Facebook Comments Scraper avatar

Facebook Comments Scraper

Pricing

from $1.79 / 1,000 results

Go to Apify Store
Facebook Comments Scraper

Facebook Comments Scraper

Scrapes comments from public Facebook posts by URL — comment text, commenter name and profile link, reaction count and timestamp — with a comment limit you can set per post.

Pricing

from $1.79 / 1,000 results

Rating

0.0

(0)

Developer

Mina

Mina

Maintained by Community

Actor stats

1

Bookmarked

45

Total users

2

Monthly active users

2 days ago

Last modified

Features

Limited permissions

Share

What does Facebook Comments Scraper do?

Facebook Comments Scraper collects the comments on public Facebook posts from nothing more than the post's link, with no cookies and no Facebook login. For every comment it saves the comment text, the commenter's name, profile link and profile picture, the reaction count, the time it was posted and the comment's id. You set how many comments to collect per post, and one run can read many posts.

Try it with the prefilled post: open the Actor, click Start, and its comments arrive in the Output tab. Because it runs on the Apify platform, you also get API access, scheduled runs, integrations (Google Sheets, Make, Zapier, webhooks), run monitoring and residential proxies built in, so there is nothing to set up to read Facebook.

Why scrape Facebook comments?

  • Sentiment analysis: feed what people say under a brand's, a competitor's or a public figure's posts into a sentiment or topic model.
  • Customer feedback: find the complaints, questions and product requests left under your Page's posts.
  • Campaign tracking: compare how people respond to different posts and announcements, by what they write and how many reactions their comments get.
  • Social listening and research: collect the public discussion around news, events or launches for analysis.
  • Community management: export comments to review or report on them outside Facebook.

How to scrape comments from a Facebook post

  1. Open a public Facebook post, video or reel in your browser and copy its link from the address bar, for example https://www.facebook.com/<page>/posts/<post id> or https://www.facebook.com/reel/<reel id>.
  2. Paste it into Start URLs. Add as many post links as you like, one per row.
  3. Set Comments Limit, the most comments to save from each post (100 by default).
  4. Click Start. Up to 5 posts are read at the same time.
  5. Download the comments when the run finishes, or read them through the API.

Input

Everything is set on the Input tab. In JSON, the input looks like this:

{
"start_urls": [
{ "url": "https://www.facebook.com/OrangeEgyptOfficial/posts/pfbid02dMgTpxN321sqGFEaZCtwToNVBjXZHocyzwXhm2cy2khE3j8JuhvJSnZwx9mq9y7pl" },
{ "url": "https://www.facebook.com/<page>/posts/<post id>", "comments_limit": 500 }
],
"comments_limit": 100
}
FieldTypeDefaultWhat it does
start_urlsarray(required)Links to public Facebook posts, videos or reels, each as a {"url": "..."} object.
comments_limitinteger100The most comments to save from each post, from 1 up.

In JSON input, a post's object can carry its own comments_limit, which replaces the run-wide limit for that post only, as in the second link above. Add each link as its own row: loading links from a remote text file is not supported.

Output

One row per comment. An example row (the commenter, ids and text are made up):

{
"comment_id": "1093847261538402",
"gender": "FEMALE",
"user_id": "100004829105736",
"user_name": "Jane Sample",
"user_url": "https://www.facebook.com/jane.sample.example",
"avatar": "https://scontent.fcai19-1.fna.fbcdn.net/v/t39.30808-1/123456789_n.jpg",
"content": "Does this offer work on prepaid lines too?",
"text": "Does this offer work on prepaid lines too?",
"likes_count": "3",
"create_time": 1784827441000,
"post_url": "https://www.facebook.com/OrangeEgyptOfficial/posts/pfbid02dMgTpxN321sqGFEaZCtwToNVBjXZHocyzwXhm2cy2khE3j8JuhvJSnZwx9mq9y7pl"
}

Every row carries the link of the post it came from, so comments from several posts in one run can be told apart. You can download the dataset in various formats such as JSON, HTML, CSV, or Excel, or read it through the Apify API.

