Email & Phone Verifier

Email & Phone Verifier checks a list of e-mail addresses and phone numbers and tells you which ones you can actually use. Every address is checked for syntax and then looked up against its domain's mail servers, so you learn whether that domain accepts mail at all. Every number is validated against the North American numbering plan and tagged with its type, the US state of its area code and the time zone to call it in. Each row comes back with a verdict, the reason behind it, and a 0–100 contact score you can sort an outreach list by.

It can also work the other way round: give it a company name and it finds the company's website and any address published on it, so a lead with no contact stops being a dead row.

It takes a plain list, a set of contact objects, or the dataset of another Actor — so a lead list you scraped five minutes ago can be cleaned without moving data by hand. No login, no API key, no third-party service: the checks are DNS lookups plus open reference data, which is why they cost almost nothing and leak nothing.

What is Email & Phone Verifier?

It is a bulk email verification and phone number validation tool for people who send cold e-mail, run dialers, or buy lead lists and suspect half of it is junk. Instead of a single yes/no it returns the evidence: whether the domain exists, whether it publishes mail servers, who runs that mail, whether the address is a role account like info@, whether the mailbox is a free consumer one, whether the domain is a known throwaway service, and whether a number is even dialable.

What it checks

Check	What you get	Why it matters
Syntax	Bad addresses rejected before anything else	Typos and pasted junk never reach your sending tool
Domain exists	domain_not_found when the domain does not resolve	The single most common cause of hard bounces
Mail servers (MX)	deliverable when the domain accepts mail, domain_accepts_no_mail when it refuses	A domain that publishes no mail server can never receive your message
Throwaway domains	emailIsDisposable, from a list of 8,700 services	Sign-up spam and burner addresses out of your CRM
Role accounts	emailIsRole for info@, sales@, billing@ …	Deliverable, but nobody's personal inbox — different campaign, different expectations
Free mailboxes	emailIsFree for Gmail, Yahoo, Outlook …	A business contact on a free mailbox is usually a sole trader
Mail host	emailProvider: google, microsoft, zoho, proofpoint …	Tells you what filtering your message will face
Typos	emailSuggestion when the domain is one or two letters off a big provider	gmial.com becomes a recovered contact instead of a bounce
Numbering plan	valid / invalid with the rule that failed	Impossible numbers never reach a dialer
Number type	geographic, toll-free, premium, non-geographic, international	A toll-free number is a switchboard, not a person
Location	phoneState, phoneTimezone from the area code	Call inside business hours instead of at 6 a.m.
Formats	phoneE164 and a readable form	Ready for a CRM field or a dialer, whatever it expects
Website discovery	website, websiteStatus from the company name	A lead with no contact becomes a lead with a site
Address discovery	foundEmails, foundEmailSource	Reads the address the company publishes itself

What it does not do, and why

This is the part most tools in this category are quiet about, so here it is plainly.

We do not knock on the mailbox. Confirming that one specific mailbox exists needs an SMTP conversation on port 25, and cloud platforms — Apify included — block outbound port 25 entirely. We measured it rather than assumed it: a lookup of the mail servers answers in about seven milliseconds, and a connection to those same servers on port 25 times out. Any serverless tool promising per-mailbox certainty is either routing through a private relay it does not mention, or guessing.

And even where port 25 is open, the answer lies. Gmail and Microsoft 365 accept any recipient at that stage and bounce later, so a "verified" verdict from that method is worth little on exactly the domains most of your list sits on.

So deliverable here means the domain accepts mail and the address is well formed — the check that removes the large majority of hard bounces. It does not mean a person is behind it. That distinction is why the reason and the flags are on every row: you decide what to send where.

Line type is missing on purpose too. Whether a US number is a mobile or a landline cannot be determined from free data — the authoritative databases are licensed commercially, and number portability makes the underlying block data wrong anyway. We would rather leave the field out than invent it.

Turning a company name into a contact

Registry and directory data is full of companies with no e-mail at all. Switch on 🔎 Find the company website and e-mail and every row that carries a company name gets a second pass:

1. Candidate domains are built from the name, with the legal suffixes stripped: Acme Holdings LLC becomes acme.com, acme-holdings.com, acme.io and so on.
2. Each candidate is resolved, and the home page is fetched.
3. The name has to be on the page. This is the step that separates a real find from a plausible-looking wrong company, and most tools skip it.
4. The home, contact and about pages are read for a published address, preferring one on the company's own domain.

websiteStatus tells you how much to trust the answer. verified means two or more distinctive words of the company name appear on the page. probable means the name is a single word — a real match, but a namesake is possible, so check before you send. Single-purpose entities that never have a site, such as property vehicles and film companies, come back as skipped_no_website_expected without a single request being made, which keeps the run cheap.

