Facebook Posts Scraper

Scrape Facebook Page and Profile posts — text, reactions, comments, shares, video and photo media, and structured hashtag/link entities — plus a dedicated cheap lookup for a single post or Reel by its permalink URL. No login, no cookies, no proxy setup — just paste a Facebook URL and run. Works out of the box on the free Apify plan.

What this actor does

• Two input modes in one run: paste a Page/Profile URL to scrape its recent posts (including Reels), or a single post permalink to fetch just that one post — mix both in the same startUrls list
• Single-post lookups are genuinely cheap: a post permalink page is fully rendered by Facebook itself, so this actor fetches it directly with no browser — faster and lower-cost than a full Page/Profile scrape. (Standalone /reel/... links specifically are less reliable this way — see the FAQ.)
• Real server-side date filtering: onlyPostsNewerThan/onlyPostsOlderThan are applied by Facebook itself, not discarded after the fact — large date-scoped runs don't need to scroll through everything in between
• Full reaction breakdown: all 7 real Facebook reaction types (Like, Love, Care, Haha, Wow, Sad, Angry), not a partial subset
• Real video captions: for video/reel posts, fetch Facebook's own auto-generated .srt caption file as a transcript (optional toggle)
• Structured text entities: hashtags, mentions, and external links are parsed out of the post text automatically — external links are the clean destination URL, not Facebook's tracking redirect
• Direct video download links: HD and SD progressive MP4 URLs for video/reel posts, not just a thumbnail
• Empty fields are omitted, not returned as null

Output

Every target produces one record per post found (a single-post/reel URL always produces exactly one; a Page/Profile URL produces up to maxItems).

Post fields

• postId — the post's numeric Facebook ID
• permalinkUrl — direct link to the post
• postType — photo, video, reel, link, or text
• text — the post's full message text
• hashtags — hashtags found in the post text
• mentions — tagged people/Pages found in the post text
• externalLinks — external links shared in the post, as the real destination URL (not Facebook's l.facebook.com tracking redirect)
• creationTime — Unix timestamp of when the post was published
• author — {id, name, url, profileUrlIsFallback}; profileUrlIsFallback is true when Facebook didn't expose a direct profile URL and this actor constructed one from the author's numeric ID instead
• attachments — media on the post:
  • Photos: url, width, height, accessibilityCaption (Facebook's own OCR/alt-text)
  • Videos/reels: videoUrlHd, videoUrlSd (direct downloadable MP4 links), thumbnailUrl, durationMs, width, height, isReel, captionsAvailable (list of {locale, languageName, url}), and — when fetchVideoCaptions is on — videoTranscript (the real .srt file content) and videoTranscriptLocale
• reactionsTotal — total reaction count
• reactionBreakdown — [{type, count}, ...] covering all reaction types Facebook returns for the post (not capped at a fixed set)
• commentsCount — total comment count (including replies)
• sharesCount — total share count
• isAd — true if Facebook flags the post as sponsored
• topComments — top-level comments (present when includeTopComments is on): {text, author, createdTime, reactionsTotal, replyCount}
• allComments — up to 100 top-level comments per post (present when fetchAllComments is on, Page/Profile batches only), same shape as topComments
• inputUrl — the original URL from your input
• scrapedAt — ISO 8601 timestamp of when this record was scraped

Input

Field	Type	Default	Description
startUrls	array	[{"url": "https://www.facebook.com/nike"}]	Facebook Page/Profile URLs and/or single post/reel permalink URLs
maxItems	integer	20	Posts to fetch per Page/Profile URL (ignored for single-post/reel URLs)
onlyPostsNewerThan	string	–	Only posts published after this date: 2026-01-01, full ISO 8601, or relative (1 day, 2 weeks, 3 months)
onlyPostsOlderThan	string	–	Only posts published before this date, same formats
includeTopComments	boolean	true	Attach each post's top-level comments (no extra request)
fetchVideoCaptions	boolean	false	Download the real .srt caption file for video/reel posts as a transcript
fetchAllComments	boolean	false	Page/Profile batches only: fetch up to 100 top-level comments per post as allComments (one extra request per post)

Example: recent posts from a Page

{
  "startUrls": [{ "url": "https://www.facebook.com/nike" }],
  "maxItems": 30
}

