Facebook Search Scraper

Search Facebook across every tab — people, pages, events, posts and videos — and get clean structured data. No login, no cookies, no browser. Enter a keyword, pick a search type, and the actor queries Facebook's own search and returns names, profile/page URLs, profile pictures, verification badges, page descriptions, the business's own website, and post text with author IDs.

Pure HTTP. No account, no cookies, no headless browser. Results come straight from Facebook's own search — not scraped from Google.

Why use this scraper

• No login and no cookies. You don't provide a Facebook account or session. The actor bootstraps a guest token and calls Facebook's search directly.
• Six search types, one actor. People, pages, events, posts, videos, or "top" (all) — selected with a single searchType field.
• Native Facebook results. Straight from Facebook's ranking, not a Google site: workaround.
• City filtering by name. Type "Manchester, NH" or "London, UK" and the search is scoped to that city — Facebook's own City filter, on people, pages, places and events. The actor resolves the name to Facebook's place UID itself; no separate lookup step, no ID database to maintain.
• Auto-pagination. Set maxItems and the actor walks the result cursor across pages for you.
• Business websites. Pages and places return the site the business links to, already unwrapped from Facebook's l.php redirect — feed it straight into enrichment.
• Rich fields. Names, URLs, pictures, verified badges, page descriptions, the profile's "Lives in" line, and for posts and videos the body text, author name + numeric ID + profile URL, permalink and timestamp.

How it works

1. You give a keyword (or several) and a search type.
2. We fetch — bootstrap a guest token, call Facebook's search API (one query serves every tab), and page through the results via the cursor. FB traffic egresses through residential proxies.
3. You get structured rows (JSON, CSV, or Excel), each tagged with a rowType.

Supported search types

Use cases

• Lead generation — find people and pages matching a niche, brand, or location.
• Brand & competitor monitoring — track public posts and pages mentioning a keyword.
• Event discovery — pull events matching a topic or city.
• Audience & influencer research — surface profiles and pages around a theme.
• Social listening — collect public post text + authors for a keyword over time.

Input configuration

Example input

{
  "searchType": "people",
  "searchQueries": ["coffee roaster", "specialty coffee"],
  "maxItems": 200
}

Searching one city

location applies Facebook's own City filter, so places searches return businesses in that city and people searches return profiles who list it as their location:

{
  "searchType": "pages",
  "searchQueries": ["tree service"],
  "location": "Manchester, New Hampshire",
  "maxItems": 50
}

At run start the actor asks Facebook's place lookup for the name, picks the city page, and logs what it chose along with the runners-up:

[LOCATION] "Manchester, New Hampshire" → Manchester, New Hampshire (uid 107634079257154)
[LOCATION] other candidates: Downtown Manchester (7,436 people checked in here) = 1173291086177843; …

Name the region. Facebook's lookup is a string matcher, so

"Manchester, New Hampshire"

Already have the UID? Put it in locationUid and the lookup is skipped. That field accepts the bare number, a facebook.com/places/<uid>/ URL, or a Facebook search URL with the City filter applied, and it takes precedence over location. A few: London 106078429431815, Paris 110774245616525, Portland OR 112548152092705.

Two things to know. If Facebook has no place for the name, the run ends immediately with a status message saying so — nothing is charged. And Facebook has no city filter for posts, videos or top; set a location there and the actor runs unfiltered and logs a warning instead of failing.

Output

Null fields are omitted from the samples below; every row carries the full key set, so CSV and Excel exports line up.

rowType: "page" / "place"

{
  "rowType": "page",
  "id": "100064803603692",
  "name": "Coffee Island",
  "url": "https://www.facebook.com/CoffeeIslandCyprus",
  "profilePicture": "https://scontent.../...jpg",
  "isVerified": true,
  "description": "Το ταξίδι στη χώρα του καφέ… Welcome to Coffee Island!",
  "website": "http://www.coffeeisland.com.cy/",
  "query": "coffee island",
  "searchType": "pages"
}

rowType: "person"

