Facebook Business Page Finder by Keyword & City

Find Facebook business pages either by pasting direct page URLs or by discovering them through a keyword and city, then get back page identity, engagement metrics, contact details, business hours, ratings, and reviews for each one.

What this actor does

• Scrapes direct Facebook page URLs, or discovers pages by combining search terms with locations (e.g. "Coffee Shop" + "London")
• Returns page identity (name, ID, URL, categories), engagement (followers, likes, check-ins), and contact details (phone, email, address, website, Messenger link) when the page owner has listed them
• Pulls business hours, star/recommend rating, and an ad library reference for each page
• Optionally fetches real reviews per page (reviewer name, recommend verdict, review text, date) via includeReviews
• Scales to large result sets by combining multiple search terms with multiple locations and raising maxItems
• Pages that turn out to be deleted, unavailable, or shown only as a login wall are skipped entirely rather than appearing as empty rows

Input

Field	Type	Default	Description
startUrls	Array of {url}	–	Direct list of Facebook page URLs to scrape — skips discovery for these. Either this or searchTerms is required
searchTerms	Array of strings	–	Plain search terms for page discovery (e.g. ["Coffee Shop", "Bakery"]). Either this or startUrls is required
locations	Array of strings	–	Optional locations combined with each search term (e.g. ["London", "Manchester"])
maxItems	Integer	60	Maximum number of page rows to output (1–5000)
includeReviews	Boolean	true	Fetch and include up to 10 reviews per page (adds one extra request per page)
cookies	String	–	Facebook session cookies — strongly recommended. See Cookies below

Example Input — direct URLs

{
  "startUrls": [
    { "url": "https://www.facebook.com/McDonalds" },
    { "url": "https://www.facebook.com/BBCNews" }
  ],
  "cookies": "..."
}

Example Input — keyword + location discovery

{
  "searchTerms": ["Coffee Shop", "Cafe", "Bakery"],
  "locations": ["London", "Manchester", "Birmingham"],
  "maxItems": 300,
  "cookies": "..."
}

That single input produces 9 search combinations (3 terms × 3 cities), each yielding up to ~40 discovered URLs — giving a pool of up to ~360 candidates to fill 300 results from. startUrls and searchTerms can also be combined in the same run — results from both are merged into the same output dataset.

Cookies (Authentication)

Strongly recommended for any run.

Without session cookies, Facebook serves a compact login-wall stub for many discovered pages — you'll get some data but contact fields (phone, address, website) are gated. Providing cookies dramatically increases both field coverage and the number of pages you can successfully retrieve.

How to export cookies from your browser:

1. Log in to Facebook in Chrome or Firefox
2. Open DevTools → Application → Cookies → facebook.com
3. Export the cookies for .facebook.com using a browser extension like EditThisCookie (export as JSON) or Cookie Quick Manager on Firefox
4. Paste the exported JSON array into the cookies field

Minimum cookies needed: c_user, xs, datr, fr, sb.

Accepted formats:

1. JSON array (recommended — export directly from your browser extension):

   [
     { "name": "c_user", "value": "100074...", "domain": ".facebook.com", "path": "/", "expires": 1815212390, "httpOnly": false, "secure": true },
     { "name": "xs", "value": "8%3A2vaT...", "domain": ".facebook.com", "path": "/", "httpOnly": true, "secure": true },
     { "name": "datr", "value": "7JhPar...", "domain": ".facebook.com", "path": "/", "httpOnly": true, "secure": true }
   ]
   

2. Cookie header string: "c_user=100074...; xs=8%3A...; datr=7JhP..."

Cookies expire after a while, or when you log out of that browser session. If you see high skip rates, refresh your cookies.

Scaling to 1K+ Results

The actor is designed to scale — here is the fastest path to large datasets:

1. Use multiple search terms and locations. Each (term, location) pair runs its own discovery pass. Example: 5 terms × 10 cities = 50 combinations × up to ~40 pages each = up to ~2,000 candidates in the pool.

2. Set maxItems to your target. The actor over-discovers ~2× maxItems URLs to absorb login-walls and fetch failures without leaving you short.

3. Provide session cookies. This is essential at scale — cookies enable Facebook's own search as the primary discovery source, which is far more reliable than discovering pages through other search engines, and unlock contact fields on login-gated pages.

4. Vary your search terms. Generic terms (e.g. "Restaurant") produce many duplicate pages across locations. More specific terms (e.g. "Italian Restaurant", "Sushi Bar") yield more unique pages.

Example for 1K results:

{
  "searchTerms": ["Restaurant", "Cafe", "Bar", "Pub", "Bistro"],
  "locations": ["London", "Manchester", "Birmingham", "Glasgow", "Leeds", "Bristol", "Edinburgh", "Liverpool", "Sheffield", "Cardiff"],
  "maxItems": 1000,
  "cookies": "..."
}

50 search combinations, each yielding up to ~40 unique URLs — a pool of up to ~2,000 candidates. How many become rows depends on how many of them Facebook actually serves; cookies make the biggest difference.

