Facebook Posts & Pages Scraper

Public Facebook page metadata and posts in one run. HTTP-first, tokenless by default, optional Graph API token mode for comments and deep pagination.

Features

• 🏢 Page info — name, followers, likes, category, about, website, verified flag
• 📝 Recent posts — text, timestamp, reactions, comment count, shares, media
• 🔓 Tokenless mode — no login, no access token, pure HTTP requests
• 🔑 Graph token mode (optional) — per-emoji reaction breakdown, comments, deep pagination
• 📇 Contact extraction — email, phone, website, address from page/about (always on) and post text (opt-in)
• ⚡ Cheap — HTTP-only (no headless browser): ~0.01–0.05 CU per page
• 🔄 Incremental mode — skip posts already scraped, only emit new ones

---

What it does

Give it Facebook page URLs (or bare usernames like NASA). For each URL the Actor:

1. Fetches the public page over HTTP (no browser, no login) and emits a page item: name, followers, likes, category, about, and publicly listed contact details.
2. Collects ~1–6 recent posts per URL from the page HTML, then optionally fetches permalinks to extend coverage up to maxPosts per page. Each post is emitted with text, timestamp, engagement counts, and media URLs.
3. Optionally regex-extracts emails and phone numbers from post text (off by default for PII minimization).

Optionally, provide a Graph API access token to unlock comments per post, per-emoji reaction breakdowns, and deep pagination beyond what the HTML yields.

---

Tokenless vs Graph token mode

Capability	Tokenless (default)	Graph token
Page info & about	✅	✅
Recent posts (~1–6 per URL)	✅	✅
Extended posts via permalinks (up to maxPosts)	✅	✅
Post reactions total	✅	✅
Per-emoji reaction breakdown	❌	✅
Comments + comment counts	❌	✅ (capped by maxCommentsPerPost)
Deep pagination	❌	✅
Setup	None	Facebook Graph API access token
Cost per page	~0.01–0.05 CU	Slightly higher (per-comment fetching)

---

Input parameters

Parameter	Type	Default	Description
startUrls	array (required)	—	Facebook page URLs or usernames (e.g. https://www.facebook.com/NASA or NASA). Max 100.
maxPosts	integer	20	Max posts per page (1–500). Tokenless: HTML yields ~4–6; permalink fetches extend up to this cap.
since	string	—	ISO 8601 date. Posts older than this are skipped.
until	string	—	ISO 8601 date. Posts newer than this are skipped.
keyword	string	—	Only keep posts whose text contains this keyword (case-insensitive). Max 200 chars.
includePageInfo	boolean	true	Emit a page dataset item per start URL.
extractContactsFromPosts	boolean	false	Regex-extract emails/phones from post text (not comments). Page/about contact extraction is always on. Off by default for PII minimization.
incremental	boolean	false	Skip posts already emitted by a previous run (key-value store state). Only new posts are pushed.
graphAccessToken	string (secret)	—	Optional Graph API token. Unlocks comments, reaction breakdown, deep pagination. Leave empty for tokenless HTTP mode.
maxCommentsPerPost	integer	50	Cap on comments fetched per post in Graph mode (1–200). Prevents runaway cost on viral posts.
proxyConfiguration	object	—	Residential proxy recommended for Apify datacenter IPs. Facebook blocks datacenter ASNs aggressively.

Example input

{
  "startUrls": ["https://www.facebook.com/NASA", "nike"],
  "maxPosts": 30,
  "keyword": "launch",
  "includePageInfo": true,
  "extractContactsFromPosts": false,
  "incremental": true
}

---

Output

Each dataset item carries a type field (page or post) plus a status / httpStatus result indicator.

