Contact Details Scraper: Emails, Phones, Socials

Give it a list of website URLs. It crawls each site's contact-style pages and returns one item per site with emails, phone numbers (normalised to E.164) and social profile links.

Use it as a website email scraper, email extractor, contact info scraper, phone number scraper or social media links scraper for lead lists, CRM enrichment or supplier research. You pay $0.03 only for sites where a contact was found, plus $0.001 per page.

What it does

• Follows only links on the same registrable domain whose URL or link text matches contact|about|impressum|kontakt|team|legal|privacy|support|company, plus the start page. Pages per site are capped.
• Emails: from mailto: links and page text; decodes Cloudflare email protection (data-cfemail and /cdn-cgi/l/email-protection#...); decodes name [at] domain [dot] com style obfuscation; lowercases, de-duplicates, drops false positives (logo@2x.png, Sentry-style tracking ids, example.com, name@company.com placeholders).
• Phones: from tel: links, validated and normalised to E.164 with libphonenumber. Numbers without an international prefix use defaultCountry. Optionally parses phone-like text too (includePhonesUncertain).
• Socials: LinkedIn, Facebook, Instagram, X/Twitter, YouTube, TikTok, Pinterest, Discord.
• sourcePages tells you the first page where each email or phone was seen.

Built on Crawlee CheerioCrawler (plain HTTP, no browser), concurrency 10. No proxies, captcha solving or stealth tricks: pages that block plain requests are not scraped.

Input

Field	Type	Default	Meaning
startUrls	array	required	Website URLs, one result per site
maxPagesPerSite	integer	10	Page cap per site, including the start page
maxDepth	integer	2	Link depth from the start page (0 = start page only)
defaultCountry	string	US	ISO-2 code for parsing local phone numbers
includePhonesUncertain	boolean	false	Also parse phone-like numbers in plain text (more hits, more false positives)

{ "startUrls": [{ "url": "https://www.nzpost.co.nz/" }], "maxPagesPerSite": 8, "maxDepth": 2 }

Output

Real item from a local run (see samples/sample-output.json for all nine sites):

{
  "domain": "nzpost.co.nz",
  "startUrl": "https://www.nzpost.co.nz/",
  "emails": ["oversize@nzpost.co.nz", "youshop@nzpost.co.nz", "yspickup@nzpost.co.nz", "cyber.security@nzpost.co.nz"],
  "phones": ["+6439831425", "+6448975675", "+6469528600", "+6479600468", "+6499770102"],
  "phonesRaw": ["+6439831425", "+6448975675", "+6469528600", "+6479600468", "+6499770102"],
  "socials": {
    "linkedin": ["https://linkedin.com/company/nzpost"],
    "facebook": ["https://facebook.com/newzealandpost"],
    "instagram": ["https://instagram.com/newzealandpost"],
    "twitter": ["https://x.com/nzpost"],
    "youtube": [], "tiktok": [], "pinterest": [], "discord": []
  },
  "pagesScraped": 8,
  "sourcePages": {
    "+6439831425": "https://www.nzpost.co.nz/contact-support",
    "cyber.security@nzpost.co.nz": "https://www.nzpost.co.nz/contact-support/scams-and-fraud"
  },
  "scrapedAt": "2026-10-26T12:57:33.189Z"
}

(sourcePages shortened here.) Social links are canonical profile URLs: https, no www, no tracking query, twitter.com → x.com, and share/embed/post links removed. A site that cannot be loaded returns an error field and is not charged contact-site.

You are responsible for using extracted contact data lawfully (e.g. GDPR/CAN-SPAM). The actor only reads publicly published pages.

Pricing (pay per event)

Event	Price	Charged
page-scraped	$0.001	per page fetched and parsed
contact-site	$0.03	per site where at least one email or phone was found
apify-actor-start	$0.00005	once per run (Apify's standard start event)

Platform compute is included: you pay only the events above. Sites that fail to load are not charged contact-site. The actor stops gracefully when your max-charge limit is reached.

Measured locally

9 public sites, maxPagesPerSite 8, maxDepth 2: 69 pages, about 19 s crawl (about 29 s including start-up), all 9 sites returned at least one email or phone. On the Apify platform: 3 sites in 19.7 s at 1 GB memory.

Limitations

• Static HTML only. Contacts rendered by JavaScript, or behind forms, images or logins, are not found.
• Sites that return 403 or captchas to plain HTTP requests are not bypassed.
• Phones come from tel: links by default. Numbers shown only as text need includePhonesUncertain, and can include false positives (for example, fragments of longer numbers).
• Same-domain detection uses the public suffix list: for hosts that are themselves public suffixes (for example nhs.uk), the exact hostname is used, so www and non-www variants count as different sites. Use the hostname that the site actually serves.
• Test coverage: decoder, filters, phone and social extraction are tested against real captured pages (including a real Cloudflare-protected page from wpbeginner.com). [at]/[dot] de-obfuscation is tested only on synthetic text because no real page using it was found.
• Email addresses and phone numbers are personal data in some contexts. Make sure your use complies with applicable law and each site's terms.