Email spoofing: what it is, and how to stop it

Last updated: November 21, 2026

Email spoofing is when someone sends mail with your domain in the From address the reader sees, even though that mail did not come from you. SPF and DKIM alone do not stop this, because they can pass for a completely different domain. DMARC stops it only when the authenticated domain aligns with that From address, and only for your exact domain and its subdomains, once the policy is at p=quarantine or p=reject.

If you want to see who is already sending as your domain, start a free DMARC trial. If mail using your domain is already reaching inboxes, use the knowledge base article What to do when your email domain is spoofed for the practical next steps.

Start a free DMARC trial · Book a meeting

The From address the reader sees is not the envelope

Internet email has two different "from" values, and they do not have to match.

The envelope is the SMTP transaction, described in RFC 5321. It includes the MAIL FROM address (sometimes called the bounce address or RFC5321.MailFrom), which recipients almost never see. The header is the message the mail app shows, described in RFC 5322. That is where the From address lives, and that is the identity people trust.

A sender can use one domain in MAIL FROM and a different domain in the From header. SMTP, on its own, does not stop that. Microsoft's explanation of email authentication puts it plainly: SMTP "makes no effort to validate that the message sender is who they claim to be."

There is a third layer people also trust: the display name. A message can appear as Finance Team <billing@sender.example> and still have nothing to do with example.com. The name is just text. DMARC does not read it.

Why an SPF or DKIM pass can still be a spoof

SPF (RFC 7208) answers a narrow question: is this sending host allowed to use this domain in MAIL FROM or HELO? It does not authenticate the From header the reader sees. Section 11.2 of the RFC is explicit: a sender can pass SPF for their own domain "and yet the message can easily list other identities in its header," including the From field.

DKIM (RFC 6376) answers a different narrow question: did this signing domain take some responsibility for the message, and was the signed content altered in transit? The signature's d= tag is the signing domain, and the RFC says that signing identity "is not required to match an address in any particular header field."

So this pattern is valid under SPF and DKIM, and it is still a spoof of example.com:

  • The reader sees support@example.com in From.
  • MAIL FROM is on sender.example, and SPF passes for sender.example.
  • DKIM is signed with d=sender.example, and the signature verifies.

Both checks passed. Neither one asked whether the sender was allowed to use example.com in From. Microsoft documents the same attack in How email authentication works in Microsoft 365: an attacker can register a domain, pass SPF or DKIM for it, and still put someone else's domain in the From address.

That is why a green SPF or DKIM result, on its own, is not proof that the From address is yours.

DMARC alignment is what ties authentication to the From domain

DMARC (RFC 9989) adds the missing test, Identifier Alignment. A message passes DMARC when at least one authenticated domain, from SPF, DKIM, or both, matches the domain in the RFC5322.From address. A DKIM signature from an unrelated domain is not enough. An SPF pass on a bounce or ESP domain is not enough, unless that domain aligns with From.

By default, the match is relaxed: the organizational domains must be the same, not necessarily the exact hostnames. For example, a validated DKIM signature with d=example.com aligns, in relaxed mode, with a From address of alerts@news.example.com. In strict mode (adkim=s or aspf=s), only an exact hostname match counts. If you omit those tags, the default is relaxed. Use strict alignment only after you have inventoried From domains, DKIM d= values, and MailFrom domains.

The policy itself is a DNS TXT record at _dmarc.example.com:

v=DMARC1; p=none; rua=mailto:dmarc-feedback@example.com

p=none asks receivers to send reports and otherwise take no special action. p=quarantine asks them to treat failing mail as suspicious, typically by junking it. p=reject asks them to refuse failing mail. You can set a separate subdomain policy with sp=. If sp= is absent, p= applies to subdomains as well. So a record on example.com can cover news.example.com. It cannot cover a different registration that merely looks similar.

For the alignment tags themselves, see Identifier alignment for SPF and DKIM and Relaxed vs strict alignment.

p=none watches spoofing. It does not stop it.

A record at p=none is useful. Receivers that support DMARC reporting will send aggregate data (the rua address) about mail that claims to be from your domain. You can see authorized senders, misaligned third parties, and obvious spoofing. What you cannot do is ask those receivers to keep the spoofed mail out of the inbox. The policy you published told them not to.

Mailbox-provider sender rules can hide this distinction. Google's email sender guidelines require bulk senders to set up SPF, DKIM, and DMARC, and they say the DMARC policy "can be set to none." Yahoo's sender requirements ask bulk senders for a valid DMARC policy with at least p=none, and that "DMARC must pass." Microsoft's requirements for high-volume senders to Outlook.com, in place since May 5, 2025, require SPF, DKIM, and DMARC for domains sending 5,000 or more messages a day to Outlook.com consumer addresses, and they accept at least p=none with alignment to SPF or DKIM.

