schedule 15-min read

Why Am I Getting Emails From My Own Address? Spoofing Explained and Fixed

DMARC AI editorial team · Last updated

Getting emails from your own address? How to tell spoofing from a hacked mailbox in the headers, why it gets delivered, and how MSPs stop it for good.

01

Why am I getting emails from my own address?

You are getting emails from your own address because someone forged it. SMTP (Simple Mail Transfer Protocol) lets any server put any address in the From line, so a criminal can send mail "from" you without touching your mailbox. It reaches your inbox when your domain's DMARC policy, or your own mail filtering, does not block it.

The classic version is a sextortion email: "I hacked your account, pay in Bitcoin." Your own address as sender makes the claim look real, but the mailbox is almost always fine. This guide shows an MSP (managed service provider) helpdesk how to prove that from the header, why the mail got through, and which settings stop it. For background on spoofing types, see what is email spoofing. For DMARC itself, start at the DMARC pillar guide.

02

Key takeaways

  • An email that shows your own address as the sender is almost always spoofed, not sent from your account. The header proves it within a minute.
  • Look for dmarc=fail with header.from set to your domain, a compauth=fail reason=6xx or reason=010 in Microsoft 365, a Return-Path that is not yours, and no copy in Sent Items.
  • It gets delivered because the domain sits at p=none or has no DMARC record, because anonymous Direct Send is open, because a gateway breaks authentication, or because someone allow-listed the domain.
  • The fix is DMARC enforcement (p=quarantine, then p=reject), Microsoft 365 spoof protection with "honor DMARC" on, Reject Direct Send where safe, and no allow-list entries for your own domain.
  • Tell the client plainly: do not pay, do not reply, change any password the email quotes, and let us check the header.
03

Is my account hacked, or is the email spoofed?

In most cases it is spoofed. A genuine compromise leaves traces that a forged From line does not: a copy in Sent Items, a sign-in you do not recognise, new inbox rules, and a message that passes SPF, DKIM and DMARC. A spoofed message fails DMARC for your domain and never touched your mailbox.

Check these five points, in this order:

  1. Sent Items. Mail sent from the real account normally leaves a copy. No copy is a strong hint, though an attacker could delete it.
  2. Authentication-Results. Read the header added by your own mail system. Spoofed mail shows dmarc=fail against your domain.
  3. Return-Path. Spoofed mail usually carries an envelope sender on another domain, or a forged one on yours that fails SPF.
  4. Message trace. In Microsoft 365, trace the message in the Exchange admin center. Spoofed mail arrives inbound from an external IP, not from your tenant's outbound path.
  5. Sign-in logs. In the Microsoft Entra admin center, review the user's sign-in logs for unfamiliar locations or failed MFA (multi-factor authentication) prompts, and check for new forwarding or inbox rules.

If all five point to spoofing, move on to why it was delivered. If any point to real access, treat it as a compromise: reset the password, revoke sessions, review MFA methods and audit mailbox rules, as covered in business email compromise and DMARC.

04

How do you read the header of a self-sent spoof?

Open the full header and find the topmost Authentication-Results line written by your own mail platform. Compare header.from with smtp.mailfrom and any header.d. If neither SPF nor DKIM passed for your domain, DMARC fails and the From address is forged. In Microsoft 365, compauth and SFTY confirm the verdict.

Here is a typical sextortion message received by a Microsoft 365 tenant that owns example.co.uk, with DMARC at p=none:

Return-Path: <info@mail.example-relay.net>
From: <jane.smith@example.co.uk>
To: <jane.smith@example.co.uk>
Subject: I have access to your device
Authentication-Results: spf=pass (sender IP is 203.0.113.77)
 smtp.mailfrom=mail.example-relay.net; dkim=none (message not signed)
 header.d=none; dmarc=fail action=none header.from=example.co.uk;
 compauth=fail reason=601
X-Forefront-Antispam-Report: CIP:203.0.113.77;...;SFV:SPM;...;CAT:SPOOF;SFTY:9.11;...

