Corporate Email Finder & Verifier

B2B email discovery from a person name and company domain. Syntax, MX, disposable-domain, and role-based signals are checked in the same run. Results are heuristic signals for small, controlled lead-validation work—not delivery guarantees.

Modes

Finder + Verifier

Provide fullName and preferably domain. The Actor generates up to five simple patterns, verifies each, and saves only the best result.

{
  "fullName": "Taro Yamada",
  "domain": "example.co.jp",
  "maxCandidates": 5
}

Verifier only

Provide emails to verify one address per Dataset record.

{
  "emails": [
    "taro.yamada@example.co.jp",
    "not-an-email"
  ]
}

domain is preferred. If it is absent, a domain-like value embedded in companyName can be used. The Actor does not query a company directory or LinkedIn.

Output example

{
  "inputName": "taro yamada",
  "inputDomain": "example.co.jp",
  "email": "taro.yamada@example.co.jp",
  "patternUsed": "first.last",
  "isValidSyntax": true,
  "hasMx": true,
  "isDisposable": false,
  "isRoleBased": false,
  "score": 85,
  "status": "risky",
  "checkedAt": "2026-08-05T00:00:00.000Z"
}

valid means the lightweight checks passed. risky means the address has a usable syntax/MX signal but also has an uncertainty or risk flag. invalid means syntax or MX failed. not_found means no usable candidate could be resolved.

Billing

The PPE event is email-resolved, initially $0.03 per event in code. Configure the matching event in Apify Console only after human approval; this repository does not publish the Actor or change production pricing.

Only a result with status valid or risky is billable, and charging is attempted after that result is saved to Dataset. not_found, invalid, and failed requests are not charged. The run checks chargedCount and stops safely when the buyer's charge limit is reached.

Dataset contains one record per Finder request or one record per supplied email. Therefore the billing reconciliation is billableSavedCount == chargedCount; total Dataset rows also include non-billable invalid and not_found records.

Limits and safe use

• Candidate generation uses only five simple patterns: first.last, firstlast, f.last, first_last, and first.
• Names are normalized with Unicode compatibility normalization, whitespace collapsing, and lowercasing. Japanese and English names use simple token splitting; transliteration is not attempted.
• MX and a short SMTP TCP connection are signals only. The Actor does not send mail, perform RCPT TO, or guarantee delivery, mailbox ownership, or consent.
• Disposable and role-based lists are small built-in baselines and are not complete.
• Do not send large repeated runs or use the output for unsolicited bulk outreach. Test a small set first.
• No paid enrichment API, LinkedIn login, cookie, CRM integration, dashboard, or AI copy generation is used.

How to use

1. Select Try for free in Apify.
2. Run one Finder input and one or two Verifier emails.
3. Inspect the Dataset and score/status before using any result.
4. Increase volume only after a small test is useful and the billing preview is understood.