Those are rules for sending mail to those inboxes, not a claim that p=none stops spoofing. Yahoo's FAQ is more direct if spoofing is the problem you have: "If you have a spoofing problem, you should be implementing a DMARC enforcement policy (p=quarantine or p=reject) regardless."

Do not stay at p=none once legitimate mail is aligned and you can read the reports. Monitoring is the first step, not the fix.

What DMARC will not catch: cousin domains and display names

RFC 9989 is honest about its own reach. Section 2.2 says DMARC "can only be used to combat specific forms of exact-domain spoofing directly," and that it "does not address the use of visually similar domain names or abuse of the RFC5322.From human-readable display name." Display-name attacks are also called out of scope in Section 11.4.

In practice, two common tricks still work after you reach p=reject on example.com:

  • A lookalike domain. Mail from a domain that looks like yours, but is a different registration, is judged under that other domain's DMARC policy, if it has one. Your record never applies.
  • A borrowed display name. Mail from billing@sender.example can still show a name that resembles your brand. DMARC looks at the address domain, not the name next to it.

Subdomains are a narrower case, and DMARC can cover them if you configure it. If example.com is at p=reject and you omit sp=, that reject policy applies to subdomains. If you set sp=none, you have asked receivers not to enforce the policy on those subdomains. BIMI Group's implementation guide wants enforcement on the organizational domain and the subdomains (p= and sp= both at quarantine or reject) before a logo is considered. Leaving sp=none is a hole in subdomain spoofing protection, and it also blocks BIMI.

One more limit, because it is easy to over-claim: mail sent from a compromised mailbox on your own domain can pass SPF, DKIM, and DMARC. The domain really did send it. Authentication will not flag that as spoofing.

This page is not a promise that DMARC will take down cousin domains. If that is the attack you are dealing with, say so when you book a meeting.

Brand spoofing is a wider problem than exact-domain email

Brand spoofing is the broader pattern: someone pretends to be your organization in order to steal credentials, take a payment, or damage trust. Email is a common channel, but it is not the only one. The same campaign can mix a cousin domain, a familiar display name, a fake login page, and a lookalike social profile.

Exact-domain email spoofing is the part DMARC is designed to stop. If receivers honor a p=reject policy, a message that claims to be from example.com (or a covered subdomain) and fails alignment should not reach the inbox as ordinary mail. That is a real, specific win, not a complete brand-protection program. Lookalike domains and display-name abuse need a different response than a DMARC record. When a campaign is already in circulation, the incident steps are in What to do when your email domain is spoofed.

How to tell if your domain is being spoofed

Publish a DMARC record with a rua address, even if the policy is still p=none, and read the aggregate reports. Those reports list sources that sent mail using your domain in From, and whether SPF, DKIM, and alignment passed.

Sources you recognize, with alignment you can fix if it is broken, are your legitimate mail. Sources you do not recognize, with no alignment to your domain, are the spoofing, or at least the unauthenticated use, of that exact From domain. The knowledge base article Legitimate mail vs spoofing in reports is the how-to for reading those rows. This page will not repeat that walkthrough.

A mailbox screenshot or a single bounced message is a weaker signal. Headers can be forged in the parts that are not authenticated, and a customer complaint does not tell you whether the From domain was yours, a cousin domain, or only a display name. If a campaign is already live, start with the incident article, then use the reports to see the scale.

Start a free DMARC trial if you want those reports in one place, instead of parsing raw XML yourself.

The order that actually stops exact-domain spoofing

Each step depends on the one before it.

1. Authenticate the mail you really send. Publish one valid SPF record that names every service that sends as your domain. Sign the mail with DKIM. Third-party tools (billing, marketing, support desks) need this too, or they will fail later, when you enforce. The sequence is in The right order: SPF, DKIM, DMARC, then BIMI. Setup help is on our DMARC setup page.

2. Publish DMARC at p=none with reporting. You need the inventory before you ask receivers to quarantine or reject. Google also recommends DMARC reports so you can monitor mail "sent from your domain, or appears to have been sent from your domain." Yahoo strongly recommends a rua tag during initial setup.

3. Align every legitimate sender, then move the policy. When the reports show that your real mail passes DMARC, tighten the policy to p=quarantine, then to p=reject. The operational walkthrough is in Transitioning from DMARC monitoring to enforcement. Do not skip the reports and jump to reject: forwarders and mailing lists can break alignment for genuine mail, and those cases need a decision first.

A typical enforcement record looks like this:

v=DMARC1; p=reject; rua=mailto:dmarc-feedback@example.com

If you still have a pct tag from an older rollout, remove it before you treat the domain as fully enforced. RFC 7489 allowed pct as a staged rollout. The current DMARC specification, RFC 9989 (May 2026), deprecates that tag. Leave it out of new records.

