Facebook Comments Scraper by Post URL: Text, Likes, Replies avatar

Facebook Comments Scraper by Post URL: Text, Likes, Replies

Pricing

from $1.39 / 1,000 comment returneds

Go to Apify Store
Facebook Comments Scraper by Post URL: Text, Likes, Replies

Facebook Comments Scraper by Post URL: Text, Likes, Replies

For sentiment analysis, giveaway picking and brand monitoring: paste public Facebook post, video or reel URLs and get one row per comment and reply with text, date, reactions by type and commenter name, as CSV, Excel or JSON. Newest first. No login, no start fee. You pay only for comments returned.

Pricing

from $1.39 / 1,000 comment returneds

Rating

0.0

(0)

Developer

NeverEmpty

NeverEmpty

Maintained by Community

Actor stats

0

Bookmarked

17

Total users

1

Monthly active users

18 days ago

Last modified

Features

Limited permissions

Share

For sentiment analysis, giveaway picking and brand monitoring: paste public Facebook post, video or reel URLs and download their comments as CSV, Excel or JSON, one row per comment or reply with text, date, reactions by type, commenter name and verified badge. No login, no cookies, no start fee, no extra charge for the date filter. Comments come newest first by default, including the ones Facebook's default view leaves out. You pay $2.49 per 1,000 comments on the Free plan, down to $1.39 on Gold and above, and only for comments returned.

What can I do with this Facebook comments scraper?

  • Read what people say about a post. Collect the comments on your own Page posts, a campaign or a competitor's launch, and sort them by reactions to see what stands out.
  • Run sentiment analysis and topic research. Get clean comment text with dates, reaction counts and reaction types (like, love, haha, wow, sad, angry, care), ready for a spreadsheet, an AI model or a dashboard.
  • Follow the conversation, not just the top level. Replies that Facebook lists come in their own rows with isReply and parentCommentId, so you can rebuild those threads.
  • Pick giveaway entrants. Get the names of people who commented, with a profile link where Facebook shows one.
  • Check videos and reels too. Posts, photos, videos, reels, watch/?v= links, permalink.php links, /share/ links and fb.watch links all work with the same input.
  • Follow up on new comments only. Set onlyNewerThan to 7 days and pay only for recent comments.

What data do you get?

GroupFields
CommentcommentId, commentUrl, position, text, commentType (text, attachment, no-text), attachmentType, likesCount, reactions, replyCount, createdAt, createTime
ThreadisReply, depth, parentCommentId
CommenterauthorName, authorProfileUrl, authorIsVerified, authorType (person or page)
PostpostUrl, postId, postOwnerName, postCommentCount, inputUrl
Runstatus, note, sortOrder, source, scrapedAt. The last row can also hold notCheckedUrls (URLs to put in a new run)

position is the place of the row in the list read for that post, starting at 1. Replies are placed after the comments of the same page. likesCount is the total of all reaction types, and reactions splits it by type. postCommentCount is the comment count Facebook shows on the post.

Each post that was read also gets one free post-summary row with commentsReturned, repliesReturned, pagesRead and postCommentCount, so you can compare what was returned with the count on the post.

How to scrape Facebook comments from a post URL

  1. Paste your post, video and reel URLs into urls, one per line. Links with tracking parameters work. startUrls, directUrls and postUrls are accepted as other names for urls.
  2. Set maxCommentsPerPost to the most rows you want from each post (comments and replies together).
  3. Choose sortOrder and whether to includeReplies.
  4. If you only want recent comments, set onlyNewerThan to a date such as 2026-10-01 or a span such as 7 days.
  5. Press Start, then download the rows as JSON, CSV or Excel, or read them through the API.

