YouTube Comments Scraper — Author, Likes, Replies & Text

Scrape all the comments from any YouTube video, channel or search in bulk. This YouTube comments scraper extracts every top‑level comment with its text, author, verified/creator flags, like count, reply count and timestamp, plus the video title, channel and view count on every row — no login, no API key, no coding. When you need to scrape YouTube comments at scale, point it at video URLs, a channel or a keyword and press Start.

Export to JSON, CSV, Excel, or API.

Built for data analysts, market researchers, social‑media managers, brand & community teams, moderators, creators and growth marketers who need clean, analysis‑ready YouTube comment data.

---

🔎 Why scrape YouTube?

YouTube comments are where a video's audience actually talks — the feedback, questions, praise, complaints, jokes and real‑world language you can't find anywhere else. Every popular video is a focus group of thousands of viewers reacting in their own words, and that YouTube comment data is pure, unfiltered voice‑of‑customer.

The problem is access. YouTube loads comments a few at a time behind infinite scroll, sorts them opaquely, and gives you no export button. Copying hundreds or thousands of comments by hand is impossible, and the official Data API is rate‑limited, quota‑capped and requires a Google Cloud project and OAuth setup. A purpose‑built YouTube comments scraper collects the whole conversation in one run and hands you a tidy spreadsheet.

This scraper does exactly that: it walks YouTube's internal pagination for you and returns every comment it can reach — up to your limit — with full author and engagement detail attached.

Common reasons people scrape YouTube:

• Sentiment analysis — measure how an audience feels about a product, launch, movie, team or campaign.
• Audience & market research — surface the questions, objections and desires viewers repeat again and again.
• Content ideas — mine the most‑liked comments and most‑asked questions for your next video or post.
• Competitor analysis — see exactly what audiences love and criticize on rival channels' videos.
• Moderation & brand safety — pull the comments on your own videos to review, flag or archive at scale.
• Community & influencer insight — find your most engaged, highest‑liked and verified commenters.
• Academic & trend research — build datasets of public discourse around a topic, event or creator.

✨ What this YouTube scraper does

• ✅ Three ways to target comments — paste exact video URLs / IDs, discover videos by search keyword / topic, or pull the latest uploads from one or more channels (handles, channel links or channel IDs). Mix all three in a single run.
• ✅ Automatic pagination at scale — follows YouTube's continuation tokens to collect hundreds to thousands of comments per video, de‑duplicated, up to your maxCommentsPerVideo limit (max 100,000).
• ✅ Rich per‑comment fields — commentId, author, authorChannelId, authorIsVerified, authorIsCreator, text, likeCount (numeric), likeCountText (as shown, e.g. 266K), replyCount, and publishedTime.
• ✅ Video context on every row — videoId, videoUrl, videoTitle, channel and videoViewCount, so each comment is self‑describing for analysis.
• ✅ Sort control — choose Top comments (most‑liked, most‑relevant) or Newest first to monitor fresh reactions.
• ✅ Keyless & no login — runs on YouTube's public web data with no API key, OAuth or cookies to manage.
• ✅ Fast & concurrent — processes multiple videos in parallel with configurable concurrency, built‑in retries and IP rotation for reliability.
• ✅ Export anywhere — download as JSON, CSV, Excel or HTML table, or stream straight from the Apify API.

🎯 Use cases

• Brand & product teams — track sentiment and FAQs under review, unboxing and launch videos for your products.
• Market researchers — build voice‑of‑customer datasets across an entire niche or product category.
• Social‑media & content managers — generate a content backlog from the questions and themes viewers raise.
• Agencies & competitor analysts — benchmark audience reaction on competitors' channels.
• Data scientists & ML teams — assemble labeled comment corpora for sentiment, toxicity or topic models.
• Community managers & moderators — review and archive comments on your own channel without endless scrolling.
• Creators & influencers — discover your most engaged fans and the questions worth answering on camera.
• Journalists & academics — study public discourse and reactions around news, events and public figures.

