The Mine Works
Public Facebook Data Without the Graph API: Which Actor for Which Job
← All posts
comparison November 20, 2026 · 8 min read Updated November 20, 2026

Public Facebook Data Without the Graph API: Which Actor for Which Job

Five actors read public Facebook as a logged out visitor: keyword post search, Page timelines, events, group discovery and Page contacts. Here is which one fits which job, with inputs, prices and limits.

Try the scraper

The actor referenced in this article. Pay only for results delivered.

View the scraper →

You want public Facebook data in a spreadsheet: what people are saying about a brand, what a competitor’s Page posted last quarter, which events are coming up in a city, which groups are worth joining, or who to call at a list of local businesses. The usual advice is Meta’s Graph API, which starts with a developer app and an access token before you see a single post. Most people who ask for “Facebook data” just want the public posts, events, groups and Page details any visitor can already see.

Try it live: Facebook Post Search Scraper. Pay per result delivered.

We run five Facebook actors on Apify. None of them logs in, asks for cookies or needs a token. Each one does a single job and returns one flat row per item, so the hard part is picking the right one. This guide covers which actor fits which job, a working input for each, what the rows hold, the exact prices and the limits you will hit.

Which actor for which job

Your jobActorWhat you give itOne row is
Find posts about a topic or brandFacebook Post Search ScraperKeywords or hashtagsA public post
Read everything a Page postedFacebook Posts ScraperPage links, names or idsA post on that Page
List events in a city, Page or groupFacebook Events ScraperKeywords, cities or linksAn event
Discover groups on a topicFacebook Groups Search ScraperKeywords, or group linksA group
Get phone, email and website for PagesFacebook Page Contacts ScraperPage links, names or idsA business Page

Most users start with keyword search, because it answers “who is talking about this?” before you know which Pages to watch. The other four take over once you have names.

Search public posts by keyword

Facebook Post Search Scraper searches Facebook’s public post search for words or hashtags and returns posts from Pages, public profiles and public groups. This input watches a topic in recent order and keeps only posts that contain the exact phrase:

{
  "keywords": ["electric cars", "#electricvehicles"],
  "sortBy": "latest",
  "onlyPostsContainingKeyword": true,
  "maxPostsPerKeyword": 100
}

Each row has post_id, post_url, text, created_at, author_name, author_id, post_type, reactions_count with a reactions split by type, comments_count, shares_count, media links, and group_name when the post sits in a public group. The field that matters most for brand monitoring is keyword_in_text. Facebook’s search also returns related posts that do not contain your words; in testing about two thirds of posts held the exact keyword. onlyPostsContainingKeyword drops the rest for free.

Two more settings are worth knowing. startDate and endDate take YYYY-MM-DD, and posts Facebook returns outside the range are dropped without charge. location takes a city name, but Facebook treats it as a preference rather than a strict filter, so check place_name if you need posts tagged in one place.

Read a Page’s own timeline

When you already know the Pages, Facebook Posts Scraper reads each one’s timeline newest first, up to 5,000 posts per Page. This weekly competitor check returns only the last seven days of posts:

{
  "startUrls": [
    { "url": "https://www.facebook.com/Starbucks" },
    { "url": "https://www.facebook.com/nike" }
  ],
  "resultsLimit": 200,
  "onlyPostsNewerThan": "7 days"
}

The rows carry the same engagement columns as search, plus video_views on videos and reels, media with photo and video links, link card fields and shared_post for reshares. onlyPostsOlderThan asks Facebook to start the Page at a past date, so you can pull one month from last year without paying for everything posted since. Page names and numeric Page ids go in usernames instead of startUrls.

List events by city, Page or group

Facebook Events Scraper takes keywords, cities, or links to Pages, groups and single events, and mixes them in one run. This input reads each city’s whole event feed for the coming week:

{
  "locations": ["Austin, Texas", "Dallas, Texas"],
  "when": "next_week",
  "maxEventsPerSearch": 200
}

An event row holds name, description, start_at, end_at, venue_name, address, latitude, longitude, organiser_name, organiser_url, every host in hosts, going_count, interested_count, ticket_url and categories. With includeDetails on (the default) the counts are exact integers; with it off the run is about five times faster but large counts may be Facebook’s rounded figures, flagged by counts_rounded.

Discover groups before you join them

Facebook’s own group search needs a login. Facebook Groups Search Scraper finds groups through site:facebook.com/groups searches on DuckDuckGo and Brave, then reads each group’s public About data from Facebook itself. This input keeps only large public groups:

{
  "searchQueries": ["real estate investors"],
  "privacy": "public",
  "minMembers": 10000,
  "maxGroupsPerKeyword": 50
}

A group row has group_id, name, url, privacy, the exact members count, posts_today, posts_last_month, posts_per_day, created_at, locations, rules and admins_and_moderators. Sorting by posts_per_day separates live groups from large dead ones. Private groups still return these numbers, because Facebook shows a private group’s About data to anyone; only their posts stay hidden.

