Extract Website Contact Data – Email, Phone, Social

Apify Actor to extract contact data from websites: emails, phone numbers, and social media profiles.

Description

This Actor processes a list of domains and automatically extracts:

1. ✅ Emails: Detection from mailto links, raw text, and JSON-LD schemas. Automatic normalization, filtering, and deduplication.
2. ✅ Phone Numbers: Extraction from tel: links and raw text. E.164 normalization with automatic country detection.
3. ✅ Social Media: LinkedIn, Facebook, Instagram, Twitter/X, TikTok, YouTube, Pinterest, Google Maps. Filtering of share links and service links.

Features

• ✅ Intelligent Crawling: Automatic detection of key pages (contact, about, legal, privacy). Adaptive crawl (8-15 pages depending on site structure)
• ✅ Fast and Efficient: Uses Cheerio/HTTP by default, Playwright only as fallback for dynamic pages
• ✅ Deterministic: Stable and traceable results (sourceUrl + snippet for each extraction)
• ✅ Deduplication: Emails and phones automatically deduplicated
• ✅ Smart Selection: Primary email and phone selected according to precise rules
• ✅ E.164 Normalization: Phone numbers normalized to international format with automatic country detection
• ✅ Smart Filtering: Automatic exclusion of test emails, public authorities, invalid numbers
• ✅ Resilience: Automatic error handling, retry on timeout, attempts on URL variants (http/https, www/non-www)

Input

Required Parameters

• startUrls (array or string): List of URLs to process. Array format [{ url: "https://example.com" }] or multi-line text (one URL per line).

Optional Parameters

• timeoutSecs (number, default: 30): Request timeout in seconds (5-120)
• usePlaywrightFallback (boolean, default: true): Use Playwright for dynamic pages if HTTP fails
• includeContacts (boolean, default: true): Extract emails and phones
• includeSocials (boolean, default: true): Extract social media links
• keyPaths (array, default: []): Custom paths to override default key paths

Input Example

{
  "startUrls": [
    { "url": "https://example.com" },
    { "url": "https://another-domain.com" }
  ],
  "timeoutSecs": 30,
  "includeContacts": true,
  "includeSocials": true
}

Output

A single JSON record per domain in the default dataset.

Record Structure

{
  "domain": "example.com",
  "finalUrl": "https://example.com",
  "keyPages": {
    "contact": "https://example.com/contact",
    "about": "https://example.com/about",
    "legal": "https://example.com/legal",
    "privacy": "https://example.com/privacy"
  },
  "pagesVisited": [
    "https://example.com",
    "https://example.com/contact",
    "https://example.com/about"
  ],
  "emails": [
    {
      "value": "contact@example.com",
      "type": "general",
      "priority": "primary",
      "signals": ["mailto", "same_domain"],
      "sourceUrl": "https://example.com/contact",
      "snippet": "Contact us at contact@example.com",
      "foundIn": "mailto"
    }
  ],
  "primaryEmail": "contact@example.com",
  "phones": [
    {
      "valueRaw": "+33 1 23 45 67 89",
      "valueE164": "+33123456789",
      "priority": "primary",
      "signals": ["tel", "footer_or_contact"],
      "sourceUrl": "https://example.com/contact",
      "snippet": "Call us: +33 1 23 45 67 89"
    }
  ],
  "primaryPhone": "+33123456789",
  "socials": {
    "linkedin": [
      {
        "url": "https://www.linkedin.com/company/example-corp",
        "handle": "example-corp",
        "sourceUrl": "https://example.com"
      }
    ],
    "facebook": [
      {
        "url": "https://www.facebook.com/examplecorp",
        "handle": "examplecorp",
        "sourceUrl": "https://example.com"
      }
    ]
  },
  "errors": []
}

Main Fields

• domain: Registrable domain (e.g., "example.com")
• finalUrl: Final URL after redirects
• keyPages: Detected key pages (contact, about, legal, privacy)
• pagesVisited: List of crawled pages for this domain
• emails: List of extracted emails with metadata
• primaryEmail: Primary email selected (same-domain > mailto > contact page)
• phones: List of extracted phones with E.164 normalization
• primaryPhone: Primary phone selected (footer/contact > tel: > E.164)
• socials: Social media by platform
• errors: Errors encountered during crawl (if present)

Crawl Strategy

Priority Key Pages

The Actor automatically detects and visits the following key pages:

• Contact: /contact, /contact-us, /nous-contacter
• About: /about, /about-us, /a-propos
• Legal: /legal, /mentions-legales, /imprint
• Privacy: /privacy, /politique-de-confidentialite

Crawl Tiers (Internal)

The Actor uses two internal crawl tiers (non-configurable):

• Standard: Maximum 8 pages per domain (default)
• Deep: Maximum 15 pages per domain (automatic activation)

Deep mode is automatically activated if:

• The site is highly structured (4+ relevant key pages)
• A Playwright fallback is required for dynamic pages

Important: Tier change does not affect output. A single record is always produced per domain.

Extraction

Emails

• Detection: mailto: links, raw text (regex), JSON-LD schema.org
• Normalization: Lowercase, trim, final punctuation removal
• Filtering: Excludes noreply, donotreply, example, test, public authorities (agpd.es, cnil.fr, etc.), test emails (mail.com, example.com, etc.)
• Deduplication: On normalized email (lowercase)
• Primary selection: Same-domain > mailto > contact page > first valid
• Validation: Exclusion of emails concatenated with phone numbers

Phone Numbers

• Detection: tel: links, raw text (international regex)
• Normalization: valueRaw (original) + valueE164 (if possible via libphonenumber-js)
• Country detection: Automatic from URL (TLD, subdomain) and context
• Filtering: Excludes SIRET, VAT, non-phone numbers, fax, GPS coordinates, dates
• Deduplication: On valueE164 if available, otherwise digitsOnly(valueRaw)
• Primary selection: Footer/contact > tel: > E.164 > first valid
• Validation: Exclusion of invalid numbers (>15 digits, incorrect formats)

Social Media

• Platforms: LinkedIn (company), Facebook, Instagram, Twitter/X, TikTok, YouTube, Pinterest, Google Maps
• Filtering: Excludes share links, settings/policies, services (Wix, Dropbox, Google Drive, OneDrive)
• Deduplication: By normalized URL and handle
• Validation: Exclusion of individual Instagram posts, internal links

Error Handling

• Retry: Automatic attempts on timeout/network/429/5xx only
• No retry: On 404 (page not found)
• Timeout: Per request (timeoutSecs), no global timeout per domain
• Resilience: Errors are recorded in errors[] without blocking processing
• URL Variants: Automatic attempts on variants (http/https, www/non-www, hyphens)

Limitations

• Maximum 200 domains per execution
• No proxy (direct crawl)
• No configurable robots.txt respect
• No OCR or image scraping
• Single result per domain (www/non-www canonicalization)