{
  "rowType": "person",
  "id": "100044619540852",
  "name": "John Smith",
  "url": "https://www.facebook.com/thejohnsmith",
  "profilePicture": "https://scontent.../...jpg",
  "isVerified": true,
  "subtitle": "Lives in San Antonio, Texas",
  "searchType": "people"
}

rowType: "event"

{
  "rowType": "event",
  "name": "Cavendish Beach Music Festival 2026",
  "url": "https://www.facebook.com/events/1108414837875098/"
}

rowType: "post"

{
  "rowType": "post",
  "id": "1023706227132483",
  "authorName": "White Oak Mtn Coffee",
  "authorId": "100084793454538",
  "authorUrl": "https://www.facebook.com/whiteoakmtncoffee",
  "text": "Coffee connects people from all over the world. 🌎☕️…",
  "postUrl": "https://www.facebook.com/whiteoakmtncoffee/posts/pfbid02EY…",
  "creationTime": "2026-08-17T15:25:28.000Z"
}

rowType: "video"

{
  "rowType": "video",
  "id": "122127066963218787",
  "authorName": "The Home Coffee",
  "authorId": "61586563628977",
  "text": "🥛🍫 Nutella Iced Coffee ☕️🧊…",
  "url": "https://www.facebook.com/reel/1004992302321918/",
  "profilePicture": "https://scontent.../thumbnail.jpg",
  "creationTime": "2026-07-01T10:14:39.000Z"
}

Key output fields

FAQ

Do I need a Facebook account or cookies? No. The actor works fully logged-out with a guest token — nothing to provide.

Is this Facebook's real search? Yes — it calls Facebook's own search endpoint, so results and ranking match what you'd see on the site (not a Google site:facebook.com workaround).

Can I search a date range? Yes, on the posts tab. Set onlyPostsNewerThan and/or onlyPostsOlderThan. Facebook's own date filter is month-granular at the edges and lets a few out-of-window posts slip through — verified live: a 2019 window returned three posts dated into January 2020. So the filter is used to narrow the crawl, and results are then trimmed to your exact dates. Anything outside the window is never pushed and never charged, and the run log reports how many were trimmed. Date windows do not apply to the other tabs; ask for one there and the run says so rather than pretending to filter.

Can I search a specific city? Yes — set location to a city name ("Manchester, New Hampshire", "Manchester, NH", "London, UK") and people, pages, places and events searches are scoped to that city. The actor resolves the name to Facebook's place UID and logs the match plus the runners-up. If you already know the UID, locationUid skips the lookup. Facebook offers no city filter for posts, videos or top.

How do I search many cities? One run per city, each with its own location — for example from a loop or a scheduled task list. Every row is stamped with locationUid and locationName, so results from different runs merge cleanly.

Do I get the business's website? Yes, when the page publishes one — website holds the real destination, already unwrapped from Facebook's l.php redirect, so it drops straight into an enrichment step.

Do posts come with like and comment counts? No. Facebook withholds engagement counts from logged-out requests, and this actor never logs in. You get the body, author (with numeric ID), permalink and timestamp.

Which proxy should I use? Residential. Facebook rate-limits datacenter and cloud IPs, so residential proxies give reliable throughput.

How many results can I get? It auto-paginates until maxItems or the results run out. Each keyword returns its own paginated set.

Why isn't group search supported? Facebook gates group search behind a login, which this no-cookie actor deliberately avoids. People, pages, events, posts and videos are all available logged-out.

Notes & limitations

• Only public search results are returned (logged-out visibility).
• Group search is not available without a login and is intentionally omitted.
• City filtering (location / locationUid) covers people, pages, places and events; Facebook exposes no city filter for posts, videos or top.
• City names are resolved with Facebook's own place lookup, which is a plain string match: add the state or country for common names, and spell out anything that is not a US/Canadian region code ("Manchester, NH" works; "Madrid, ES" needs "Madrid, Spain").
• Reaction, comment and share counts are not returned — Facebook does not send them to logged-out requests.
• Event start times are not in Facebook's search response; event rows carry the name, URL and cover image, and the date lives on the event page itself.
• description and website are present only when the page publishes them; subtitle appears on the minority of profiles that show a public location.
• Facebook changes its internals periodically; the actor is maintained to keep pace.