Read it line by line, using Microsoft's anti-spam message header reference:

  • Return-Path and smtp.mailfrom are on an unrelated domain, so SPF passed for that domain, not yours. dkim=none: the message was not signed.
  • dmarc=fail action=none header.from=example.co.uk: nothing authenticated your domain, so DMARC failed. action=none because the policy is p=none, which asks receivers to take no action.
  • compauth=fail reason=601: Microsoft lists 601 as implicit failure where the sending domain is one of your accepted domains, which it calls self-to-self or intra-org spoofing.
  • SFTY:9.11: Microsoft's anti-spoofing protection page says .11 indicates intra-org spoofing. CAT:SPOOF is the spoofing category.

Once the domain moves to an enforcing policy, the same attack shows compauth=fail reason=010. Microsoft defines 010 as DMARC failed with p=quarantine or p=reject where the sending domain is one of your accepted domains.

Paste any header into the email header analyzer to parse it. For the Received chain, see the MSP guide to reading email headers.

What does a genuinely sent message look like?

Mail that really left the user's mailbox in Microsoft 365 shows aligned authentication, for example dmarc=pass with header.from=example.co.uk and compauth=pass reason=100 or reason=115. Microsoft defines 115 as mail sent from a Microsoft 365 organisation with an accepted From domain. If you see that on a threatening email, take the compromise path seriously.

05

Why does spoofed mail from my own domain get delivered?

Spoofed self-sent mail gets through when nothing tells the receiving system to block it. The four usual causes are a weak or missing DMARC policy, anonymous Direct Send into Microsoft 365, a gateway or connector that hides the original sender, and allow-list rules that trust your own domain.

Is the domain at p=none or missing DMARC?

This is the most common cause. With p=none or no record, DMARC asks receivers to do nothing with failing mail. Microsoft's implicit checks may still send it to Junk, but a well-written lure can reach the inbox. RFC 9989, Section 4.7, defines p=reject as the owner saying failures are "a clear indication that the use of the domain name is not valid."

Check the record with the DMARC checker. If the client is stuck at monitoring, what p=none means explains why it does not protect anyone.

Is Microsoft 365 Direct Send open?

Direct Send lets a device or application connect anonymously to the tenant's MX endpoint on port 25 and deliver mail to internal recipients using the tenant's own domain. Scanners use it. Attackers can use the same path to drop forged "internal" mail into mailboxes, because Direct Send requires no login. Our guide to Direct Send, printers and DMARC covers the legitimate uses and alternatives.

Is a gateway or connector breaking the checks?

When a secure email gateway sits in front of Microsoft 365, Microsoft sees the gateway's IP instead of the original sender's. SPF evaluates the wrong address and the spoof can look like trusted relay traffic. Without Enhanced Filtering for Connectors, or a gateway that enforces DMARC itself, forged mail slips through. Also lock the tenant so only the gateway can deliver, or attackers can skip it entirely. Microsoft's header reference lists compauth=none reason=905 for DMARC not enforced "due to complex routing".

See why DMARC fails with a gateway in front of Microsoft 365 and Enhanced Filtering for Connectors.

Has someone allow-listed the client's own domain?

Allow-lists are a frequent, self-inflicted cause. A transport rule that skips spam filtering for @example.co.uk, an allowed domain in an anti-spam policy, or a user's Safe Senders entry for their own address all invite spoofing. In the header, look for SFV:SKN (bypassed by a mail flow rule), SFV:SKA (allowed sender or domain list) or SFV:SFE (user Safe Senders list).

Microsoft is explicit. Its safe sender lists guidance says: "Don't use domains you own (also known as accepted domains) or popular domains (for example, microsoft.com) as conditions in mail flow rules." It also warns never to configure mail flow rules with only the sender domain as the condition to skip spam filtering.

06

How do you triage a "from my own address" ticket?

Match what the user sees and what the header shows to the likely cause, then act on the cause rather than the symptom. The table covers the patterns an MSP service desk meets most. Header evidence always wins over the user's description.