Page item

{
  "type": "page",
  "status": "SUCCESS",
  "httpStatus": 200,
  "pageId": "1234567890",
  "pageUrl": "https://www.facebook.com/NASA",
  "name": "NASA",
  "followers": 75400000,
  "likes": 73000000,
  "category": "Government organization",
  "about": "Explore the universe and discover our home planet with NASA.",
  "email": "public-inquiries@nasa.gov",
  "phone": "+1 202-358-0001",
  "website": "https://www.nasa.gov",
  "address": "300 E St SW, Washington, DC 20546",
  "messenger": "https://m.me/NASA",
  "verified": true
}

Post item (tokenless)

{
  "type": "post",
  "status": "SUCCESS",
  "httpStatus": 200,
  "pageId": "1234567890",
  "postId": "1234567890_10159876543210000",
  "url": "https://www.facebook.com/NASA/posts/10159876543210000",
  "text": "Our newest mission is a go! Read more about what's next for exploration…",
  "timestamp": "2026-10-08T14:32:00.000Z",
  "reactionsTotal": 48210,
  "reactions": {},
  "commentsCount": 512,
  "comments": {},
  "shares": 3105,
  "media": ["https://scontent.xx.fbcdn.net/.../photo.jpg"],
  "extractedContacts": {}
}

Field notes

• reactions (per-emoji breakdown) and comments (comment objects) are only populated in Graph token mode; they stay empty {} in tokenless mode.
• extractedContacts applies to post items only: extracted from post text when extractContactsFromPosts=true (default false). Page/about contact fields (email, phone, address, messenger) come from the page's own published info.
• httpStatus reflects the last HTTP response for the item; status is one of SUCCESS, PARTIAL, NOT_FOUND, BLOCKED, ERROR.

---

Use cases

• Brand monitoring — track followers, engagement, and recent posts on your own and partner pages; keyword-filter for mentions.
• Competitor analysis — compare categories, follower counts, posting cadence, and top-performing posts across rival pages.
• Market research — profile public pages in a niche (categories, contact info, audience size) in bulk.
• Lead extraction — harvest publicly listed emails, phones, and websites from business pages; opt into post-text contact extraction when relevant.

---

FAQ

Do I need to log in? No. Tokenless mode never logs in and never handles cookies. All data comes from publicly served page HTML. A Graph API token is optional and unlocks extra fields — you obtain it from Facebook yourself; the Actor never stores it.

What are the tokenless limits? Tokenless HTML yields roughly 1–6 recent posts per URL (typically ~4–6), extendable toward maxPosts via permalink fetches. No comments, no per-emoji reaction breakdown, no deep pagination in this mode.

Does incremental mode work? Yes — set incremental: true and the Actor records emitted post IDs in its key-value store. Subsequent runs push only new posts.

How does contact extraction work? Emails/phones/websites/addresses found on the page and its about information are always extracted. Post text is only scanned when extractContactsFromPosts is enabled (default false, for PII minimization).

Do I need a proxy? Strongly recommended. Facebook blocks datacenter IP ranges aggressively. Configure proxyConfiguration (residential recommended) for reliable runs — Apify's default residential proxy works out of the box.

---

Cost & performance

• HTTP-first — every request is plain HTTP via got-scraping with browser-like headers. No headless browser, no CDP, no page rendering.
• Cost: ~0.01–0.05 CU per page in tokenless mode (page fetch + up to maxPosts cheap permalink fetches).
• Runtime: tens of seconds for a handful of pages; scales linearly with URLs and maxPosts.
• To keep costs down: reduce maxPosts, use since/until windows, and enable incremental for repeated monitoring runs.

---

Personal data & lawful basis

[!IMPORTANT] This Actor extracts publicly available Facebook data, which may include personal data (names of page admins implied by page names, contact details listed on pages, usernames in post text).

• PII disclosure: Page-level contact fields (email, phone, address, messenger) are published by page owners voluntarily. extractedContacts from post text may contain personal contact details of commenters/authors — that is why it is off by default.
• Buyer responsibility: You, the buyer, are the data controller for whatever you scrape. Under GDPR and similar laws you need a lawful basis (e.g. legitimate interest assessment) before processing personal data. Do not scrape private profiles, do not build contact lists for spam, and honor data-subject rights (deletion requests) for data you store.
• This Actor only accesses content Facebook serves without authentication. If you cannot articulate a lawful basis for your use case, do not run it.

---

Limitations

• Tokenless mode cannot read posts older than what the public page serves.
• Private pages, groups, and login-gated content are never accessed.
• Facebook markup changes can intermittently affect extraction — the status/httpStatus fields on each item let you monitor success rates.

---

License

LICENSED — all rights reserved. © 2026 REXREUS