DMARC tells receiving mail servers what to do when a message claims to be from your domain but fails authentication, and asks them to send you reports about it. It builds on SPF and DKIM by tying their results to the From address people actually see. Set up carefully, it stops criminals sending email that looks like it came from you, without blocking your own legitimate mail.
What DMARC adds to SPF and DKIM
SPF and DKIM each check a domain, but not necessarily the one in your From header. SPF checks the envelope sender; DKIM checks whatever domain signed the message. A spammer can pass both using their own domain while putting yours in the From line.
DMARC closes that gap with alignment. A message passes DMARC if either:
- SPF passes and the envelope sender domain aligns with the From domain, or
- DKIM passes and the signing (
d=) domain aligns with the From domain.
Only one needs to succeed. That is why DKIM matters so much: it survives forwarding, whereas SPF usually breaks when a message is forwarded.
A basic DMARC record
DMARC is a TXT record at _dmarc on your domain:
_dmarc.example.co.uk. 3600 IN TXT "v=DMARC1; p=none; rua=mailto:dmarc@example.co.uk"
That record asks receivers to take no action on failures (p=none) but to send daily aggregate reports to the given address. It is the right first step for almost every domain, and it meets the minimum required by Gmail and Yahoo for bulk senders since February 2024 and by Microsoft for Outlook.com senders of over 5,000 messages a day since 5 May 2025.
Every DMARC tag explained
| Tag | Meaning | Values and default |
|---|---|---|
v | Version. Must be first. | DMARC1 |
p | Policy for the domain. Required. | none, quarantine (spam folder), reject (block) |
sp | Policy for subdomains. | Same values as p. Defaults to the p value. |
pct | Percentage of failing mail the policy applies to. | 0 to 100. Default 100. |
rua | Where to send aggregate reports. | One or more mailto: addresses, comma-separated |
ruf | Where to send failure (forensic) reports. | mailto: addresses |
adkim | DKIM alignment mode. | r relaxed (default) or s strict |
aspf | SPF alignment mode. | r relaxed (default) or s strict |
fo | When to generate failure reports. | 0 both fail (default), 1 either fails, d DKIM fails, s SPF fails |
A few practical notes:
- pct lets you apply a policy to a sample of failing mail. Support is inconsistent between receivers, and the revised DMARC specification replaces it with a simpler testing flag, so do not build a rollout plan that depends on fine-grained percentages.
- ruf reports contain parts of real messages. Many large mailbox providers do not send them at all, and those that do may include personal data, so only use an address you are happy to receive that at.
- External report addresses: if
ruapoints at a different domain (for example a reporting service), that domain must publish an authorisation record, or receivers will not send reports there. Reporting services set this up for you.
Relaxed vs strict alignment
With relaxed alignment, the domains only need to share the same organisational domain. A DKIM signature from d=mail.example.co.uk aligns with From: hello@example.co.uk. So does an envelope sender of bounce@news.example.co.uk.
With strict alignment, they must match exactly. mail.example.co.uk would no longer align with example.co.uk.
Relaxed is the default and the right choice for almost everyone. It lets email platforms use a subdomain for bounces and signing while still counting towards your DMARC. Strict alignment adds little real protection for most businesses and breaks easily.
Domains, DNS & emaildijitul DNSRegister and transfer .uk and .com domains, with Cloudflare DNS management and business email. SPF, DKIM and DMARC set up properly.Fix your DNS with dijitulThe staged rollout: none, quarantine, reject
Jumping straight to p=reject is how businesses discover, the hard way, that their invoicing system or website contact form was never authenticated. Do it in stages.
Stage 1: monitor with p=none
v=DMARC1; p=none; rua=mailto:dmarc@example.co.uk
Collect reports for several weeks, long enough to cover monthly jobs like invoice runs and newsletters. Use them to build a list of every service sending as your domain and fix each one so it passes aligned SPF or DKIM. Usually that means setting up DKIM signing with your own domain and a custom return-path on each platform.