Support

Found a bug or need a search type that isn't exposed? Open an issue on the Actor's Issues tab on Apify, or reach out via the Apify Store profile. Include your input JSON and the run ID.

Explore more scrapers

More Facebook and social tools from the same developer:

• Facebook Marketplace Scraper — newest Marketplace listings, fast
• Facebook Public Group Posts & Comments — no cookies needed
• Facebook Comments Scraper — full comment threads for any public post
• Facebook Ads Library Scraper — ads intel with reach included
• Google Flights Scraper — fares & multi-city from $1/1k

🤖 For AI Agents & LLM Apps

Compact reference for AI agents calling this actor via the Apify MCP server or the Apify API (actor: memo23/facebook-search-scraper).

Purpose: Facebook's own native search across people, pages, places, events, posts and videos — fully logged-out (no account, no cookies) via a bootstrapped guest token, so results match Facebook's ranking rather than a Google site: workaround.

Minimal input:

{
  "searchType": "people",
  "searchQueries": ["coffee roaster"],
  "maxItems": 50
}

Output: one row per result — rowType, id, name, url, profilePicture, isVerified, query, searchType. Pages and places add description and website (the business's own site, already unwrapped from Facebook's l.php redirect); people add subtitle (the "Lives in …" line); posts and videos add text, authorName, authorId, authorUrl, postUrl and creationTime. Every row carries the full key set with nulls where a field does not apply.

Behaviors an agent should know:

• searchType selects the tab: top, people, pages, places, events, posts, videos (default people); one query serves every tab.
• location scopes the search to one city by name ("Manchester, New Hampshire", "Manchester, NH", "London, UK"); the actor resolves it to a Facebook place UID at run start. Include the state or country for common names — a bare "London" picks London, Ontario. If Facebook knows no such place the run ends with a status message and zero rows, nothing charged. Works on people, pages, places, events; on posts, videos and top Facebook has no city filter, so the actor ignores it and logs a warning.
• locationUid is the advanced form: pass the Facebook place UID directly (e.g. 106078429431815 = London) to skip the lookup. It overrides location. A UID that is not a place returns zero rows rather than an error.
• Output rows carry locationUid and locationName when a city was set, so multi-city runs merge cleanly.
• Always set maxItems (total across all queries, default 100) — the actor auto-paginates the result cursor until the cap or the results run out.
• Group search is not supported (login-gated and deliberately omitted); only public, logged-out results are returned.
• Datacenter proxy is the default and works well; switch to RESIDENTIAL only if you hit rate limits at high volume.
• Engagement counts (reactions, comments, shares) and event start times are not available logged-out and are never returned; do not prompt for them.
• Billing: Pay-per-event billing — see the Pricing tab on the actor page.

⚠️ Disclaimer

This actor collects publicly available data from Facebook search for legitimate purposes such as market research, lead generation, and social listening. You are responsible for how you use the data. Comply with Facebook's Terms of Service, applicable data-protection law (including the GDPR when handling personal data), and all local regulations. Do not use collected personal data for unsolicited contact where prohibited. This tool is not affiliated with, endorsed by, or connected to Facebook or Meta.

SEO Keywords

Facebook search scraper, Facebook scraper, Facebook people search, Facebook pages scraper, Facebook posts scraper, Facebook events scraper, Facebook search API, scrape Facebook without login, Facebook data extraction, social media scraper, Facebook profiles scraper, Facebook keyword search, Facebook lead generation, social listening tool, Facebook public data, no-cookie Facebook scraper, Facebook video search, Meta search scraper, Apify Facebook scraper, Facebook search results, Facebook search by city, Facebook search by location name, Facebook location filter, Facebook city UID, Facebook place ID search, Facebook location UID lookup, Facebook local business search, Facebook business website scraper, Facebook page description scraper, Facebook reels scraper.