On real registry names this finds a site for roughly four companies in ten, and an address on about four of every ten sites found. Billing follows the result: a search that finds nothing costs nothing.

How much does it cost to verify a list?

Pricing is pay per event: you are charged per contact checked, and nothing else. There is no browser and no proxy in this Actor — a DNS lookup and some in-memory lists — so platform usage is negligible and a list of ten thousand contacts finishes in well under a minute. See the Pricing tab for the current rate.

How to verify a list

1. Paste your addresses into 📧 E-mail addresses, your numbers into 📱 Phone numbers, or both.
2. If one person has both, use 👥 Contacts and pass objects — any extra fields you include, such as your own record id, are copied to the output row.
3. To clean a list you just scraped, paste that run's dataset id into 🔗 Check a dataset from another Actor and name the fields to read.
4. Pick 🎯 Keep only if you want the run to return just the usable rows.
5. Click Start, then export as CSV, JSON or Excel, or read the dataset through the API.

⬇️ Input

{
  "emails": ["owner@acme.com", "info@gmail.com", "hello@mailinator.com"],
  "phones": ["+1 206 235 0243", "(212) 211-5678"],
  "keep": "all"
}

Contacts with both channels

{
  "contacts": [
    { "email": "jane@acme.com", "phone": "+1 415 555 2671", "crmId": "A-1042" },
    { "email": "carlos@ejemplo.mx", "phone": "+52 55 1234 5678" }
  ]
}

crmId is not a field this Actor knows — it is copied through untouched, so the output lines up with your own records.

Cleaning another Actor's output

{
  "inputDatasetId": "aBcD1234efGh5678",
  "emailField": "email, contactEmail",
  "phoneField": "phone, contactPhone",
  "keep": "usable"
}

Point it at the dataset of a lead scraper — for example our US New Business Leads — and it reads the contacts out of that run directly.

⬆️ Output

One row per contact.

{
  "type": "contact",
  "email": "info@gmail.com",
  "emailStatus": "deliverable",
  "emailDomain": "gmail.com",
  "emailProvider": "google",
  "emailIsRole": true,
  "emailIsFree": true,
  "emailIsDisposable": false,
  "emailSuggestion": null,
  "phone": "+1 206 235 0243",
  "phoneStatus": "valid",
  "phoneE164": "+12062350243",
  "phoneFormatted": "(206) 235-0243",
  "phoneType": "geographic",
  "phoneCountry": "US",
  "phoneAreaCode": "206",
  "phoneState": "WA",
  "phoneTimezone": "America/Los_Angeles",
  "contactScore": 75,
  "checkedAt": "2026-09-03T11:59:00.000Z"
}

A row that failed

{
  "email": "sales@acmeholdings-llc.com",
  "emailStatus": "undeliverable",
  "emailReason": "domain_not_found",
  "phone": "(212) 211-5678",
  "phoneStatus": "invalid",
  "phoneReason": "reserved_exchange",
  "contactScore": 0
}

reserved_exchange means the exchange code can never be assigned to a subscriber — 211 is a service code, like 911. Fourteen such codes exist, and they are checked for you.

Use cases

Cold outreach without burning the domain

Bounces above a few percent damage sender reputation for everything you send afterwards. Dropping undeliverable rows and treating risky ones carefully is the cheapest protection there is.

Cleaning bought or scraped lists

Lists sold by the thousand are full of dead domains and throwaway sign-ups. Run the list first, and the emailReason column tells you exactly what you paid for.

Dialers and call centres

phoneStatus keeps impossible numbers out of the queue, phoneType separates switchboards from direct lines, and phoneTimezone stops calls landing before breakfast.

Sorting a lead list

contactScore ranks contacts by how reachable they are: a deliverable company address beats a Gmail one, which beats a role account, which beats a throwaway.

Sign-up and form validation

Reject disposable domains and obvious typos at the point of entry, and suggest the correction with emailSuggestion.

Integrations

Run it from the Apify API, the JavaScript and Python clients, a schedule, or a webhook when another run finishes. It plugs into n8n, Make and Zapier through the Apify connectors, and results can be pushed straight to Google Sheets, Slack or a CRM.

The most useful pattern is chaining: a lead scraper writes its dataset, and this Actor is called with that inputDatasetId to clean it before anything reaches your outreach tool.