Turn Pages into contact rows

Facebook Page Contacts Scraper reads the header and Intro box of public business Pages and returns one row per Page:

{
  "pages": [
    "https://www.facebook.com/copperkettleyqr",
    "https://www.facebook.com/PatoisToronto/",
    "kissatanto",
    "barmarcopgh",
    "TheBiltmoreMayfair"
  ]
}

Rows hold name, category, phone, email, website, address, city, country, Instagram, X, YouTube, TikTok, LinkedIn and WhatsApp links, followers, rating_text, verified and ad_library_url, which opens the Page’s ads in Meta’s Ad Library. In a test of 193 business Pages from 44 countries, 89% had a website, 70% a phone and 67% an email. Pages that are missing, personal profiles and duplicates go to a separate free dataset, facebook-page-status, with a reason for each.

Chain two of them in Python

The actors are built to feed each other. A common one: find the people who run events in your niche, then get their contact details. organiser_url from the events actor goes straight into pages on the contacts actor, and the contacts row keeps your entry in input, so the two join cleanly.

from apify_client import ApifyClient

client = ApifyClient("YOUR_APIFY_TOKEN")

events_run = client.actor("themineworks/facebook-events-scraper").call(run_input={
    "keywords": ["wedding expo", "bridal show"],
    "locations": ["Chicago"],
    "maxEventsPerSearch": 100,
})

organisers = {}
for row in client.dataset(events_run["defaultDatasetId"]).iterate_items():
    if "_type" in row:
        continue  # summary, info and error rows: free, no event in them
    url = row.get("organiser_url")
    if not url:
        continue
    seen = organisers.get(url)
    if seen is None or row["going_count"] > seen["going_count"]:
        organisers[url] = {"event": row["name"], "going_count": row["going_count"]}

if organisers:
    contacts_run = client.actor("themineworks/facebook-page-contacts-scraper").call(run_input={
        "pages": list(organisers),
    })
    for page in client.dataset(contacts_run["defaultDatasetId"]).iterate_items():
        if "_type" in page:
            continue
        top = organisers.get(page["input"], {})
        print(page["name"], page.get("phone"), page.get("email"),
              page.get("website"), top.get("event"), top.get("going_count"))

An organiser that turns out to be a personal profile is skipped by the contacts actor with a free personal_profile status, so you never pay for it.

What each actor costs

All five charge per row delivered plus a small flat fee per run. Gold, Platinum and Diamond plans pay the Gold price.

ActorFree planGold and aboveStart fee
Facebook Post Search Scraper$3.60 per 1,000 posts$1.80 per 1,000 posts$0.001, only after the first post
Facebook Posts Scraper$4.50 per 1,000 posts$1.80 per 1,000 posts$0.0009, only after the first post
Facebook Events Scraper$11.70 per 1,000 events$6.30 per 1,000 events$0.001, only after the first event
Facebook Groups Search Scraper$1.71 per 1,000 groups$1.25 per 1,000 groups$0.003, only after the first group
Facebook Page Contacts Scraper$11.70 per 1,000 Pages$5.94 per 1,000 Pages$0.0009, on every run

Duplicates, rows dropped by your filters, and the summary, info and error rows are never charged on any of them. Full event details and group About data cost nothing extra.

Limits to plan around

  • You see what a logged out visitor sees. No private group posts, no friends only content, no personal profiles. Facebook may also personalise results for logged in accounts, so counts can differ from what you see in your own app.
  • Keyword search depth is set by Facebook. Across 12 test keywords a search list ended between 71 and 303 posts in top order, and latest order is shorter. Add related keywords to collect more; a post found twice is charged once.
  • Timelines are slow by design. Facebook hands out a Page’s timeline three posts at a time, about one post every two seconds per Page. For more than about 1,500 posts from one Page, raise the run timeout above the default hour.
  • Events default to the US. With no city set, Facebook picks events near the proxy, which is in the US. Add locations for anywhere else.
  • Group discovery covers what the web can see. Popular keywords reached 100 groups in every test, but that is not every group Facebook holds, and hidden groups cannot be found from outside.
  • Contact fields depend on the business. Big brands often list only a website. Restaurants, hotels, bars and local services fill phone and email far more often.
Related Actor

Explore the scraper referenced in this article: inputs, outputs, and pricing, then run it on Apify.

Frequently asked questions

Do any of these actors need a Facebook account, cookies or a token? +

No. All five read Facebook the way a visitor who is not logged in sees it, over Apify Proxy, which is included in the price. You need an Apify account and nothing else. The flip side is that private groups' posts, friends only content and personal profiles are out of reach.

Which actor should I start with? +

If you do not yet know which Pages matter, start with Facebook Post Search Scraper: type a keyword and see who is posting about it. Once you have a list of Pages, Facebook Posts Scraper reads their full timelines and Facebook Page Contacts Scraper turns them into phone, email and website rows.

Is a run that finds nothing charged? +

For post search, Page posts, events and groups, no: the start fee is charged only after the first row reaches your dataset. Facebook Page Contacts Scraper is the exception. Its $0.0009 start fee is charged when every run starts, although Pages that cannot be read are never charged.

Can I get the comments on the posts these actors find? +

Not from these five. Post search and Page posts give comments_count, not the comment text. Pass the post_url values into Facebook Comments Scraper to get the comments and replies themselves.

Apify Store

Find a ready-made scraper for your job

Apify has over 80,000 scrapers and automations, ours included. Start free with $5 of platform credit every month.