Developers tend to say validation for the rules a form can apply on its own: one @ sign, a legal local part, a real-looking domain. That's syntax validation. It proves the address is well-formed and says nothing about whether it exists.
Domain validation goes one step further and looks up the domain's MX record. A domain with no MX and no A record can't receive mail, so every address on it is invalid, however tidy it looks.
Verification is the SMTP step. Your verifier connects on port 25, says EHLO and MAIL FROM, then asks RCPT TO about the address. A 250 means accepted, 550 means no such user, and a 4xx means try again later. The connection closes before DATA, so no email is ever sent.
When you compare tools, ignore the label and ask which steps run. A format-only check passes jonh@company.com when the real mailbox is john@, and only the SMTP reply catches it. Around 28% of a typical B2B list sits on catch-all domains, where the server says 250 to everything, so a good tool reports those as a result of their own.
You can see why pattern rules fall short on is regex enough to validate an email. Try every step on one address with the free email verifier, or follow the full pipeline on how ZapBounce verifies.
Related questions
Should I use email validation or email verification?
Use both. Validation catches typos and dead domains, and verification confirms the mailbox with the receiving server. ZapBounce runs both in one check from $5 for 1,000.
Can I verify an email address without sending an email?
Yes. The SMTP check asks the receiving server about the address with RCPT TO and closes before DATA, the step where a message would be sent. The person on your list receives nothing.
What does email verification check?
Syntax, the domain's MX record, the SMTP reply to RCPT TO, and a catch-all probe with a made-up address at the same domain. Role and disposable flags come from list lookups.