📥 Input

Provide any combination of topics, search keywords, video URLs and channel URLs, then tune how many comments and videos to collect. All fields are optional — run it with the prefilled topics to try it instantly.

Field	Type	Description
topics	array	Pick one or more popular topics (dropdown) and the scraper finds trending videos and pulls their comments — no links needed.
searchKeywords	array	Type any search term in your own words to discover videos and scrape their comments, e.g. iphone 16 review or your brand name.
videoUrls	array	Exact YouTube video links, youtu.be links, /shorts/ links or 11‑character video IDs to scrape directly.
channelUrls	array	Channel links, @handles or channel IDs — scrapes comments from each channel's latest videos.
sortBy	string	top (most popular first) or newest (most recent first). Default top.
includeVideoDetails	boolean	Add videoTitle, channel and videoViewCount to each comment row. Default true.
maxCommentsPerVideo	integer	Top‑level comments to collect per video. Default 100, max 100000.
maxVideosPerSource	integer	Videos to collect for each topic / search term / channel. Default 20.
maxConcurrency	integer	Videos processed in parallel. Default 5, max 15.
proxyConfiguration	object	Apify Proxy. Off by default; enable for very large runs or if you hit rate limits.

Example — comments from specific videos

{
  "videoUrls": ["dQw4w9WgXcQ", "https://youtu.be/arj7oStGLkU"],
  "maxCommentsPerVideo": 500,
  "sortBy": "top"
}

Example — newest comments from a channel

{
  "channelUrls": ["@mkbhd"],
  "maxVideosPerSource": 10,
  "maxCommentsPerVideo": 200,
  "sortBy": "newest"
}

Example — discover by keyword

{
  "searchKeywords": ["iphone 16 review"],
  "maxVideosPerSource": 20,
  "maxCommentsPerVideo": 200
}

📤 Output

Each dataset item is one YouTube comment with its author, engagement and video context. Here is a realistic output record:

{
  "videoId": "FANEYwR-6R0",
  "videoUrl": "https://www.youtube.com/watch?v=FANEYwR-6R0",
  "videoTitle": "iPhone 16 In 2026! (Still Worth Buying?) (Review)",
  "channel": "Simple Alpaca",
  "videoViewCount": "18521",
  "commentId": "Ugxke4pVzDnCY2BnMqN4AaABAg",
  "author": "imkevson",
  "authorChannelId": "UCXJFYkJ6upAnIcZ7bUIK_Uw",
  "authorIsVerified": "false",
  "authorIsCreator": "false",
  "text": "fam it hasn't even been 2 years since this released",
  "likeCount": "41",
  "likeCountText": "41",
  "replyCount": "1",
  "publishedTime": "5 months ago",
  "scrapedAt": "2026-10-15T13:03:15.785Z"
}

Field	Description
videoId	YouTube video ID the comment belongs to
videoUrl	Full watch URL for the video
videoTitle	Video title (present when includeVideoDetails is on)
channel	Channel name that published the video
videoViewCount	Video view count as a number (present when available)
commentId	Unique comment ID
author	Comment author's display name
authorChannelId	Author's channel ID (UC…)
authorIsVerified	"true" if the author's channel is verified
authorIsCreator	"true" if the author is the video's creator
text	Full comment text
likeCount	Like count as a number (e.g. 148000)
likeCountText	Like count as shown on YouTube (e.g. 148K)
replyCount	Number of replies to the comment
publishedTime	Relative post time (e.g. 1 year ago, 5 months ago (edited))
scrapedAt	ISO timestamp of when the comment was scraped
error	Present only when a video has comments disabled or unavailable

Fields are returned as strings for stable CSV/Excel export. videoViewCount populates when the view count is available on the page; a video with comments turned off yields a single row carrying an error note instead of comments.

🔁 Pagination & scale

YouTube serves comments in small batches behind infinite scroll. This scraper calls YouTube's internal (innertube) next endpoint and follows the continuation token from page to page, appending each batch until it reaches your maxCommentsPerVideo limit or runs out of comments. Video discovery works the same way — keyword search and channel listings are paginated via continuation tokens to gather up to maxVideosPerSource videos per source.

