What validation checks

Validation answers the question “could this be a working email address?” without contacting the recipient’s mail server. It usually covers:

  • Syntax. Does the address follow the structure defined in RFC 5322: a local part, an @, and a domain, with only permitted characters? Length limits also apply; RFC 5321 caps the local part at 64 octets. RFC 3696 is a useful guide for applications here, and makes clear that characters such as + and ' are legal in the local part, so overly strict filters end up rejecting real people.
  • Domain existence. Does the domain resolve in DNS at all?
  • Mail routing. Does the domain publish MX records, or at least an A or AAAA record that senders can fall back to? Does it publish a null MX that explicitly refuses mail? See what is an MX record and what is a null MX record.
  • Typo detection. Is gmial.com likely meant to be gmail.com?
  • Domain lists. Is the domain a known disposable or temporary inbox provider?

These checks are fast and cheap because they rely on string rules and DNS lookups. They catch a meaningful share of bad data, especially typos at signup. What they cannot do is tell jane.doe@example.com apart from jane.d0e@example.com when both sit on the same healthy domain.

What verification adds

Verification includes everything above and then asks the recipient’s server directly. The verifier connects to the domain’s mail server, runs the SMTP handshake up to the RCPT TO command, and reads the reply without sending a message (see how SMTP verification works). That reply is what separates “this domain receives mail” from “this specific mailbox exists.”

Verification can also surface conditions validation never sees:

  • A mailbox that has been disabled or is over quota.
  • A domain that accepts every address (a catch-all), which makes the mailbox result inconclusive.
  • Temporary deferrals such as greylisting, which need a retry before a verdict.
  • Policy blocks against the verifier itself, which say nothing about the mailbox.

Because it requires live network connections to many different mail servers, managed sending infrastructure, and retry logic, verification is slower and costs more to run than validation. That is why it is usually the metered layer of a verification service.

Why the distinction matters

A list that has only been validated can still bounce heavily. Validation cannot detect an address with perfect syntax, on a domain with healthy MX records, where the mailbox simply does not exist. That is the classic source of hard bounces, and those bounces hurt sender reputation. Mailbox providers expect bulk senders to manage bounces; Google’s sender guidelines, for example, ask senders to unsubscribe recipients after repeated bounces. If your goal is a low bounce rate, verification is the layer that predicts it.

Validation still matters. It is the right first line of defense on a signup form, because it is instant and can prompt the user to fix a typo before they leave the page. It also filters out addresses that should never reach the slower verification step at all, such as malformed strings or domains that do not exist.

Using both in practice

  1. At signup: validate syntax and domain instantly, suggest corrections for obvious typos, and run a real-time verification before accepting the address.
  2. Before a campaign: verify any list that is new, imported, or has not been mailed in a few months, since addresses decay over time (see what is list decay).
  3. When reviewing results: treat undeliverable as remove, unknown as retry later, and catch-all or risky addresses as a separate segment to send cautiously.

Where RocketVerifier helps

RocketVerifier runs validation and verification together and returns a single status (deliverable, undeliverable, risky, or unknown) with a sub_status that shows which layer caught the issue, such as invalid_syntax, invalid_domain, possible_typo, disposable, or mailbox_not_found. Use the email verification API for real-time checks at signup, or RocketShield to protect forms directly.