtube.com/watch?v=aircAruvnKk'],
  maxComments: 200,
  sort: 'top',
};
const run = await client.actor('juliob.rangel/youtube-comments-fast').call(input);
const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items.length, 'comments');

Python:

from apify_client import ApifyClient

client = ApifyClient('YOUR_TOKEN')
run_input = {
    'videos': ['https://www.youtube.com/watch?v=aircAruvnKk'],
    'maxComments': 200,
    'sort': 'top',
}
run = client.actor('juliob.rangel/youtube-comments-fast').call(run_input=run_input)
items = client.dataset(run['defaultDatasetId']).list_items().items
print(len(items), 'comments')

You can also call it over plain HTTP with the Apify API (for example the run-sync-get-dataset-items endpoint of juliob.rangel~youtube-comments-fast).

The robot works with Make, Zapier, n8n, Google Sheets export, webhooks (to get notified when a run finishes) and AI agents through the Apify MCP server.

FAQ

Do I need a login, cookies or a YouTube API key? No. The robot reads the public comment section the same way a logged-out visitor sees it. You only need an Apify account.

Can I export YouTube comments to Excel or CSV? Yes. After the run, click Export in the Output tab and choose Excel, CSV, JSON, HTML or XML. You can also pull the dataset through the API.

How many comments can I get per video? As many as YouTube serves in its comment list, up to your maxComments value. There is no fixed cap in the input. Very large comment sections take longer and cost more, so start with a smaller limit to test.

Does it scrape replies? No. It returns top-level comments with their replyCount. Reply threads are not opened.

Can I get the exact date of each comment? No. YouTube shows comment dates as relative text (2 weeks ago), and the robot saves that text in publishedTime.

Why did I get an error item? The video could not be loaded after several retries, or the video has its comments turned off. Check that the video is public and opens in a browser. Error items are not charged.

Is scraping YouTube comments legal? The robot only collects publicly available data. You are responsible for complying with YouTube's terms and with privacy laws such as GDPR and LGPD. Comments are personal data of their authors, so you need a legitimate reason to collect and store them.

Can I run it every day? Yes. Create a schedule in Apify Console with sort set to newest to track new comments over time.

Related scrapers

• YouTube Transcript Scraper (Fast & Cheap)
• YouTube Video Details Scraper (Views, Likes, Comments)
• YouTube Search Scraper (Fast & Cheap)
• YouTube Channel Videos Scraper (Fast & Cheap)
• YouTube Shorts Scraper (Channels & Hashtags)
• TikTok Comments Scraper (Text, Likes, Replies, Users)

---

Comentários do YouTube

Comentários de qualquer vídeo do YouTube, por relevância ou mais recentes: texto, autor, curtidas, respostas e data. Para análise de sentimento, ideias de conteúdo e dados para IA. Sem login e sem chave de API; você paga só por comentário retornado.