Input

FieldWhat it doesDefault
urlsPublic Facebook post, video and reel URLs, one per line. Up to 1,000 are read per run.One free sample post when no URLs are given
maxCommentsPerPostReading a post stops after this many rows, comments and replies together (1 to 100,000).500
includeRepliesAlso return the replies under each comment, each in its own row with isReply: true.On
sortOrdernewest (newest first, including comments the default view leaves out), all (Facebook's ranked order without its usual filter) or relevant (Facebook's default Most relevant view, which shows fewer comments).newest
onlyNewerThanReturn only comments newer than a date (2026-10-01) or a span back from now (12 hours, 7 days, 2 weeks, 3 months). Older comments are skipped and not charged.Empty (any date)
{
"urls": [
"https://www.facebook.com/NASA/posts/1671137804381589/",
"https://www.facebook.com/watch/?v=1234567890"
],
"maxCommentsPerPost": 500,
"includeReplies": true,
"sortOrder": "newest",
"onlyNewerThan": "30 days"
}

Output example

{
"status": "ok",
"inputUrl": "https://www.facebook.com/NASA/posts/1671137804381589/",
"postUrl": "https://www.facebook.com/NASA/posts/1671137804381589/",
"postId": "1671137804381589",
"postOwnerName": "NASA",
"postCommentCount": 12,
"commentId": "1234567890123456",
"commentUrl": "https://www.facebook.com/NASA/posts/1671137804381589/?comment_id=1234567890123456",
"position": 1,
"text": "What a view of the launch!",
"commentType": "text",
"attachmentType": null,
"likesCount": 14,
"reactions": { "like": 11, "love": 3 },
"replyCount": 2,
"createdAt": "2026-10-28T15:04:31.000Z",
"createTime": 1793199871,
"isReply": false,
"depth": 0,
"parentCommentId": null,
"authorName": "Orbit Watcher",
"authorProfileUrl": "https://www.facebook.com/orbit.watcher",
"authorIsVerified": false,
"authorType": "person",
"sortOrder": "newest",
"source": "facebook-comment",
"scrapedAt": "2026-11-01T09:12:44.000Z"
}

The comment, names, IDs and counts in this example are made up to show the shape of a row.

How much does it cost?

$2.49 per 1,000 comments on the Free plan, with lower prices on paid Apify plans:

Apify planPrice per 1,000 comments
Free$2.49
Bronze$1.99
Silver$1.69
Gold and above$1.39

There is no start fee, no charge per post and no extra charge for onlyNewerThan or sortOrder. Replies are charged like comments. You are charged only for rows with status: "ok", and each comment is returned once per run. Every other row is free, and the list is below.

To keep a run inside a budget, set maxCommentsPerPost, or set a maximum total charge when you start the run. The run stops at that amount and says so in a budget-reached row.

FAQ

How do I export Facebook comments to Excel or CSV? Paste the post URL into urls and run the Actor. When the run ends, open the Output tab and download the dataset as CSV or Excel. Each comment and each reply is one row.

Can I scrape Facebook comments without logging in? Yes. You don't need a Facebook account, a cookie or an API key.

Can I get the replies to a comment? Yes, the replies Facebook lists, one level deep. Not every reply is listed, so the count can be lower than on the post. With includeReplies on, replies come in their own rows with isReply: true and the parentCommentId of the comment they answer.

How many comments can I get from one post? As many as Facebook lists for that post to visitors who are not logged in, up to your maxCommentsPerPost. This is often fewer than the count shown on the post; see Limitations.

Can I scrape comments from Facebook reels and videos? Yes. Use the reel or video URL, such as https://www.facebook.com/reel/1234567890/ or https://www.facebook.com/watch/?v=1234567890. Reels, videos and posts return the same fields.

Can I run this from an API or an AI agent? Yes. Call the Actor through the Apify API, or from an AI agent through the Apify MCP server, with a list of URLs. The rows come back as JSON.

How do I scrape comments from many posts at once? Paste a list of URLs (the limit is 1,000 per run). Every post the run reaches returns the comments Facebook lists for it, or one free row that says why there are none. If Facebook stops answering, the run ends early, and the posts it did not reach are listed in notCheckedUrls in the last row, ready for a new run a few hours later.