Example: a single post by URL

{
  "startUrls": [
    { "url": "https://www.facebook.com/bbcnews/posts/pfbid02wicKt1z7teFytZajyGT72VCz5tcuVQmCZC4HrZTHZZ8SEyxZBgfxMn37pnRey6Sql" }
  ]
}

Example: posts from the last 30 days, with video transcripts

{
  "startUrls": [{ "url": "https://www.facebook.com/nike" }],
  "onlyPostsNewerThan": "30 days",
  "fetchVideoCaptions": true
}

Example: mixing a Page batch and a single reel lookup in one run

{
  "startUrls": [
    { "url": "https://www.facebook.com/nike" },
    { "url": "https://www.facebook.com/reel/752115997659579/" }
  ]
}

Use cases

• Social media monitoring — track a brand or competitor Page's recent posts and engagement
• Content research — pull a Page's or Profile's post history for analysis
• Fact-checking and journalism — fetch a specific viral post by URL with its exact text, reactions, and comments
• Video content archiving — download direct MP4 links and captions for video/reel posts
• Sentiment and engagement analysis — reaction-type breakdowns and comment data for a batch of posts
• Link and hashtag tracking — structured extraction of hashtags and external links shared in posts

FAQ

Why does a single post URL return results faster than a Page/Profile URL? A post permalink page is fully rendered by Facebook itself, so this actor fetches it directly with a lightweight HTTP request — no browser needed. A Page/Profile URL only shows one post without scrolling, so scraping more requires a real browser session to trigger Facebook's own pagination.

A standalone /reel/... URL sometimes returns nothing — will that be fixed? This is a genuine, confirmed Facebook restriction, not a bug in this actor: Facebook applies significantly stricter anti-bot checks to a Reel reached by pasting its permalink directly than to the exact same Reel reached by scrolling through a Page's timeline. This actor automatically retries a blocked single-Reel lookup before giving up — but confirmed via live testing across multiple accounts and previously-untouched Reel IDs that Facebook can still block the standalone deep link even after retrying. If you need Reels reliably, scrape them via their Page/Profile URL instead of a standalone /reel/... link — Reels returned as part of a Page/Profile batch (via startUrls) are not affected by this restriction.

Why only 7 reaction types, and why does a post sometimes show fewer? Facebook has exactly 7 reaction types (Like, Love, Care, Haha, Wow, Sad, Angry). This actor returns whichever of them a post actually received — a post with no Angry reactions simply won't have an Angry entry in reactionBreakdown, not a zero.

Are video view/play counts included? No. Facebook does not reliably expose video view counts on the public post surface this actor reads, so this field is never included — we don't fabricate or guess a number.

Is the video transcript AI-generated? No — videoTranscript (when fetchVideoCaptions is on) is the real .srt caption file Facebook itself generated for the video, not independent speech-to-text transcription.

Why is a post author's profile URL sometimes constructed rather than a direct Facebook link? Facebook doesn't always expose a direct profile URL for a post's author. When that happens, this actor builds a valid profile.php?id=... URL from the author's numeric ID and sets profileUrlIsFallback: true so you know it was constructed rather than returned directly by Facebook.

Does this work on private profiles or restricted posts? No — a private, restricted, or removed target simply produces no record for that URL, rather than a fabricated empty-looking result.

Do I need cookies or to log in? No. This actor reads only Facebook's publicly accessible pages — no login, no cookies, ever.

Is this actor affiliated with Facebook or Meta? No. This is an independent, third-party actor that reads Facebook's own publicly accessible pages; it is not affiliated with, endorsed by, or operated by Meta Platforms, Inc.

How fresh is the data? Every run fetches live from Facebook at request time — there's no caching, so results reflect what Facebook shows at the moment the actor runs.

Are the comment/reaction/share counts accurate for posts with thousands of comments? Yes. These are Facebook's own aggregate totals, not something derived by counting visible items, so they're exact regardless of scale — verified directly against a real, logged-in view of a post with 4,169 comments: commentsCount, reactionsTotal, sharesCount, and every value in reactionBreakdown matched exactly. (The topComments/allComments arrays themselves are intentionally capped previews, not the full comment thread — see the field list above.)