website-contact-social-link-extractor

Production-oriented Apify Actor for shallow crawling of public websites to extract contact details, company clues, and social profile links.

It is intentionally narrow:

• Public data only
• No login, no session bypassing, no protected pages
• Best suited for company websites, agency sites, startup sites, directories, and public landing pages

What It Extracts

• Emails
• Phone numbers
• Contact form URLs
• Contact page URLs
• About, team, and legal/privacy page URLs
• Social profile URLs for LinkedIn, X, Twitter, Facebook, Instagram, YouTube, TikTok, and GitHub
• Schema.org organization clues and company-name hints
• Page title and meta description
• Confidence score and crawl metadata
• Source evidence showing where each signal was found

Input

Use startUrls, domains, or row-oriented inputs: googleMapsRows, businessRows, shopifyRows, serpRows, and generic rows.

Supported row fields are intentionally CSV-friendly: website, url, domain, link, displayedLink, storeUrl, shopDomain, shopifyDomain, and sourceUrl.

All seeds are normalized to public site homepages and deduped by site.

Example:

{
  "googleMapsRows": [{ "name": "Example Co", "website": "https://example.com" }],
  "serpRows": [{ "title": "Example Org Contact", "link": "https://example.org/contact" }],
  "rows": [{ "domain": "example.net" }],
  "maxPagesPerSite": 8,
  "maxDepth": 1,
  "maxConcurrency": 10,
  "requestTimeoutSecs": 20
}

See examples/sample-input.json.

Output

The actor writes one record per site to the Apify Dataset.

Example:

{
  "inputSource": "https://example.com",
  "siteUrl": "https://example.com/",
  "domain": "example.com",
  "title": "Example Company",
  "metaDescription": "Example Company public contact page.",
  "emails": ["hello@example.com"],
  "phones": ["+14155550101"],
  "contactFormUrls": ["https://example.com/contact"],
  "contactPageUrls": ["https://example.com/contact"],
  "aboutPageUrls": ["https://example.com/about"],
  "teamPageUrls": ["https://example.com/team"],
  "legalPageUrls": ["https://example.com/privacy"],
  "socialProfiles": {
    "linkedin": ["https://www.linkedin.com/company/example-company"],
    "x": ["https://x.com/examplecompany"],
    "twitter": [],
    "facebook": [],
    "instagram": [],
    "youtube": [],
    "tiktok": [],
    "github": []
  },
  "schemaOrg": {
    "types": ["Organization"],
    "names": ["Example Company LLC"],
    "sameAs": ["https://www.linkedin.com/company/example-company"]
  },
  "companyClues": {
    "names": ["Example Company", "Example Company LLC"],
    "legalMentions": ["Example Company LLC"]
  },
  "pagesCrawled": 4,
  "crawl": {
    "pagesCrawled": 4,
    "maxPagesPerSite": 8,
    "maxDepth": 1,
    "sameHostOnly": true,
    "shallow": true
  },
  "confidenceScore": 0.95,
  "evidence": {
    "emails": [{ "value": "hello@example.com", "sourceUrl": "https://example.com/contact" }]
  }
}

See examples/sample-output.json.

Crawl Strategy

• Seeds each site with /, /contact, /contact-us, /about, /about-us, and /team
• Follows same-host links only
• Prioritizes contact/about/team/legal-like links
• Stays shallow through maxDepth
• Caps work per site with maxPagesPerSite
• Surfaces crawl metadata in each result for easier QA downstream

Local Development

Install dependencies:

$npm install

Run the actor locally with Apify local storage:

$npm start

Run tests:

$npm test

Run deterministic smoke verification against a local HTML fixture server:

$npm run smoke

Run with Apify CLI:

$npx --yes apify-cli run

Notes

• The actor does not attempt to bypass access controls.
• The actor is intended for generic public website extraction, not scraping protected platforms.
• Some websites hide contact details behind JavaScript apps or images; those cases may produce partial results.