What happens if a run reaches its timeout? The run stops about 90 seconds before its timeout. Comments already returned stay in the dataset. A post that was cut short gets a stopped-reading row, and posts it did not start are listed in notCheckedUrls in the last row. To read them, start a new run with those URLs, or set a longer timeout.

Why is text empty on some rows? commentType tells you why. attachment means the comment is only a sticker, a photo, a GIF or a link, and attachmentType says which. no-text means Facebook gave the comment without text or an attachment. These rows are still real comments and are charged like any other.

Rows you are not charged for

A post that was read but did not return comments gets its own row, so you can see what happened to it. Posts the run did not check (past the 1,000 URL limit, past the maximum total charge, stopped because Facebook stopped answering, or stopped before the run timeout) are counted in one last row, and notCheckedUrls lists them for a new run.

statusWhat it means
post-summaryOne row per post that was read: how many comments and replies were returned, how many pages were read, and the count shown on the post.
sampleNo URLs were given, so up to 20 comments of one sample post were read to show the output. These rows are free. Add your own URLs to urls.
no-public-postFacebook shows no public post at this address. It may be deleted, not public, not available in some countries, or mistyped.
login-requiredFacebook asks visitors to log in before it shows this post, for example a post in a group, or a post that was deleted or never public.
comments-not-shownFacebook shows the post but has comments turned off for it.
no-comments-listedThe post has no comments listed, or none newer than onlyNewerThan.
max-comments-reachedmaxCommentsPerPost rows were returned and Facebook showed that the post may list more. Raise the limit to read further.
stopped-readingReading stopped part way, because Facebook did not answer the next page or the run was about to reach its timeout. The comments before that point are in the dataset.
blockedFacebook did not let this run open the post. Run the URL again later.
unreadableThe page loaded, but it showed a login page or the comment data was not in the expected form. In our tests this was 1 public post in about 300; run the URL again later. Nothing is guessed.
invalid-inputThe entry is not a Facebook post, video or reel URL, such as a Page or profile link, or an input value could not be read.
duplicateThe same post was already in this run's list.
not-checkedOne row at the end: how many posts were not read and why. The URLs are in notCheckedUrls.
budget-reachedThe run reached the maximum total charge you set. Comments that did not fit are not in the dataset and were not charged.

Limitations

  • The Actor returns the comments Facebook lists to visitors who are not logged in. This is often fewer than the count shown on the post (postCommentCount), which also counts comments Facebook hides. In our tests on 1 November 2026, a post with about 3,200 comments returned 85% of that count in newest-first order, and posts with 12 to 90 comments returned 58% to 100% of it. relevant returns fewer.
  • Replies are read one level deep. Answers posted under a reply are not in the output.
  • With onlyNewerThan and newest order, reading a post stops at the first page of comments that are all older than the date, so a new reply under an older comment may be missed.
  • Only public posts can be read. Posts in groups, and posts that are deleted or not public, return a free login-required or no-public-post row. Facebook shows the same login page for a deleted post as for a post that needs a login, so the two cannot always be told apart.
  • likesCount is the total of all reaction types, not only likes. reactions holds the split for the types Facebook lists.
  • Numeric account IDs and profile links that contain one are not collected, so authorProfileUrl is empty for many commenters (52% to 70% of comment rows in our tests).
  • On long lists Facebook may stop answering, so the run can end early. The rest are not charged and are listed in notCheckedUrls.
  • Email addresses and phone numbers in comment text are masked on a best-effort basis. Not every way of writing them can be caught, so some may remain. Masking can also go too far: a long number such as a reference code may be masked as well.
  • Use the data in line with the rules that apply to you, including privacy law for the personal data in comments.

Need help or a missing field?

Open the Issues tab and tell us what you need. If a row looks wrong, include the run ID.

Thanks for using this Actor

We build these tools for people who run them every day, and we improve them from what users tell us.

  • Missing a field, or need another filter? Tell us in the Issues tab. If the data is there, we add it.
  • Found a bug or a wrong value? Post the run ID in the Issues tab. Wrong data is the thing we fix first.

If this Actor saved you time, a short review helps other people find it.