Website Contact & Email Extractor

Extract email addresses, phone numbers, social media profile links, and physical addresses from any website — fast, without a browser.

What it does

For each URL you provide, the actor crawls the main page plus the most common contact sub-pages:

• /contact
• /about
• /team
• /support
• /impressum
• /contact-us
• /about-us

From each page it extracts:

Signal	Method
Email addresses	Regex over HTML source + visible text; validated and deduplicated
Phone numbers	Regex supporting international and US formats
Social media links	Anchor tag scanning for Twitter/X, LinkedIn, Facebook, Instagram, GitHub
Physical address	JSON-LD structured data (schema.org/PostalAddress) with HTML fallback (<address> tag, class/id heuristics)

Results from all crawled pages are merged and deduplicated before being stored.

Input

{
  "urls": [
    "https://apify.com",
    "https://example.com"
  ],
  "contactPaths": ["/contact", "/about", "/team", "/support", "/impressum", "/contact-us", "/about-us"],
  "requestTimeoutSecs": 15,
  "maxRetries": 2,
  "proxyConfiguration": { "useApifyProxy": true }
}

Field	Type	Required	Default	Description
urls	string[]	Yes	—	List of website URLs to process
contactPaths	string[]	No	See above	Sub-paths to probe on each domain
requestTimeoutSecs	integer	No	15	Per-request timeout (5–60)
maxRetries	integer	No	2	Retries on transient failures (0–5)
proxyConfiguration	object	No	{}	Apify proxy settings

Output

Each website produces one dataset record:

{
  "url": "https://apify.com",
  "domain": "apify.com",
  "emails": [
    "hello@apify.com"
  ],
  "phones": [
    "+1 (555) 123-4567"
  ],
  "socialMedia": {
    "twitter": ["https://twitter.com/apify"],
    "linkedin": ["https://linkedin.com/company/apify"],
    "github": ["https://github.com/apifytech"]
  },
  "address": "Vodickova 704/36, Prague 1, 110 00, CZ",
  "pagesScraped": [
    "https://apify.com",
    "https://apify.com/contact"
  ],
  "scrapedAt": "2024-01-15T10:30:00+00:00",
  "error": null
}

If a page returns a 404, times out, or otherwise fails, it is skipped silently and the rest of the run continues. Any non-fatal per-page errors are noted in the top-level error field of the result.

Pricing

This actor uses pay-per-event charging: one charge event per website processed, regardless of how many sub-pages were crawled.

Technical details

• Runtime: Python 3.11
• HTTP client: httpx (async, no browser overhead)
• HTML parser: BeautifulSoup4
• Concurrency: All sub-pages for a single website are fetched concurrently; websites are processed sequentially to stay polite
• Proxy support: Pass an Apify proxy configuration to rotate IPs and avoid blocks

Limitations

• Does not execute JavaScript — contact information loaded dynamically by JS will not be found. Use the Web Scraper for JS-heavy sites.
• Phone number extraction is heuristic and may include false positives (e.g. dates, zip codes) on some pages.
• Email blacklisting filters obvious placeholders and CDN/framework addresses but cannot catch every false positive.