Stage 2: quarantine
v=DMARC1; p=quarantine; rua=mailto:dmarc@example.co.uk
Once your reports show legitimate sources passing, move to quarantine. Failing mail goes to the spam folder rather than disappearing, so if you have missed something, it is recoverable. Keep watching reports and listen for colleagues saying customers have not received things.
Stage 3: reject
v=DMARC1; p=reject; rua=mailto:dmarc@example.co.uk
When quarantine has run cleanly for a few weeks, move to reject. Failing mail is now refused outright, which is what stops spoofing. Keep the rua address: reports remain your early warning when someone adds a new tool without telling you.
Reading aggregate reports
Aggregate reports arrive as compressed XML attachments, typically one per receiving provider per day. Each contains rows like this:
<row>
<source_ip>203.0.113.25</source_ip>
<count>412</count>
<policy_evaluated>
<disposition>none</disposition>
<dkim>pass</dkim>
<spf>fail</spf>
</policy_evaluated>
</row>
<identifiers><header_from>example.co.uk</header_from></identifiers>
That says 412 messages from one IP claimed to be from example.co.uk; DKIM passed with alignment, SPF did not, so the messages passed DMARC overall. Raw XML is tedious, so most people use a reporting tool to group sources by sender. When reviewing, sort every source into one of three buckets:
- Legitimate and passing: nothing to do.
- Legitimate and failing: a service you use that needs DKIM or SPF fixing before you enforce. Look up the IP's owner to identify it.
- Not yours: spoofing or spam using your domain. This is exactly what
p=rejectwill block.
Forwarded mail often shows up as SPF failing from an IP you do not recognise, with DKIM passing. That is normal and does not need fixing.
Subdomains
A subdomain with no DMARC record of its own inherits the policy of your main domain, using sp= if it is set and p= otherwise. That has two uses:
- Protect unused subdomains: once your main domain is at reject, attackers cannot simply spoof
invoices.example.co.ukinstead. - Stage a rollout differently: you might run
p=reject; sp=quarantinewhile you sort out a marketing subdomain.
A subdomain can publish its own _dmarc record, which overrides the parent. That is handy when a subdomain is dedicated to one sending platform. Domains you own but never send from should publish v=DMARC1; p=reject along with an SPF record of v=spf1 -all.
Common mistakes
- Publishing the record on the root domain instead of at
_dmarc. - Having two DMARC records, which means receivers ignore both.
- Moving to reject without ever reading the reports.
- Pointing
ruaat a mailbox nobody checks, then missing a broken service for months. - Assuming SPF alone will carry you. Forwarding breaks it; aligned DKIM is what keeps forwarded mail passing.
For the full list of what the big mailbox providers now expect, see Gmail and Yahoo sender requirements and Outlook sender requirements.
Frequently asked questions
What DMARC policy do Gmail and Yahoo require?
Since February 2024, bulk senders to Gmail and Yahoo need a DMARC record with at least p=none, and messages must align with SPF or DKIM. Microsoft applies a similar minimum for high-volume senders to Outlook.com from May 2025.
Is p=none enough?
It meets the current minimum requirements and gives you reports, but it does not stop anyone spoofing your domain. Treat it as a starting point and move towards quarantine and reject once your reports are clean.
What is the difference between rua and ruf?
rua receives aggregate reports: daily XML summaries of who sent mail using your domain and whether it passed. ruf requests forensic reports on individual failures, which many large providers do not send for privacy reasons.
Does a DMARC policy on my main domain cover subdomains?
Yes. A subdomain without its own DMARC record falls back to the organisational domain's record, using the sp= policy if set and the p= policy otherwise.
How long should I stay at p=none?
Long enough to see a full cycle of your sending: usually several weeks, covering monthly invoices, newsletters and anything seasonal. Move on when every legitimate source in your reports passes aligned SPF or DKIM.