Facebook Groups Search Scraper (members, privacy, no login)

Search Facebook groups by keyword and get one flat row per group: member count, privacy, description, group id and the URL — no Facebook account, no cookies, no session.

What you get

Real rows from the example dataset (queries: ["yoga kyiv"], plus one group added via Group URLs):

name	memberCount	privacy	description
Yoga in Ukraine in English - Kyiv, Dnipropetrovsk...	378	public	"This group is intended for all activities, information and studies on Yoga in Ukraine…"
yoga.kyiv	—	private	— (login wall: name and URL only, no numbers)
Python (added via Group URLs)	149	public	"What is Python…"

Full row: id/slug, url, groupId, name, memberCount, memberCountText, privacy, visibility, description, website, email, detailsFetched, query, rank, source, scrapedAt.

Use cases

• Community managers — shortlist public groups to post announcements or promotions in, filtered to a member-size range so tiny dead groups and 500k mega-groups don't waste your time.
• Market researchers — size a niche by keyword or city: how many public groups exist, how big they are, what they say about themselves in their own description.
• Agencies — build an outreach list of public groups (group names and admin-published contacts, not individual people) for a client's niche, in minutes instead of manual searching.

Try it in 10 seconds

Hit Start/Try it — the input already works: queries: ["yoga kyiv"], resultsPerQuery: 10, maxItems: 20, nothing required. Swap in your own keywords and run again.

To track a niche over time instead of a one-off pull: save the task, set Mode to monitor, and put it on a schedule (Apify → Schedules → cron 0 8 * * 1 for weekly). Each run then returns only groups that are new or whose member count/privacy changed since last time, and can post the diff to a webhook or Telegram.

Guide: How to find Facebook groups by keyword without login walks through a real run, the monitor mode and the FAQ.

Related actors

• Facebook Pages Search: Business Contacts, Followers, Monitor — need the business's public Page instead of its Groups: contacts, followers, monitor mode.
• Telegram Channel Search Scraper (topic, subscribers, no login) — the same keyword-and-size discovery, but for Telegram channels.
• Reddit Scraper (subreddit posts, search, comments, no login) — communities built around subreddits instead of Facebook groups.
• Facebook Reels Scraper (page reels, views, MP4, no login) — once you have the Page, pull its video content and view counts.
• Threads Scraper — Posts, Profiles, Replies & Search, No Login — search Threads instead of Facebook for the same topic.

How it works

1. Discovery — each keyword is searched in a public search-engine index restricted to facebook.com/groups, so you get groups Facebook itself has published to the open web. Post links collapse into their group.
2. Enrichment (fetchDetails) — each group page is opened as a logged-out visitor to read memberCount, privacy, description and groupId. Private groups show a login wall and come back as privacy: private with no numbers.
3. Contacts — website and email are the first link and the first email address the group published in its own description. No extra requests, no personal data: these are group-level, self-published contacts.

Input

Field	Meaning	Default
queries	Keywords to search groups for, e.g. yoga kyiv	["yoga kyiv"]
groupUrls	Known group URLs to enrich directly	empty
resultsPerQuery	Search hits scanned per keyword (1-200)	10
fetchDetails	Open each group page for members/privacy/description	true
minMembers / maxMembers	Keep only groups inside that size range	empty (no filter)
maxItems	Stop after this many group rows	20
mode	scrape or monitor (only new/changed since last run)	scrape
monitorStateId, webhookUrl, telegramBotToken, telegramChatId	Monitor-mode state key and alert targets	empty

Pricing

Event	Price
result	$0.003 per group ($3 per 1,000)
monitor-check	$0.006 per monitor run
change	$0.003 per new/changed group

Charged only for groups actually pushed. Residential proxy traffic is billed by Apify on top (group pages hit a login wall from datacenter IPs).

Found it useful? A short review on the Store page helps other people find this actor and tells us what to improve. If something is wrong, open an issue on the actor page — issues are answered within a day.

Why this actor

• No login and no cookies to supply — nothing to keep alive, nothing to get banned.
• Member-size filter built in, so you can ask for "groups between 1,000 and 50,000 members" instead of filtering afterwards.
• Monitor mode with webhook and Telegram alerts on member-count and privacy changes.
• A run that finds nothing pushes nothing, so it charges no result events; the run summary says why it was empty instead of leaving you guessing.

Limits

• Facebook's own group search needs a login and is not used; discovery depends on the search-engine index, which favours groups with public activity. Very small or freshly created groups may be missing.
• Private groups: name only, no member count.
• No posts, members or admin lists — only group-level facts.

FAQ

Does it need a Facebook login or cookies? No. There is no account field and no cookie input. Everything comes from pages Facebook serves to logged-out visitors, which is also why private groups return a name and nothing else.

What happens when there are no results? The dataset is empty and no result events are charged. The RUN_SUMMARY record in the run's key-value store carries emptyReason, so you can tell "the keyword really has no public groups" from "the search index answered with a challenge".

Can I get the members or the posts of a group? No. This actor returns group-level facts only — no member lists, no posts, no comments, no personal data about individuals.

Changelog

• 0.1: initial release — keyword discovery, group enrichment, monitor mode.
• 0.2: member-size filter, website/email from description; empty search results are retried and reported as a block instead of an empty dataset.
• Yahoo discovery now reads every organic result block; runs return more hits per query.