Facebook Events Scraper

Facebook Events Scraper

Type what you are looking for — "concert", "startup meetup", "food festival" — add a city if you want, or paste Facebook event, Page or Group URLs, and get public Facebook events back as clean, flat rows. Every row is the full event, not a search snippet: the whole description, the exact start and end time with its timezone, the venue with its street address and coordinates, the organising Page, the ticket link, the cover image and how many people are going or interested. No login, no cookies, no Facebook account, no browser, and nothing for you to configure.

• Search by keyword, by city, by Facebook Page, by Group, or straight from an event URL — mix them in one run.
• Full event detail on every row, including the complete description and the venue's street address and latitude/longitude.
• Going and interested counts, and the total number of people who have responded, wherever the host makes them public.
• Date windows, categories, online-only and ticketed-only are filters on the input, not paid add-ons — events they drop are never charged.
• A city name is matched against Facebook's own location list, so "Berlin" returns events in Berlin rather than events with "Berlin" in the title.
• Runs with empty input return a labelled sample row, free, so you can see the shape first.
• A private, deleted or misspelled event produces an uncharged diagnostic row, never a failed run.

Price

$0.90 per 1,000 events, plus a $0.001 start fee per run (billed per gigabyte of run memory, so exactly that on the default 1 GB).

This is the cheapest Facebook events scraper on the market, and it is the same rate on every plan, free or paid. There are no volume tiers, no minimum spend, no subscription and no add-on fees. What you read here is what you pay on day one and on day four hundred.

What is actually charged

• One event-scraped event per event row written to the dataset. Nothing else is metered per row.
• Free: the sample row an empty run returns, and every diagnostic row — a blocked target, a dead URL, a search that matched nothing. Those rows all carry "charged": false.
• Events dropped by your date window, category, online-only or ticketed-only filters are never charged — the filters run before billing.
• Duplicate events found through more than one search term or URL are dropped before they are charged.
• Looking a city name up against Facebook's location list costs nothing.
• A search that matches nothing produces one free note row and no event charges.
• A run that finds nothing costs the start fee and nothing more.
• Rows never leave the dataset without a charge, and are never charged without a row. The billed event is a named one, so there is no price quietly attached to apify-default-dataset-item — the trick that makes some scrapers bill you for their own error messages.

Input

{
  "searchTerms": [
    "startup meetup",
    "tech conference"
  ],
  "location": "Berlin",
  "maxEvents": 20,
  "startingFrom": "2026-09-01",
  "startingBefore": "2026-12-31"
}

Run it with empty input and you get one clearly labelled sample row, free, so you can see the output shape before you spend anything.

Output

One row per event. A real row from a real run:

{
  "ok": true,
  "charged": true,
  "recordType": "event",
  "eventId": "1045755418043207",
  "eventUrl": "https://www.facebook.com/events/1045755418043207/",
  "name": "Devassas Concert",
  "description": "Rock concerts with Devassas (portuguese band - punk, grunge, metal) and Die Ausreden (powerpunk) at Regenbogenfabrik, in Berlin!\nSee ya!",
  "startTime": "2026-08-21T17:00:00.000Z",
  "startTimestamp": 1787331600,
  "endTime": null,
  "endTimestamp": null,
  "timezone": "UTC+02",
  "dateText": "Friday, August 21, 2026 at 7:00 PM CEST",
  "venueName": "Regenbogenfabrik",
  "venueAddress": "Lausitzer Straße 21a, 10999 Berlin, Germany",
  "venueCity": "Berlin, Germany",
  "venueUrl": "https://facebook.com/Regenbogenfabrik",
  "latitude": 52.495993170936,
  "longitude": 13.426580093814,
  "isOnline": false,
  "onlineUrl": null,
  "organizers": [
    {
      "id": "61580124287250",
      "name": "Devassas Band",
      "url": "https://www.facebook.com/people/Devassas-Band/61580124287250/",
      "type": "User"
    }
  ],
  "organizerName": "Devassas Band",
  "organizerUrl": "https://www.facebook.com/people/Devassas-Band/61580124287250/",
  "ticketUrl": null,
  "priceRange": null,
  "attendingCount": 4,
  "interestedCount": 2,
  "respondedCount": 6,
  "coverImageUrl": "https://scontent.xx.fbcdn.net/v/t39.30808-6/772604334_122136197643004142_4062119575631978264_n.jpg",
  "categories": [],
  "isCanceled": false,
  "isPast": false,
  "sourceType": "search",
  "sourceQuery": "concert",
  "sourceUrl": "https://www.facebook.com/events/search/?q=concert",
  "scrapedAt": "2026-08-16T13:39:24.434Z"
}

