Website Contact & Tech Enricher

Give it a list of domains and get back one clean lead row per website:

• Emails: from mailto: links, page text, JSON-LD, Cloudflare-protected emails (data-cfemail) and common obfuscations like name [at] domain [dot] com. Image names (logo@2x.png), placeholders (john.doe@example.com), tracking addresses (Sentry, Wix) and invalid TLDs are filtered out. Emails on the site's own domain are listed first.
• Phone numbers: from tel: links and labelled text ("Phone:", "Tel.", "Telefon", "电话"...), validated with libphonenumber and returned in E.164 and international format, with country and line type.
• Social profiles: LinkedIn, Facebook, Instagram, X/Twitter, YouTube, TikTok, Pinterest, GitHub, Telegram, WhatsApp, Threads, Discord, Reddit, Medium, Snapchat, Vimeo and XING. Share buttons such as facebook.com/sharer and twitter.com/intent are skipped.
• Tech stack: about 90 technologies detected from HTML and response headers: CMS (WordPress, Drupal, Ghost...), e-commerce (Shopify, WooCommerce, Magento, BigCommerce), site builders (Wix, Squarespace, Webflow, Framer), analytics and ad pixels (GA, GTM, Meta Pixel, LinkedIn Insight, TikTok), chat and CRM widgets (Intercom, HubSpot, Zendesk, Tawk.to, Crisp), payments, CDN and hosting.
• MX check: does the domain receive email at all, and on which provider (Google Workspace, Microsoft 365, Zoho, Proofpoint, Mimecast...)?
• Page meta: title, description, language, og:site_name and og:image.

How it works

For each website the Actor loads the homepage, ranks internal links by how likely they are to hold contact details (Contact, Kontakt, Contacto, Impressum, About, Team, Support, Legal and more, in several languages), and visits the top ones, up to maxPagesPerSite. It uses plain HTTP requests rather than a headless browser, so it is fast and cheap: a local test of 10 sites with 6 pages each finished in about 9 seconds.

It's polite by default. It honours robots.txt (sites that disallow it are skipped and not charged), uses one connection per site, and identifies itself in the User-Agent.

Input

{
  "websites": ["allbirds.com", "https://www.python.org", "kinsta.com"],
  "maxPagesPerSite": 6,
  "detectTechnologies": true,
  "checkMx": true
}

Output (one row per website)

{
  "domain": "allbirds.com",
  "status": "ok",
  "title": "Allbirds ...",
  "primaryEmail": "help@allbirds.com",
  "emails": ["help@allbirds.com"],
  "primaryPhone": "+1 424 363 8064",
  "phones": ["+14243638064"],
  "socials": { "instagram": ["https://www.instagram.com/allbirds"], "facebook": ["..."], "tiktok": ["..."] },
  "linkedin": null,
  "technologies": ["Shopify", "Google Tag Manager", "Microsoft Clarity", "Cloudflare", "OneTrust"],
  "mx": { "domain": "allbirds.com", "hasMx": true, "provider": "Microsoft 365", "records": ["..."] },
  "emailDetails": [{ "email": "help@allbirds.com", "page": "https://www.allbirds.com/...", "onSiteDomain": true }],
  "phoneDetails": [{ "e164": "+14243638064", "country": "US", "type": "FIXED_LINE_OR_MOBILE", "source": "tel-link", "page": "..." }],
  "pagesCrawled": [{ "url": "https://www.allbirds.com/", "status": 200 }]
}

(Values above are from a real local test run; the live site may change.)

status is one of ok, unreachable, http_error, no_html, skipped_robots_txt or invalid_input. Only ok rows count as enriched.

Typical uses

• Enrich a CSV of company domains before outreach. You get emails, phones and LinkedIn URLs in one pass.
• Qualify leads by tech: find Shopify stores without a chat widget, or WordPress sites running Meta Pixel.
• Check whether domains can receive email (MX) before you send.
• Clean up an agency's or directory's prospect lists.

Notes and limits

• Only publicly published business contact details are extracted. You are responsible for using the data in line with GDPR/CAN-SPAM and the sites' terms.
• Sites that render all content with JavaScript, or that block datacenter traffic, may return fewer results. You can enable a proxy for those.
• Big sites with no published email (for example, BBC) will correctly return no emails.

Found a site that parses wrongly? Open an issue on the Actor page with the domain and we'll fix it.