Data fields

FieldTypeDescription
comment_idstringFacebook's id of the comment.
genderstring or nullThe commenter's gender as Facebook reports it, such as MALE or FEMALE.
user_idstringThe commenter's Facebook id. For some people Facebook shows logged-out visitors only a scrambled id starting with pfbid instead of the numeric one.
user_namestringThe commenter's display name.
user_urlstring or nullLink to the commenter's profile. Null when Facebook gives no profile link.
avatarstringLink to the commenter's profile picture.
contentstring or nullThe comment text. Null when the comment has no text, such as a sticker or photo on its own.
textstring or nullThe same text as content.
likes_countstringNumber of reactions on the comment, as text, exactly as Facebook shows it, so large counts can come abbreviated (such as 1.2K).
create_timeintegerWhen the comment was posted, as a Unix timestamp in milliseconds.
post_urlstringThe post link you gave.

How much does it cost to scrape Facebook comments?

You pay per comment saved to the dataset; see the Pricing tab for the exact price. The price per comment drops on higher Apify plans, and the Pricing tab lists each tier. Each run also has a tiny charge when it starts. Posts that were deleted, could not be read or have no public comments add nothing more, because only comments are saved. The residential proxies the Actor needs are included. Comments Limit caps what each post can cost, and the Actor stops by itself when a run reaches the maximum cost you set for it.

Tips for scraping Facebook comments

  • Set Comments Limit to what you need. The Actor stops paging through a post's comments as soon as the limit is reached, so a low limit makes runs both shorter and cheaper.
  • Put many posts in one run. Up to 5 posts are read at the same time, and you pay the start charge once instead of once per post.
  • Use the post's own link. Open the post itself (for example by clicking its date), so the address bar shows the post and not the Page's timeline.
  • Convert the timestamps. create_time is in milliseconds. Divide it by 1000 for a Unix timestamp in seconds, or turn it into a date in a spreadsheet with =A2/86400000+DATE(1970,1,1).
  • Collecting new comments every day? Schedule the run. Each run collects up to the limit again, so the same comment can appear in several runs: de-duplicate on comment_id.

FAQ, disclaimers, and support

Do I need a Facebook account or cookies? No. The Actor reads posts the way a logged-out visitor sees them, so you never share a login. That also means it can only read posts Facebook shows to logged-out visitors: posts that are visible only to friends, or posted in private groups, can't be read.

Is it legal to scrape Facebook comments? The Actor collects only comments that Facebook shows publicly, without logging in. Comments contain personal data, such as commenters' names, profile links, pictures and gender, which is protected by the GDPR in the EU and by other laws elsewhere. Only collect personal data if you have a legitimate reason to. Web data scraping is legal in many cases, but if you are unsure, consult a lawyer.

Are replies to comments included? No. The Actor saves the comments listed directly under the post; replies to those comments are not collected.

Does it work on videos and reels? Yes. Post, video (/videos/), reel (/reel/) and watch/?v= links all work. A video or reel page also lists suggested videos; the Actor reads the comments of the video you linked, not theirs.

Why did I get fewer comments than the post shows? The number on a post counts replies too, and on some posts Facebook shows logged-out visitors only part of the comments. The prefilled post, for example, shows 123 but serves 14 comments to anyone not logged in. The Actor saves every comment Facebook serves, up to your limit.

Why did a post give no comments? Either the post has no public comments, or it could not be read. A post that was deleted, or whose link is wrong, is listed as not found in the FAILED_INPUTS record of the run's key-value store. A post that failed for another reason is listed there too, with the reason. Facebook sometimes answers with a page that lacks the post's data, so each post is tried with up to 3 fresh residential IPs before it counts as failed; trying it again in a later run may work. The run's status message sums up what happened, and the run only fails when every post failed with an error.

Why are some user_url values empty, or user_id values long codes? Facebook hides some people's profile links and numeric ids from logged-out visitors. The Actor saves what Facebook shows.

Found a bug or need another field? Open an issue on the Issues tab and it will be looked at. Need a custom scraper or a different output format? Get in touch through the Issues tab too.