Field notes

• eventId — Facebook's own numeric id for the event — stable, and the thing to key on when you re-run and merge.
• startTime — the exact start in ISO 8601 UTC, read from the event itself rather than from a "Sat, 21 Aug" label. startTimestamp is the same moment in epoch seconds and dateText is the sentence Facebook prints.
• endTime — null when the host never set an end time, which is common. It is not an error and it is not guessed.
• timezone — the event's own timezone as Facebook stores it, so you can put a local start time back together from startTimestamp.
• venueAddress — the one-line address Facebook shows on the event. For an event whose "venue" is a free-text line rather than a Page, this and venueName can be the same string — that is what the host typed.
• latitude / longitude — the venue coordinates. Null for online events and for the small number of events whose host typed a place name Facebook could not resolve.
• organizers — every host Facebook lists, each with name, Page or profile URL, id and type. organizerName and organizerUrl repeat the first one so a spreadsheet has a plain column to sort on.
• ticketUrl — the external ticketing link the host attached. Null when tickets are not sold through a link, and priceRange carries the price band when Facebook shows one.
• attendingCount / interestedCount — the going and interested numbers as Facebook publishes them, which are rounded on large events ("1.2K interested" becomes 1200). respondedCount is the exact total where Facebook gives one. All three are null when the host hides the guest list.
• sourceType / sourceQuery / sourceUrl — how this event was found — search, city, page, group or url — so rows from a multi-term run can be grouped afterwards.

Every real row carries "charged": true. Sample rows carry "_sample": true and diagnostic rows carry "_diagnostic": true with an errorCode you can filter on, and neither is ever billed.

How it works

• Facebook renders its public event pages, its public events search and a Page's hosted-events tab server-side, and this Actor reads those renders directly. No account is created, no cookies are planted and no login is ever attempted.
• The data is taken from the structured payload Facebook ships inside the page, not scraped out of on-screen text, so a start time is a real timestamp and a venue is a real place object with coordinates.
• Traffic leaves through a large pool of rotating addresses that cost nothing per gigabyte. If Facebook answers one address with a stub, the run takes a different address and asks again rather than giving up or charging you.
• Each page read stops as soon as the event data has arrived instead of downloading the rest of the render, which is what keeps the price where it is.
• You can point it at your own proxy servers instead, and nothing else changes.

What people use it for

• Event aggregators and "what's on" sites filling a city listing without hand-entering anything.
• Venue and promoter research — who is putting on what, where, and how many people said they are coming.
• Sponsorship and sales teams building a pipeline of upcoming conferences, meetups and festivals in a territory.
• Local media and tourism boards monitoring what is happening in a city over a date window.
• Anyone tracking their own or a partner Page's event calendar into a spreadsheet or a database.

Limitations

• Facebook's logged-out surfaces are shallow by design: one search term returns up to 8 events, one city page up to 12, and a Page or Group about 8. Going deeper needs a logged-in session, which this Actor will not use. Runs are therefore built to go WIDE — many terms, many cities, many Pages in one run — rather than deep on one term.
• Private events, events restricted to a Group's members, and events behind an age or country gate are not public. They come back as an uncharged diagnostic row rather than a partial one.
• The guest list itself is never public. attendingCount, interestedCount and respondedCount come from the counters Facebook prints on the event, they are rounded on large events, and they are null when the host hides them.
• endTime and ticketUrl are null whenever the host did not set them. That is most casual events, and no value is invented to fill the column.
• A city name is matched against Facebook's own location list. Where Facebook has no matching location the name is folded into the search text instead — you still get results near that place, but matched on words rather than geography, and a free note row says so. Paste a Facebook events URL for the city to force an exact area.
• Facebook occasionally answers a stub instead of a page. The run takes a fresh address and asks again, and only after several attempts does it write an uncharged BLOCKED row. Re-running a minute later normally clears it.
• Search results are ranked by Facebook, not by date. Use the date window if you need a specific period rather than expecting the newest events first.
• Facebook's logged-out events index does not answer every phrase. Screened over 87 ordinary event phrases, 85 returned results and a handful ("wine tasting", "pub quiz", "beer festival") returned nothing at all, from any address, with or without a city — while single words from the same phrase ("tasting") returned plenty. When a term comes back empty you get a free NO_RESULTS row; try a shorter or more common wording.
• Recurring events are returned as the occurrence Facebook links to. The series is not expanded into one row per date.

Questions

Do I need a Facebook account, cookies or a token?

No. Nothing is logged in and nothing is stored. Only public events are read, exactly as an anonymous visitor sees them.

One of my search terms returns nothing at all. Is it broken?

Probably not. Facebook's public events index answers most phrases but genuinely returns nothing for some of them, from every address and with or without a city. A single word from the same phrase usually works where the two-word version does not. An empty term costs you nothing — you get a free note row.

Why do I only get 8 events for a search term?

That is what Facebook's public events search hands an anonymous visitor for one term. Add more terms, add a city, or point the run at specific Pages — the run limit is on total events, not on sources, so going wide costs nothing extra per source.

Can I get events near a specific city rather than everywhere?

Yes. Put the city in location. It is matched against Facebook's own location list, so results are geographic rather than keyword matches. For complete control, paste a Facebook events URL for that city into location instead.

What happens if an event is private or has been deleted?

You get one uncharged row with "_diagnostic": true and a NOT_FOUND code saying so. You are never charged for it and the run still finishes successfully.

Does a failed or empty run cost me anything?

Only the start fee. Diagnostic rows, sample rows and events dropped by your filters are all free.

Can I pull a Page's past events too?

Yes — set includePastEvents to true and pass the Page URL in startUrls. A Page's public events tab mixes upcoming and past events; with the setting off, anything that has already started is dropped before it is charged.