Symptom What it means Action
Sextortion or "I hacked you" mail, dmarc=fail, no copy in Sent Items Spoofed From; mailbox not involved Reassure user, report the email, move DMARC towards p=reject
Quoted password the user recognises Password from an old data breach, not from this mailbox Change that password everywhere it was reused; enable MFA
compauth=fail reason=601, domain at p=none Intra-org spoof caught only by implicit checks Plan enforcement; confirm spoof intelligence action is Junk or quarantine
compauth=fail reason=010 but message in inbox DMARC enforcement is set but overridden locally Check anti-phishing "honor DMARC" setting and allow entries
SFV:SKN, SFV:SKA or SFV:SFE in the header A rule or allow-list bypassed filtering Remove own-domain allow entries; add an authentication condition
Arrived anonymously on port 25, no connector match Direct Send abuse Move devices off Direct Send, then enable Reject Direct Send
Sending IP is your gateway, reason=905 or SPF on gateway IP Gateway or connector hides the true source Enable Enhanced Filtering or enforce DMARC at the gateway
dmarc=pass, copy in Sent Items, odd sign-ins Real account compromise Reset password, revoke sessions, audit rules and MFA
07

How do you stop spoofed email from your own domain?

Stopping it takes two layers. Publish an enforcing DMARC policy so every receiver, including your own, knows forged mail is invalid. Then make sure your own tenant honours that policy and has no back doors: Direct Send, gateway bypasses or allow-lists. Either layer alone leaves gaps.

Step 1: Move DMARC to quarantine, then reject

Inventory every sending service, align SPF and DKIM for each, and watch aggregate reports at p=none until legitimate mail passes. Then move to p=quarantine and p=reject. RFC 9989 removed the old percentage tag, so there is no partial ramp. To soften a step, add t=y, which asks receivers to apply the policy one level lower while you test.

_dmarc.example.co.uk.  3600  IN  TXT  "v=DMARC1; p=reject; rua=mailto:dmarc@example.co.uk"

The checkpoints are in moving from monitoring to enforcement safely. Generate a valid record with the DMARC record generator.

Step 2: Set Microsoft 365 spoof protection to honour DMARC

In the Microsoft Defender portal, check the anti-phishing policy's spoof settings. Microsoft's anti-phishing policy reference describes them:

  • Enable spoof intelligence. Microsoft recommends leaving it on. It is on by default in the default policy and in new custom policies.
  • Honor DMARC record policy when the message is detected as spoof. With this on, explicit failures use the action you pick for p=quarantine (quarantine or Junk) and p=reject (quarantine or reject).
  • If the message is detected as spoof by spoof intelligence. The action for implicit failures, such as a domain at p=none.

With "honor DMARC" off, Microsoft ignores the sender's p= value and applies its own spoof action. Policy detail is in DMARC in Microsoft Defender for Office 365.

Step 3: Close Direct Send with Reject Direct Send

Exchange Online has an organisation setting that refuses anonymous mail from your own domains. Per the Set-OrganizationConfig reference, with $true Exchange Online "rejects anonymous messages sent from your own domain to your organization's mailboxes" when they match no inbound connector and the MAIL FROM domain is an accepted domain.

Get-OrganizationConfig | Format-List RejectDirectSend
Set-OrganizationConfig -RejectDirectSend $true

Microsoft introduced the setting as a preview in 2025; confirm its current status in the Message Center before rollout. Move every scanner and application off Direct Send first, or they will break. The check uses the envelope sender, not the visible From, so it complements DMARC rather than replacing it.

Step 4: Remove allow-list entries for your own domain

Search transport rules, anti-spam allowed domains, the Tenant Allow/Block List and users' Safe Senders for the client's accepted domains, and remove them. If a rule must trust a partner, Microsoft recommends pairing the sender domain with a header condition: Authentication-Results contains dmarc=pass or dmarc=bestguesspass. For internal senders, Microsoft states that mail flow rules cannot designate them as safe at all.

What about Google Workspace?

