Email Finder & Verifier – B2B Emails by Name & Company

Find a person's work e-mail from their name and company, list every reachable address on a domain, or verify the addresses you already have — live, with the pattern, a confidence and exactly what the mail server confirmed on every row. Give it Jane Doe, acme.com (or the rows of a people finder), a list of domains, or a list of addresses. Nothing is looked up in a resold database: the company's own address pattern is learned from its website, candidates are generated from the person's name, and each is checked against the domain's mail server.

E-mail tools on the market either resell stale databases by the thousand, or guess a pattern and call it "verified". This Actor does the work an outbound team does by hand — read the company site for addresses and names, learn how the company forms its addresses, generate the person's, check it — and reports the depth of every check instead of a single green tick.

Email finder & verifier: what you get

One row per person, per domain address, or per verified address:

Field	What it tells you
email, status	The address and its status — found / not_found for people; valid (mailbox confirmed), risky (catch-all domain), invalid (no mail server, rejected, bad syntax, disposable) or unknown (mail server only) for addresses
pattern, confidence, domainPattern	How the address is formed (first.last, flast, …), a 0–100 confidence (100 = seen on the company site for this person; 85 / 70 = the company's learned pattern; 45 and below = by prevalence; 95 = mailbox confirmed), and the pattern learned for the whole domain
verified, verification, catchAll, mx	verified is true only when the mailbox was confirmed; verification says how far the check went (smtp-valid, smtp-catch-all, mx-only, no-mx, syntax-only); catch-all flag and the mail exchanger
type, roleAddress, freeMail, disposable	Personal vs role address (info@, sales@…), free-mail and disposable flags
personName, firstName, lastName, company, domain	The person and company, ready for merge fields and joins
source, sourceUrl, candidates[], reason	Where the address came from (website:/team, search, pattern, input), the alternatives that were tried with their own verification, and why nothing was delivered when that is the case

Three dataset views: E-mails (one line per person / address), Verification (what was confirmed), By domain (addresses per domain with the learned pattern and the page they were found on).

How it works

1. Learn the domain — the home page and the contact, about, team and imprint pages are read for addresses (hidden ones decoded) and named people. Addresses that match a named person teach the domain's pattern with strong evidence; other on-domain addresses teach it by shape. Optionally one web search per domain adds addresses seen elsewhere.
2. Find the person — an address seen on the site for that person wins outright; otherwise candidates are generated from the person's name in the learned pattern first, then by how common each pattern is across companies.
3. Verify — syntax, disposable and free-mail checks, the domain's mail server (MX), and — where the network allows an SMTP conversation — the mailbox itself with a catch-all probe. verified never says more than the mail server did.
4. Deliver — the best address with its candidates for people; every reachable address for domains (personal and role, plus the learned pattern applied to the people named on the site); the result for each address in a verify list.

No login, no cookies, no third-party data vendor, no nested scrapers.

Chain it into a prospecting pipeline

• In: the rows of a people finder ({ name | firstName + lastName, company, companyDomain }) can be passed as people unchanged; the domains of a company list as domains.
• Out: rows keyed by domain with email, confidence and verification — filter on verified or a confidence threshold and send to a sequencer.

Who uses it

• Outbound / GTM agents — a keyword-discoverable, MCP-callable step that turns people into deliverable addresses unattended.
• SDR teams — the work address for every decision maker on a target list, with a confidence you can route on.
• Ops and data teams — a bulk verifier for existing lists, and a per-domain address inventory.

Set it up in a minute

1. Paste People (Name, domain per line), Domains, and/or E-mails to verify.
2. Start. Rows arrive as each item is finished; the run summary is in the OUTPUT record.

Verification and discovery settings are collapsed below the essentials; the defaults are tuned for unattended runs.

Use it from an agent or the API

{ "people": ["Guillaume Moubeche, lemlist.com", "Tope Awotona, calendly.com"] }

{ "people": [{ "firstName": "Jessie", "lastName": "Sloan", "companyDomain": "gong.io" }], "domains": ["syberry.com"], "emails": ["contact@lemlist.com"] }

Agents may also pass contacts / names, companies / websites, addresses, and objects with email. Results are in the default dataset (?view=emails, ?view=verification, ?view=domains); the run summary is in the OUTPUT record of the run's key-value store. Through the Apify MCP server, call inovaflow/email-finder-verifier with the same input.

Output example

{
  "mode": "person",
  "email": "tope.awotona@calendly.com",
  "status": "found",
  "type": "personal",
  "personName": "Tope Awotona",
  "firstName": "Tope",
  "lastName": "Awotona",
  "domain": "calendly.com",
  "pattern": "first.last",
  "confidence": 70,
  "verified": false,
  "verification": "mx-only",
  "catchAll": null,
  "mx": "aspmx.l.google.com",
  "source": "pattern",
  "domainPattern": "first.last",
  "candidates": [
    { "address": "tope.awotona@calendly.com", "pattern": "first.last", "confidence": 70, "verified": false, "verification": "mx-only" },
    { "address": "tope@calendly.com", "pattern": "first", "confidence": 35, "verified": false, "verification": "mx-only" }
  ]
}

Pricing

Pay per event: $0.01 per e-mail address delivered (a person with an address, or a domain with at least one address) and $0.001 per e-mail check in a verify list, plus a small per-run start fee. People and domains where nothing is found are free.

Good to know

• Verification depth depends on the network. Where outbound SMTP is blocked, addresses are verified down to the domain's mail server (mx-only) and delivered with their pattern confidence; where SMTP is reachable, mailboxes are confirmed (smtp-valid) and catch-all domains are flagged (risky).
• A company name without a domain is resolved to its website only when a page verifiably names the company; otherwise the person is delivered as not_found with reason: company_not_resolved — pass the domain for the best results.
• Role addresses (info@, sales@) are returned in domain lists (switch off with Include role addresses) and flagged roleAddress: true in verify lists.
• Results reflect what companies publish; source, sourceUrl and candidates[] carry the evidence for every row.