Email Finder — Name + Domain → Best-Guess Work Email

Give this actor a person's name and their company domain, and it returns the most likely work email — with a numeric confidence score, MX validation, and a ranked list of alternative candidates. No API key, no signup, no monthly seat.

What you actually get: a ranked, MX-gated best guess, not a hard-confirmed mailbox. The actor attempts live SMTP verification, but most cloud networks (including Apify's) block outbound port 25, so it degrades to MX + pattern ranking and labels the method mx_pattern. SMTP confirmation only fires where the run network allows it.

Why this one

Most email-finder actors hand back a single guessed address and, at best, a catch_all flag. That leaves you to guess whether the address is real. This actor instead:

• Builds frequency-ranked candidate patterns — first.last, flast, first, firstlast, and ~9 more — weighted by how common each scheme actually is in corporate email.
• Validates what it can — one MX lookup per domain to confirm the mail server exists, then (where the network allows) a catch-all probe and RCPT verification of candidates in priority order. A confirmed mailbox is marked verified; on port-25-blocked networks the best guess is marked unverified_guess with its MX host and confidence.
• Tells you the truth — every record carries confidence (0–1), the verificationMethod (smtp / mx_pattern / none), isCatchAll, isDisposable, isRoleAccount, and the ranked alternativeEmails so you can fall back if the top pick bounces.

If the run network blocks outbound port 25 (common on cloud), the finder degrades gracefully to MX + pattern ranking and labels the method mx_pattern — you still get a ranked, MX-gated best guess instead of a hard failure.

Input

{
  "people": [
    { "firstName": "Patrick", "lastName": "Collison", "domain": "stripe.com" },
    { "firstName": "Tobias", "lastName": "Lutke", "domain": "shopify.com" }
  ],
  "verifySmtp": true,
  "maxAlternatives": 5,
  "smtpTimeoutSecs": 8,
  "maxItems": 0
}

Field	Type	Notes
people	array	Each item: firstName (required), lastName or surname, optional fullName, and domain (required).
verifySmtp	boolean	Attempt live SMTP RCPT verification. Auto-falls back to MX + pattern if port 25 is blocked. Default true.
maxAlternatives	integer	Ranked alternatives returned per person. Default 5.
smtpTimeoutSecs	integer	Per-probe SMTP timeout. Default 8.
maxItems	integer	Total output cap, 0 = unlimited.

Output (per person)

{
  "firstName": "patrick",
  "lastName": "collison",
  "domain": "stripe.com",
  "email": "patrick.collison@stripe.com",
  "confidence": 0.45,
  "status": "unverified_guess",
  "verificationMethod": "mx_pattern",
  "isCatchAll": null,
  "isDisposable": false,
  "isRoleAccount": false,
  "mxFound": true,
  "mxHost": "aspmx.l.google.com",
  "patternUsed": "first.last",
  "alternativeEmails": [
    { "email": "pcollison@stripe.com", "pattern": "flast", "confidence": 0.319, "smtpStatus": "not_checked", "smtpCode": null }
  ],
  "parse_confidence": 0.9,
  "warnings": ["smtp_unreachable_port25"]
}

The example above is a real Apify-cloud result (port 25 blocked → mx_pattern). On a network with outbound SMTP open, status becomes verified/accept_all, verificationMethod becomes smtp, and confidence rises accordingly.

status values: verified (SMTP confirmed the mailbox) · accept_all (domain accepts all mail — can't single one out) · unverified_guess (MX present, SMTP not usable, ranked best guess) · no_valid_pattern (SMTP worked but no candidate accepted) · no_mx (domain has no mail server).

Pricing

Pay-per-result. You are charged once per person record produced — never for an invalid input. You run it on your own Apify account, so compute is billed to you at platform cost.

More scrapers from our toolkit

Building a data pipeline? These actors pair well with this one — each runs on your own Apify account with the same pay-per-result pricing, no subscription:

• Epa Data Scraper
• Firmy Cz Scraper
• Gelbeseiten Directory
• Google Maps Leads
• Google Maps Scraper
• Hiring Signal Monitor

Chain any of them together from the Integrations tab (the Run succeeded trigger) to build a multi-step workflow — one actor's output feeds the next.