Output

Page Identity

Field	Type	Description
facebookUrl	String	Facebook page URL
pageUrl	String	Normalized page URL
pageName	String	Page name or slug
pageId	String	Facebook page identifier
facebookId	String	Facebook ID (usually matches pageId)
title	String	Full page title with context

Classification

Field	Type	Description
categories	Array	Page categories
category	String	Primary category
info	Array	Short info lines from the page
intro	String	Page introduction or description

Contact & Business

Field	Type	Description
phone	String	Public phone number
email	String	Public contact email
messenger	String	Messenger link
address	String	Address or coordinates
addressUrl	String	Google Maps link for the address
website	String	Primary linked website
websites	Array	All detected website links
priceRange	String	Price range (e.g. "££")

Engagement

Field	Type	Description
followers	Number	Follower count
likes	Number	Like count
followings	Number	Following count
checkins	Number	Check-in count ("were here")
talkingAboutCount	Number	Talking-about count

Media

Field	Type	Description
profilePictureUrl	String	Profile picture URL
coverPhotoUrl	String	Cover photo URL

Business Details

Field	Type	Description
business_hours	String	Business hours info
ratingOverall	Number	Overall rating score, synthesized to a 0-5 scale for compatibility (Facebook Pages actually use a recommend %, not 1-5 stars — see recommendPercentage)
ratingCount	Number	Review/rating count
ratingText	String	Human-readable rating text
recommendPercentage	Number	Raw recommend percentage (0-100) as shown on the page — the authentic Facebook Pages rating metric
ownerOrganization	String	Confirmed owning organization/business ("X is responsible for this Page"), when Facebook discloses it
pageAdLibrary	Object or string	Ad library metadata when found on the page, otherwise a constructed Ad Library search URL for the page
reviews	Array	Up to 10 reviews that Facebook shows first on the page's reviews tab: [{reviewerName, recommends, reviewText, reviewDate}] (reviewDate as YYYY-MM-DD). Inside a review, reviewText or reviewDate can be null when Facebook didn't include them. Controlled by the includeReviews input (default on)

A top-level field that Facebook didn't expose for a page is left out of that record entirely rather than included as null (the review objects described above are the one exception).

Sample Output

{
  "facebookUrl": "https://www.facebook.com/ozonecoffeeuk",
  "pageUrl": "https://www.facebook.com/ozonecoffeeuk",
  "pageName": "ozonecoffeeuk",
  "pageId": "191234567",
  "facebookId": "191234567",
  "title": "Ozone Coffee Roasters | Facebook",
  "categories": ["Coffee Shop"],
  "category": "Coffee Shop",
  "intro": "Specialty coffee roaster and cafe in London.",
  "messenger": "https://m.me/191234567",
  "followers": 9100,
  "followings": 12,
  "profilePictureUrl": "https://scontent.xx.fbcdn.net/v/...",
  "coverPhotoUrl": "https://scontent.xx.fbcdn.net/v/...",
  "pageAdLibrary": "https://www.facebook.com/ads/library/?active_status=all&ad_type=all&country=ALL&view_all_page_id=191234567&search_type=page",
  "reviews": [
    {
      "reviewerName": "Jane Smith",
      "recommends": true,
      "reviewText": "Great coffee and friendly staff!",
      "reviewDate": "2025-11-02"
    }
  ]
}

Unavailable Pages

When a discovered or provided URL turns out to be unavailable or deleted, it is detected and skipped entirely — it does not appear in the output dataset and does not consume your maxItems budget. This keeps the dataset free of junk rows; check the run log if you need to know which URLs were skipped and why.

Use Cases

• Local business discovery: find businesses in specific cities or regions
• Competitive intelligence: build datasets of competitor pages in your industry
• Lead generation: collect pages with contact info for outreach
• Market research: survey the Facebook page landscape for a category
• Monitoring: run repeatedly to track new pages for a search term

FAQ

Do I need a Facebook account to use this?

No, but it's strongly recommended. Without cookies, page discovery relies on other search engines and extracts only public HTML. Providing session cookies from a logged-in Facebook account enables the most reliable discovery method (Facebook's own search) and gives access to contact fields that aren't present in the public HTML.

Why are some fields missing from a record?

A missing field means Facebook did not expose that information in the page's public HTML, the page owner did not fill it in, or the field layout for that page type doesn't include it — it's left out of the record rather than included as an empty value. Fields like phone, address, and website are only populated when the page owner has listed them on their Facebook page. Adding cookies increases the chance of extracting these for login-gated pages.

Why did I get fewer results than maxItems?

The most common reasons:

• Login walls: Facebook gates some pages behind a login session, and pages that come back as a login wall are skipped. Provide session cookies to reach more of them.
• No cookies provided: without session cookies, discovery relies solely on DuckDuckGo/Google, which are less reliable than Facebook's own search — add cookies for the best results.
• Narrow search terms: each (term, location) pair yields a limited pool. Add more search terms or locations.

Can I get 1,000+ results?

Yes — see Scaling to 1K+ Results above. The short version: combine multiple search terms × multiple locations and set maxItems to your target. The actor handles all pagination and over-discovers URLs automatically.

How often do I need to refresh my cookies?

Whenever they stop working: logging out of that browser session or Facebook expiring the session invalidates them. If you notice a sudden drop in results or many login-wall skips in the logs, export fresh cookies from your browser.

Does this work for Facebook Groups or personal profiles?

No. The scraper targets public Facebook Pages (business pages, community pages, public figures). Facebook Groups and personal profiles have different URL structures and content restrictions.

Is this affiliated with Facebook or Meta?

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