Lead Contact Verifier

Stop paying for a list of "leads" you can't tell apart from a list of dead links.

Give this actor any list of website URLs — pulled from a Google Maps scraper, a LinkedIn export, a CSV you already have — and it comes back with every email address it can find on each site, tagged with two honest signals: can this domain even receive mail (MX record present), and is this a named contact, a generic role account, or a personal free-mail address (not actually the company's own domain). Plus whatever LinkedIn/X/Instagram/Facebook links are on the page.

Why not just scrape emails?

Most email scrapers stop at the regex match. That leaves you with info@ addresses mixed in with real named contacts, and zero way to tell a working domain from a dead one without sending anything and hoping it doesn't bounce.

What you get (one row per email found)

{
  "input_url": "https://acme-plumbing.com",
  "domain": "acme-plumbing.com",
  "checked_at": "2026-08-11",
  "email": "jane@acme-plumbing.com",
  "source": "mailto",
  "deliverability": "mx_confirmed",
  "address_type": "corporate_personal",
  "linkedin": "https://linkedin.com/company/acme-plumbing",
  "x_twitter": null,
  "instagram": null,
  "facebook": "https://facebook.com/acmeplumbing"
}

• source: mailto (the address was wired into a clickable mailto: link — high confidence, a real business has no reason to fake this) or text (the address only appeared as plain text — usually a real contact too, but this is also where fictional example addresses in marketing/product copy show up; found live testing this actor against stripe.com, which surfaces a made-up customer email in a product mockup on some page variants). Treat text as "worth a glance," mailto as "trust it."
• deliverability: mx_confirmed (the domain has a real mail server — as far as any check can tell you without sending) or no_mx_record (the domain provably cannot receive mail — drop this one)
• address_type: corporate_personal (a named address on the company's own domain — highest value), role_or_generic (info@, contact@, support@ — still usable, just not a person), or personal_free_provider (gmail.com/yahoo.com etc — a real person, but not the company's own domain)

If a site has no email at all, you still get one row for it (email: null) so your input list is always fully accounted for in the output — no silent drops.

Input

• urls (required): list of website URLs or bare domains
• extra_paths (optional, default /contact, /contact-us, /about, /about-us): pages tried if the homepage has no email — stops as soon as one is found
• request_timeout_secs (optional, default 15)

Known limitation — read before you buy

No live SMTP verification. A real mailbox check (connecting to the mail server and asking if a specific address exists) needs outbound port 25, which is blocked on most cloud infrastructure — including, as far as this actor's testing could confirm, typical hosted run environments — specifically to stop spam. mx_confirmed is the strongest signal obtainable without it: it proves the domain can receive mail, not that this exact mailbox does. Treat no_mx_record as a hard "don't bother"; treat mx_confirmed as "worth trying," not as a guarantee.

Marketing copy can contain fictional example emails. Product mockups and docs sometimes show a made-up customer address as plain text (not a real contact). This is exactly why every row carries a source field — mailto addresses are wired into an actual link and safe to trust; text matches are usually real too but occasionally this kind of noise, so treat them as lower confidence.

Larger companies often don't publish an email at all. Enterprise and SaaS sites increasingly route everything through a contact form instead of a plain-text address (confirmed testing this actor against several real company sites). This actor works best on the kind of small/local business that a Google Maps lead list is usually made of — plumbers, clinics, agencies, shops — where a direct email in the footer is still the norm, not on Fortune-500 sites.

Pricing

Pay per result ($0.02/row) — no charge for input URLs that produce nothing.