Business intelligence · Facebook · 2026年9月6日
1,454,867 public Facebook Page contact records — website, email, phone, WhatsApp, category and like count — queryable over one REST API. Filter by contact field, category, like count or discovery source; sort by reach. Read from each Page's public About tab, no login and no Graph API access token, pay on success.
1,454,867
Facebook Pages — one record per Page, refreshed continuously.
66.3%
have a phone
59.3%
have a website
57.3%
have an email
Snapshot 2026年9月6日 — public fields only. A seeded index discovered mostly via business-listing search, not a sample of Facebook.
1,454,867
public Facebook Page contact records, one record per Page, read from the public About tab. 75.4% discovered business-listing search of them — Facebook's own robots.txt blocks Common Crawl on Page paths, so this index cannot lean on the broad web-archive method the rest of this catalog's social datasets use.
66.3%
of Pages publish a public phone number — the richest contact-field coverage of any dataset in this catalog. This reads as a business directory, not a follower graph.
58.6%
of Pages sit under 1,000 likes — small local businesses dominate the index, not mega-brands, though the tail reaches all the way to Netflix and Cristiano Ronaldo.
29.8%
of the index carries a self-declared category, and the tagged slice skews small-business: restaurants, local businesses, coffee shops and hair salons lead the list, not media companies.
Every number on this page describes the Pages in the index, not Facebook's Page base. Discovery works differently here than for this catalog's other social datasets: Facebook's robots.txt blocks Common Crawl on Page paths entirely, so most Pages are found through a business-listing search crawl instead, with a smaller Wikidata sweep (notable people and organizations) and a WARC domain scan filling in the rest. That business-search-heavy method is exactly why this index skews toward small local businesses with rich contact info, rather than toward globally notable brands. It is the right shape for lead generation and business enrichment; it is the wrong shape for claims like "the average Facebook Page".
This index is not a curated list of major brands: the modal band is under 1,000 likes, and only 41.4% clear that line. 2,413 Pages sit above a million — up to 205M on Netflix.
| Likes | Pages | Share |
|---|---|---|
| Under 1K | 852,568 | 58.6% |
| 1K–9.9K | 493,088 | 33.9% |
| 10K–99K | 94,207 | 6.5% |
| 100K–999K | 12,591 | 0.9% |
| 1M+ | 2,413 | 0.2% |
A Page record is only useful if the contact fields you need are populated. Two-thirds carry a public phone number and just over half publish a website or email — well above the field-coverage this catalog's follower-focused social datasets carry, since these records come from a business's own About tab rather than a personal profile.
Facebook lets Pages self-declare a category. It is a bonus field, not a universal one — only 29.8% of the index carries a value — but for the Pages that do, it is a clean way to slice by business type.
| Category | Pages | Share of tagged |
|---|---|---|
| Restaurant | 35,572 | 8.2% |
| Local business | 31,012 | 7.2% |
| Product/service | 29,577 | 6.8% |
| Coffee shop | 14,727 | 3.4% |
| Health/beauty | 14,378 | 3.3% |
| Shopping & retail | 14,010 | 3.2% |
| Hair Salon | 13,778 | 3.2% |
| Nonprofit organization | 12,308 | 2.8% |
| Beauty, cosmetic & personal care | 11,807 | 2.7% |
| Community | 11,348 | 2.6% |
| Entrepreneur | 10,780 | 2.5% |
| Medical & health | 9,836 | 2.3% |
Shares here are of the 433,175 tagged Pages, not the full 1,454,867-record index — category is sparse by construction, not a data-quality gap.
Straight from a likes_desc query — nothing hand-picked.
| Page | Category | Likes |
|---|---|---|
| Netflix | Media/news company | 205,290,748 |
| Netflix | Media/news company | 203,419,762 |
| Cristiano Ronaldo | Athlete | 174,459,436 |
| Samsung | Electronics Company | 162,703,901 |
| Samsung | Electronics Company | 162,532,871 |
| Samsung | Electronics company | 162,494,794 |
| Shakira | Musician/band | 126,849,178 |
| Mark Zuckerberg | — | 121,213,658 |
| YouTube | Product/service | 108,872,320 |
| Coca-Cola | Food & beverage | 107,070,257 |
| Coca-Cola | Food & beverage | 107,039,067 |
| Meta | Company | 106,606,348 |
One record per Page. Grouped for readability; the API returns a flat object. No post content or reviews — for those, use the live Facebook endpoints.
Identity
page_ididentifiertitlesource_urlcategoryContact
websiteemailphoneswhatsappaddressAudience
likesStatus & provenance
discovery_sourcestatusdiscovered_athydrated_atThe search endpoint takes the filters below; combine any of them. Page with page and page_size (≤100 per page, and page × page_size ≤ 10,000).
