PublicLink Dataset

The PublicLink dataset identifies relationships between business domains by matching shared public contact information — phone numbers, email addresses, and social media URLs — extracted from their websites. When two domains publish the same phone number, email, or social profile, they are linked as related entities.

Use Cases

• Corporate group mapping: Identify brands, subsidiaries, and regional sites owned by the same organization (e.g., acmehotels.com and acmeresorts.com sharing privacy@acmehotels.com)
• M&A and portfolio analysis: Detect domains that began sharing contact info after an acquisition
• Lead deduplication: Avoid reaching out to the same organization through multiple domains
• Fraud and spam detection: Flag networks of domains sharing identical contact details, which can indicate lead-gen farms or templated sites

Link Sources

Source	How it works
email	Identical email addresses extracted from both websites (e.g., info@company.com found on two different domains)
social	Identical social media profile URLs extracted from both websites (e.g., the same LinkedIn company page linked from two domains)
phone	Identical phone numbers extracted from both websites

Flat-File Layout

Name	Format	Description
domain	String	Normalized domain name (excluding any subdomain)
linked_domain	String	Normalized linked domain name (excluding any subdomain)
record_date	Date (YYYY-MM-DD)	Date the record was compiled
source	String	Link method: email, social, or phone
link_values	Array[String]	The shared values establishing the link (email addresses, phone numbers, or social media URLs)

Example Record

{
  "domain": "acmehotels.com",
  "linked_domain": "acmeresorts.com",
  "source": "email",
  "link_values": ["privacy@acmehotels.com", "dpo@acmehotels.com"],
  "record_date": "2024-07-17"
}

API Access

Query PublicLink relationships via the PublicLink API (Enterprise plan) or the public-company-links MCP tool. Filter by domain and link source.

Delivery

Available as JSON or delimited flat files. Updated periodically as new website crawl data is processed.