The result is large, de‑duplicated result sets — from a handful of comments to hundreds of thousands across many videos — controlled entirely by the maxCommentsPerVideo, maxVideosPerSource and maxConcurrency inputs. Requests retry with back‑off and rotate IPs (when a proxy is enabled) so long runs stay reliable.

⚙️ How to use

1. Open the actor on the Apify Console and click Try for free.
2. Add video URLs, search keywords / topics, or channel URLs (any combination).
3. (Optional) Set maxCommentsPerVideo, sortBy, maxVideosPerSource and maxConcurrency.
4. Click Start and watch the live progress.
5. Download results as JSON, CSV, Excel or HTML table, or pull them from the Apify API.
6. Automate it — run on a schedule, or connect to Make, Zapier, Google Sheets or a webhook to keep a dataset fresh.

Call it from the API

curl -X POST "https://api.apify.com/v2/acts/YOUR_ACTOR_ID/runs?token=YOUR_APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{ "videoUrls": ["dQw4w9WgXcQ"], "maxCommentsPerVideo": 200, "sortBy": "top" }'

$curl "https://api.apify.com/v2/acts/YOUR_ACTOR_ID/runs/last/dataset/items?token=YOUR_APIFY_TOKEN&format=csv"

💡 Tips

• Use top to capture the most‑liked, most‑relevant comments; switch to newest to monitor fresh reactions after a launch or upload.
• Raise maxCommentsPerVideo for deep per‑video analysis; lower it to sample quickly across many videos.
• Combine inputs — pass videos, keywords and channels together to build a broad dataset in one run.
• Filter after export on authorIsVerified or high likeCount to focus on influential voices.
• Enable proxyConfiguration for very large multi‑video runs or if you start hitting rate limits.
• Keep includeVideoDetails on so every comment row is self‑describing for spreadsheets and BI tools.

❓ FAQ

Is scraping YouTube comments legal? This scraper collects only public comment data that anyone can view without logging in. Comments and author names are personal data, so you are responsible for using the output lawfully — follow data‑protection laws such as GDPR and YouTube's terms when storing or processing it. The actor is provided for lawful research and analytics and does not offer legal advice.

How many comments can it scrape, and how fast? From a handful to hundreds of thousands, paginated automatically. Throughput depends on maxCommentsPerVideo, the number of videos and maxConcurrency; a single busy video with thousands of comments typically finishes in a minute or two.

Do I need a YouTube API key or login? No. This is a keyless YouTube comment scraper — no API key, OAuth, Google Cloud project or cookies required. Just add inputs and run.

Do I need proxies? Not for normal use — it runs fine on Apify's default infrastructure. For very large runs or if you hit rate limits, enable Apify Proxy in proxyConfiguration.

Can I filter by top vs newest? Yes. Set sortBy to top for the most‑liked comments or newest for the most recent.

Can I scrape a whole channel's comments? Yes — pass the channel in channelUrls (link, @handle or channel ID) and set maxVideosPerSource to cover as many of its latest videos as you need.

Does it capture replies? It collects every top‑level comment along with each comment's replyCount. Nested reply threads can be added on request.

What formats can I export to? JSON, CSV, Excel or HTML table, or straight from the Apify API as JSON or CSV.

Why does a video sometimes return no comments? Some videos have comments disabled or restricted. Those videos return a single row with an error note so your run never fails silently.

How fresh is the data? Every run scrapes live from YouTube, so results reflect the comments and like counts at the moment the run executes.

🔗 Related scrapers

• YouTube Contact Scraper
• Patreon Scraper
• Bilibili Scraper
• Xiaohongshu / RedNote Scraper
• WeChat Official Account Scraper
• Linktree Scraper

---

Ready to mine the conversation? Click Try for free, add a video, channel or keyword, and export clean YouTube comment data in minutes.