🤖 For AI Agents & LLM Apps

Compact reference for agents calling this Actor through the Apify MCP server or the Apify API (lergassy/email-phone-verifier).

Purpose: decides whether contacts are worth using. Given e-mails or phone numbers, it returns a verdict per contact with the evidence behind it. Use it to answer "is this address worth sending to", "which of these numbers can be dialled" and "clean the list I just scraped".

Minimal input:

{ "emails": ["owner@acme.com"], "phones": ["+1 206 235 0243"] }

Output: one flat row per contact — type, email, emailStatus, emailReason, emailDomain, emailProvider, emailIsRole, emailIsFree, emailIsDisposable, emailSuggestion, phone, phoneStatus, phoneReason, phoneE164, phoneFormatted, phoneType, phoneCountry, phoneAreaCode, phoneState, phoneTimezone, phoneTimezoneOptions, contactScore, checkedAt. Fields you pass in on a contact object are copied through unchanged.

Behaviors an agent should know:

• Set findWebsite: true when a row has a company name but no address — it returns website, websiteStatus, foundEmails and foundEmailSource. Trust verified, double-check probable (single-word names can hit a namesake), and read skipped_no_website_expected as "this kind of entity never has a site".
• Four inputs do the same job: emails, phones, contacts (objects, when one person has both) and inputDatasetId (read another run's dataset). They can be combined, and duplicates are removed across all of them.
• emailStatus is about the domain, not the mailbox: cloud platforms block outbound port 25, so no serverless tool can prove a mailbox exists. Treat deliverable as "worth sending", risky as "check first", undeliverable as "do not send", unknown as "retry later" — never as bad.
• emailIsRole, emailIsFree and emailIsDisposable are independent of the verdict. A deliverable address can be all three at once.
• Mobile versus landline is deliberately absent for US numbers; it cannot be derived from free data. Do not infer it.
• phoneTimezone is empty for 21 US area codes that straddle a time-zone boundary; both candidates are in phoneTimezoneOptions.
• To return only usable contacts, set keep rather than filtering rows afterwards — the run is billed per contact checked either way, but the answer is smaller and cleaner.
• Numbers written with a country code other than +1 get a structure check only, reported as phoneReason: "structure_only".
• No proxies, no browser, no third-party API: a run of a few thousand contacts finishes in seconds.

❓ FAQ

Does a "deliverable" verdict guarantee the mailbox exists?

No, and no serverless tool can guarantee that. It means the address is well formed and its domain accepts mail, which is what removes most hard bounces. Mailbox-level confirmation needs an SMTP conversation on a port cloud platforms block, and the big providers answer misleadingly on that port anyway.

Can it tell me if a phone number is a mobile?

No. For US numbers that requires commercially licensed carrier data, and number portability makes the free approximations wrong. We leave the field out rather than invent it.

How fresh are the throwaway-domain and area-code lists?

The disposable-domain list is a community blocklist updated daily; the area-code table comes from the North American numbering administrator's own public database. Both are baked into the Actor and refreshed when it is rebuilt.

Is it legal to verify e-mail addresses?

Checking the technical validity of an address you already hold is routine practice and involves no message being sent. How you obtained the list and what you send to it are governed by your local rules — GDPR, CAN-SPAM and their equivalents. Ask a lawyer about your campaign, not about this check.

Does it send any e-mail?

No. Nothing is sent, and no address leaves the run except as a DNS question about its domain.

Can I use it with the Apify API or an MCP server?

Yes. It runs from the API and the official clients, and it is available to AI agents through the Apify MCP server without any extra setup.

Can it clean the output of another Actor?

Yes — that is what inputDatasetId is for. Give it the dataset id of a previous run and the field names to read.

How does it find a company website?

It builds candidate domains from the name, resolves them, and then requires the company's own name to appear on the page before reporting a match. A guess that cannot prove itself is discarded rather than returned.

What happens to my own fields?

Anything you attach to a contact object is copied to the output row untouched, so the results line up with your records.

Your feedback

Found a case it gets wrong, or need a check that is missing? Open an issue on the Issues tab — every one gets answered, and rules that turn out to be wrong get fixed.

You might also like

Actor	What it does
Airbnb Scraper	Airbnb listings for any place and dates: nightly price, rating, beds, amenities, host
US New Business Leads	Newly registered US businesses with phone and e-mail, already checked
Google Flights Scraper	Live flight prices, layovers and booking options
Agoda Reviews Scraper	Hotel reviews and ratings, including the Booking.com reviews shown on Agoda
Mercado Libre Scraper	Latin American marketplace listings, prices and reviews