DEV Community

Cover image for DMARC: None vs Quarantine vs Reject
Regő Botond Ronyecz
Regő Botond Ronyecz

Posted on

DMARC: None vs Quarantine vs Reject

Teaser only. This is not the full article. Complete policy guide with rollout timing and audit implications: DMARC Policy: None vs Quarantine vs Reject

p=none feels safe. It is monitor-only. Receivers still see spoofing attempts; they just will not act on your policy.

p=quarantine tells mailbox providers to treat failures as suspicious. p=reject tells them to block unauthenticated mail claiming your domain. SOC2, NIS2, and enterprise vendor forms in 2026 increasingly expect quarantine or reject, not endless p=none.

Your policy lives in one TXT record:

dig _dmarc.example.com TXT +short
Enter fullscreen mode Exit fullscreen mode

Look for p=none, p=quarantine, or p=reject. The sp= tag governs subdomains separately.

In the full post on zerohook.org:

  • What each policy does at Gmail, Outlook, and Yahoo
  • When p=none is still acceptable (short answer: not long)
  • pct= and gradual enforcement
  • Compliance mapping (SOC2, NIS2, ISO)

Read the full guide: DMARC Policy: None vs Quarantine vs Reject

Top comments (1)

Collapse
 
sealedmail profile image
SealedMail

Good breakdown of the three policies. Worth adding that p=none is a monitoring mode, not a protection mode. Many domains sit at none indefinitely because no one is reading the aggregate reports. If you're not processing DMARC reports, you won't know when to move to quarantine or what you'll break when you do.