Full-text & identity
Contact fields
Audience
Dates & provenance
Sort
relevancelikes_desclikes_aschydrated_at_deschydrated_at_ascEvery query authenticates with an x-api-key header and reads the stored search index — there is no live crawl, proxy or Graph API access token to manage, and you are billed pay on success: charged for results, not failed requests. Three endpoints cover it:
GET /datasets/facebook-pages/search — filter, sort, page the population.GET /datasets/facebook-pages/items/{page_id} — one Page.GET /datasets/facebook-pages/facets — counts for category or discovery source.The same three are exposed as MCP tools — datasets_facebook_pages_search, datasets_facebook_pages_item and datasets_facebook_pages_facets — so an agent can call them directly. Need post content, reviews or a live read of a Page that is not in the index? Those are the live Facebook endpoints.
Cite this
Crawlora (2026). Facebook Pages Dataset. 1,454,867 public Facebook Page contact records, seeded index; public fields only. https://crawlora.net/datasets/facebook-pages.
Search with filters
# Restaurants with 10K+ likes, biggest first
curl "https://api.crawlora.net/api/v1/datasets/facebook-pages/search?category=Restaurant&min_likes=10000&sort=likes_desc" \
-H "x-api-key: $CRAWLORA_API_KEY"One Page by page_id
# One Page by page_id
curl "https://api.crawlora.net/api/v1/datasets/facebook-pages/items/100064500857993" -H "x-api-key: $CRAWLORA_API_KEY"Facet the population
# Facet the population by category (or discovery_source)
curl "https://api.crawlora.net/api/v1/datasets/facebook-pages/facets?facet=category" -H "x-api-key: $CRAWLORA_API_KEY"Pages with a phone number and website, biggest first
# Pages with a phone number and a website
curl "https://api.crawlora.net/api/v1/datasets/facebook-pages/search?has_phone=true&has_website=true&sort=likes_desc" \
-H "x-api-key: $CRAWLORA_API_KEY"The same figures behind the bars and the waffle — plain and machine-readable for search engines and AI answer engines that cannot parse a chart.
| Likes | Pages | Share |
|---|---|---|
| Under 1K | 852,568 | 58.6% |
| 1K–9.9K | 493,088 | 33.9% |
| 10K–99K | 94,207 | 6.5% |
| 100K–999K | 12,591 | 0.9% |
| 1M+ | 2,413 | 0.2% |
| Discovery source | Pages | Share |
|---|---|---|
| business_search | 1,096,838 | 75.4% |
| warc_domain_scan | 352,282 | 24.2% |
| wikidata | 5,714 | 0.4% |
Query 1.5M public Facebook Page contact records — website, email, phone, WhatsApp, category and like count — through one REST API. Business lead generation, local-market research or account enrichment: clean JSON, public fields only, pay on success.
The dataset holds 1,454,867 public Facebook Page contact records, deduplicated to one record per page_id. 33 records predate discovery-source tracking and carry no field at all, so the discovery-source breakdown sums slightly short of the total rather than landing in an "unknown" bucket.
No, and it matters. Facebook's own robots.txt blocks Common Crawl on Page paths, so unlike this catalog's other social datasets, this index cannot use a broad web-archive sweep as its main discovery method. Instead, 75.4% of attributed discoveries come from a business-listing search crawl, with smaller Wikidata and WARC-domain-scan slices filling in the rest. That method skews the population toward small local businesses with contact info on file: 58.6% of Pages have under 1,000 likes. Use it as a business-contact directory, not as a census or a basis for population-level claims about Facebook Pages.
Every field is a public, credential-free Page field read from the Page's public About tab: page_id, identifier, title, source URL, website, email, phone numbers, address, category, WhatsApp contact, like count, the discovery source, and a hydrated_at timestamp. Nothing behind authentication is collected — no login, no Graph API access token. There is no post content, no reviews and no private messages; for those, use the live Facebook endpoints instead.
Because these are business Pages' own self-published contact details, not personal-profile bios. 66.3% carry a public phone number, 57.3% an email, and 59.3% a website — a Page's About tab exists specifically to be found by customers, so the fields it publishes are contact-oriented by design, unlike a personal follower-focused social profile.
category is Facebook's own self-declared category field for Pages (e.g. "Restaurant", "Local business", "Coffee shop"). It is sparse by design — only 29.8% of the 1,454,867-record index carries a value. Filter on category to slice the tagged slice by business type; do not treat its absence as a data-quality gap.
Every record in this snapshot was hydrated between 2026年7月30日 and 2026年9月6日. Each record carries its own hydrated_at timestamp, and you can filter or sort on it (hydrated_after, hydrated_before, sort=hydrated_at_asc) to find the stalest records yourself.
Yes. Three endpoints cover it: search (full-text and faceted, with contact-field, like-count and category filters), one Page by page_id, and facet counts for category or discovery source. The same three are exposed as MCP tools — datasets_facebook_pages_search, datasets_facebook_pages_item and datasets_facebook_pages_facets — so an agent can call them directly. Reads hit the stored index and never trigger a live crawl; billing is pay-on-success.