4. Only after enforcement, consider BIMI. A logo in the inbox is a visible trust layer. It is not the control that stops spoofing, and it is not the starting point.

Start a free DMARC trial · Book a meeting

After enforcement, BIMI can show a verified logo

BIMI (Brand Indicators for Message Identification) lets supporting mailbox providers show your logo next to qualifying mail. BIMI Group's sender FAQ states the order clearly: "Authenticate: SPF/DKIM aligned to the visible From: domain, with DMARC at quarantine/reject." Most BIMI-supporting providers "require enforcement before considering your logo for display," and they do not accept p=none or a pct value below 100.

Gmail follows that bar, and it also requires a mark certificate. That certificate is a VMC if the logo is a registered trademark, or a CMC if you are using a mark route that does not need a completed trademark registration. SSL.com's VMC requirements, as one certificate authority's public checklist, include DMARC at p=quarantine or p=reject before you apply. For a comparison of the two certificate types, see Is a VMC still worth it now that CMCs exist?.

BIMI does not change whether a spoofed message is accepted. DMARC policy does that. The logo, where a provider shows it, helps the reader recognize the mail that already passed. Coverage is uneven. Yahoo may display a logo for bulk mail with a BIMI record, enforcement, and sufficient reputation, and it does not require a VMC or CMC. Official Microsoft documentation still says emails sent from Exchange Online do not yet support BIMI. Do not count on an Outlook logo just because you ordered a certificate.

If you want help with the logo step after enforcement, start at Get your BIMI.

Our recommendation

Start with visibility, then enforcement, then the logo. Start a free DMARC trial, identify every service that sends as your domain, fix alignment, and move to p=quarantine and p=reject. That is what stops exact-domain From spoofing.

Do not buy a certificate first, and do not treat Google, Yahoo, or Microsoft bulk-sender minimums as the finish line. p=none keeps you in compliance with those sending rules. It does not keep spoofed mail out of other people's inboxes.

If a campaign is already using your exact domain, follow the incident article and tighten the policy as soon as the reports say your legitimate mail will survive it. If the campaign is only a lookalike domain or a display name, do not expect the DMARC record on example.com to take it down.

Start a free DMARC trial · See plans · Book a meeting

Frequently asked questions

What is email spoofing?

Email spoofing is sending a message with a From address, or sometimes only a display name, that makes the mail look as if it came from someone else. The form DMARC can stop is exact-domain spoofing: use of your actual From domain without alignment.

Does SPF stop email spoofing?

No. SPF authorizes a sending host to use a domain in MAIL FROM or HELO. It does not authenticate the From header. RFC 7208 Section 11.2 describes how a message can pass SPF and still list other identities in the header.

Does DKIM stop email spoofing?

No. DKIM proves that a signing domain took responsibility for the signed content. That signing domain does not have to match the From address.

Does DMARC stop all spoofing?

No. It stops exact-domain From spoofing, for the domain that published the policy and, depending on sp=, its subdomains, when receivers honor p=quarantine or p=reject. It does not stop cousin domains or display-name abuse. RFC 9989 Section 2.2 states both limits.

What is brand spoofing?

Brand spoofing is impersonating an organization across any channel, including email, lookalike websites, and social accounts. DMARC addresses exact-domain email spoofing, one slice of that problem, not a complete brand-protection program.

Is a DMARC policy of p=none enough to stop spoofing?

No. p=none collects reports. It does not ask receivers to junk or reject failing mail. Yahoo's sender FAQ advises an enforcement policy if you have a spoofing problem.

Do Google, Yahoo, and Microsoft require DMARC enforcement?

Not as their bulk-sender minimum. Google, Yahoo, and Outlook.com all accept a DMARC policy of p=none for high-volume sending, with SPF, DKIM, and alignment so that DMARC can pass. Enforcement is what stops spoofing, and it is also what BIMI requires.

How do I tell if my domain is being spoofed?

Publish DMARC with a rua address and read the aggregate reports. Unrecognized sources that fail alignment are the usual sign of exact-domain spoofing. Legitimate mail vs spoofing in reports explains how to read the rows. For a live campaign, use What to do when your email domain is spoofed.

When should I add BIMI, a VMC, or a CMC?

After DMARC is at p=quarantine or p=reject on the organizational domain and the From domain, with no pct below 100, and after legitimate mail is passing. The certificate does not replace enforcement.

Will DMARC stop a lookalike domain that is spoofing my brand?

No. A lookalike is a different domain. Publish DMARC on the domains you own, and handle lookalikes as a separate issue.

Sources