Google Workspace has a matching control in its advanced phishing and malware protection settings. "Protect against inbound emails spoofing your domain" targets messages "not authenticated with either SPF or DKIM", and its default action is to keep the email in the inbox with a warning. For a client receiving self-sent lures, change the action to spam or quarantine, alongside DMARC enforcement.

08

What should you tell the client?

Keep the message calm and specific. Users are often frightened or embarrassed by sextortion content, so lead with the facts: the email is forged, the account was not used, and here is what to do. The NCSC's sextortion guidance advises not to pay.

A script your service desk can adapt:

Subject: The email that appeared to come from your own address

Thank you for reporting this. We have checked the message header. The email was sent by an outside server that put your address in the From line. It did not come from your mailbox, and we found no sign that anyone has logged in to your account.

These emails are sent in bulk to thousands of people. Please do not reply, click any link or pay anything. If the email quoted a password, it most likely came from an old data breach on another website. If you still use that password anywhere, change it there and switch on multi-factor authentication.

We are tightening the settings on your domain so that forged emails like this are blocked rather than delivered. You do not need to do anything else. If you receive another one, forward it to us as an attachment.

In the UK, staff can also forward suspicious emails to the NCSC's Suspicious Email Reporting Service at report@phishing.gov.uk, as GOV.UK's reporting page explains.

09

How should an MSP handle this across a client portfolio?

Treat every "email from my own address" ticket as a sign that the client's domain is not enforced or the tenant has a bypass. Fix the cause once and the tickets stop. Across a portfolio, a few checks catch most exposure.

  1. List domains below p=reject. Each one can be spoofed to its own users and to everyone else.
  2. Audit every tenant for own-domain allow entries in transport rules, anti-spam policies and the Tenant Allow/Block List.
  3. Record Direct Send use per tenant and schedule the move to SMTP AUTH or a connector, then Reject Direct Send.
  4. Standardise the anti-phishing baseline, with spoof intelligence on and "honor DMARC" set to quarantine or reject.
  5. Watch DMARC aggregate reports for spoofing volume against each domain. A spike usually precedes the next wave of tickets.

The DMARC for MSPs page explains how DMARC AI processes aggregate reports across client domains and classifies sending sources, separating spoofing from legitimate senders still to fix.

10

FAQ

Can someone send an email from my address without my password?

Yes. SMTP does not check the From address, so anyone can type yours into it. That is spoofing, and it needs no access to your account. Whether the forged message is delivered depends on your domain's DMARC policy and the receiver's filtering. At p=reject, receivers that honour DMARC refuse it.

Why does the spoof email include my real password?

The password almost certainly comes from an old data breach on another website, not from your mailbox. Criminals buy breach lists and paste the password in to make the threat look credible. The NCSC advises changing it straight away if you still use it anywhere, and turning on multi-factor authentication.

Will DMARC p=reject stop these emails completely?

It stops exact-domain spoofing at receivers that honour DMARC, including your own Microsoft 365 tenant when "honor DMARC" is set correctly. It does not stop lookalike domains, display-name tricks or mail from a genuinely compromised account. It also cannot help if a transport rule or allow-list bypasses filtering, so remove those too.

Should I block my own domain to stop the spam?

No. Blocking your own domain would also block legitimate internal mail and replies. Fix the cause instead: move DMARC to p=quarantine or p=reject, turn on Microsoft's "honor DMARC" setting, close Direct Send, and remove any allow-list entries for your domain. Those steps block forgeries without touching genuine mail.

11

Final thoughts

An email from your own address is usually a forged From line and a missing control, not a breached mailbox. The header answers the security question quickly: dmarc=fail, an intra-org compauth code, a foreign Return-Path and no Sent Items copy. Then fix the delivery path: enforce DMARC, honour it in Microsoft 365, close Direct Send, and stop trusting your own domain in allow-lists. For warnings on mail your clients send, see why Outlook and Gmail say a message may be spoofed.

Related articles

Related tools

Ready to Implement?

Get authenticated mail moving in minutes: start free, book a guided demo, or talk to the team about your stack.