The terms “email validation” and “email verification” are often used interchangeably - but they refer to different technical operations with different accuracy profiles, different latency characteristics, and different use cases. Here is a precise breakdown of what each does and when to use it.
Checks an email address for correctness without contacting any mail server. Runs entirely client-side or against DNS records.
Includes everything in validation, plus a live SMTP connection to the recipient’s mail server to test whether the specific mailbox accepts email.
| Check | Validation | Verification | What it catches |
|---|---|---|---|
| RFC 5321 syntax | ✓ | ✓ | Missing @, invalid characters, malformed TLD |
| Domain DNS lookup | ✓ | ✓ | Domains that don’t exist at all |
| MX record existence | ✓ | ✓ | Domains with no mail server configured |
| Disposable domain | ✓ | ✓ | Temp-mail services (Mailinator, Guerrilla Mail) |
| Role-address detection | ✓ | ✓ | info@, admin@, noreply@ - low deliverability |
| SMTP connection | ✗ | ✓ | Mail server is reachable and accepting connections |
| Mailbox existence | ✗ | ✓ | Specific address accepted or rejected by server |
| Catch-all detection | ✗ | ✓ | Domains that accept all addresses (uncertain status) |
| Provider intelligence | ✗ | ✓ | Gmail / M365 / Yahoo specific signals |
Some mail servers accept any RCPT TO probe - they don’t reject unknown addresses at the SMTP level, so they respond as if every address exists. This is called a catch-all or accept-all domain. Verification returns ‘catch-all’ for these (not ‘valid’ or ‘invalid’). In practice, ~30-50% of catch-all addresses are actually deliverable.
Some mail servers limit probe frequency. A 4xx temporary rejection is ambiguous - it may mean ‘not now’ or ‘I don’t know’, not ‘doesn’t exist’. Good verification tools retry with exponential backoff and classify ambiguous responses appropriately.
Even a verified address decays over time. B2B addresses decay at ~2%/month. An address that was ‘valid’ 3 months ago has a ~6% chance of being invalid now. Always verify as close to send-time as possible, not months in advance.
Email validation checks whether an email address is syntactically correct and has a valid domain with MX records - without contacting the mail server. Email verification goes further: it makes an SMTP connection to the recipient’s mail server and tests whether the specific mailbox exists. Validation is cheap and fast (milliseconds). Verification is more accurate but slower. Most modern tools do both as part of a single check.
For signup forms and real-time API calls, email validation (syntax + MX check) is usually sufficient. For outbound email lists before a campaign, email verification (SMTP-level mailbox probe) is essential - it catches the 5-15% of syntactically valid addresses that belong to inactive or deleted mailboxes. Best practice: validate at point of capture, then verify the full list before each campaign.
Yes - email verification is the most direct tool for reducing hard bounce rate before a campaign. A hard bounce rate above 2-3% triggers ISP-level spam filtering and can cause domain blacklisting. Verifying a cold email list typically reduces hard bounces by 85-95% by removing addresses that no longer exist.
Syntax check, MX lookup, disposable detection, SMTP probe, catch-all flagging - all in a single request. $3/1K. 100 free credits.
Ayoub built BounceZero's 5-stage validation pipeline, its dedicated BGP-announced IP infrastructure, and the Patroni HA PostgreSQL cluster behind every verification. Previously built high-volume email delivery infrastructure. Trained at 1337 Benguerir (École 42 network, 2019). Open-source: bgp_analyzer.
Deep-dive guides on how email verification and inbox placement work
272,446-domain census: DMARC gap, provider divide, catch-all rates
10.2M verifications: 12.3% of addresses are dead, and where they hide
826K re-verifications: only 19% of valid addresses survive 90 days
True catch-all is 1.4% - most of what looks catch-all is unprobeable providers
info@ bounces 4.5x more than personal addresses - measured, not guessed
The 3x invalid-rate gap that vanishes when you control for domain size
Explore other topics