To set up DMARC, make sure every service that sends mail as your domain passes SPF or DKIM, then publish a _dmarc TXT record that starts at p=none with a rua reporting address. Read the aggregate reports for about two weeks, fix the senders that fail, and tighten the policy to p=quarantine and then p=reject. In Google Workspace and Microsoft 365 that is three admin screens and a handful of DNS records, and most small businesses finish the setup in an afternoon and the rollout in four to six weeks.
Why this matters for your website
Without DMARC, anyone can put your domain in the From address of an email. That is how fake invoices, fake password resets and fake supplier bank-change requests reach your customers while your brand takes the blame.
DMARC, defined in RFC 7489, does two things. It tells receiving servers what to do with mail that fails authentication, and it sends you daily reports about who is sending mail as you. Google also lists a DMARC record among its requirements for bulk senders in the Gmail sender guidelines.
If you want the underlying concepts first, read SPF, DKIM and DMARC explained. Otherwise, start at step one.
How to set up DMARC in seven steps
Work in order. Each step is safe on its own, and nothing blocks mail until step five.
Step 1: List every service that sends mail as you
Write down every system that puts your domain in the From address. Typical list for a small business:
- Your mailbox provider (Google Workspace or Microsoft 365)
- Newsletter or marketing tool
- CRM and sales sequencing tool
- Helpdesk or ticketing system
- Your website: WordPress or WooCommerce transactional mail, often via an SMTP plugin
- Invoicing, booking or e-signature tools
Then check your SPF record. You should have exactly one TXT record on the root domain, and it should include a mechanism for each sender, for example: v=spf1 include:_spf.google.com include:servers.mcsv.net ~all. One SPF record only; two records make SPF fail entirely.
Step 2: Enable DKIM in Google Workspace and Microsoft 365
DKIM signs your messages so they survive forwarding, which SPF often does not. Neither platform signs with your own domain until you switch it on.
Google Workspace: in the Admin console go to Apps > Google Workspace > Gmail > Authenticate email, select the domain, generate a 2048-bit key, publish the supplied TXT record at google._domainkey, then return and choose Start authentication. Google documents the process in its DKIM help page.
Microsoft 365: in the Defender portal go to Email & collaboration > Policies & rules > Threat policies > Email authentication settings > DKIM, select the domain, publish the two CNAME records for selector1._domainkey and selector2._domainkey, then set Sign messages for this domain with DKIM signatures to Enabled. Microsoft's DKIM configuration guide lists the record values.
Enable DKIM in your newsletter tool, CRM and helpdesk as well. Each one gives you its own CNAME or TXT records to publish.
Step 3: Publish a DMARC record with p=none
Create a TXT record at the host _dmarc on your domain. Monitoring only, no effect on delivery:
_dmarc.example.com TXT "v=DMARC1; p=none; rua=mailto:[email protected]; fo=1"p=none means take no action. rua is where daily aggregate reports are sent; use a dedicated mailbox or a report viewer's address. fo=1 asks for failure detail when supported. Leave out ruf unless you have a reason to receive forensic copies.
Step 4: Read the aggregate reports for two weeks
Aggregate reports arrive as compressed XML, which is unreadable by hand. Upload them to any free DMARC report viewer or use the reporting dashboard included with many mailbox and DNS providers.
You are looking for three things: sending IPs you recognise that fail alignment, sending IPs you do not recognise, and the volume behind each. Fix your own senders first by adding the missing SPF include or enabling DKIM. Ignore low-volume noise from forwarders and mailing lists.
Step 5: Move to p=quarantine with pct
When your known senders pass, apply the policy to a slice of traffic:
"v=DMARC1; p=quarantine; pct=25; rua=mailto:[email protected]"Failing mail in that 25 per cent goes to spam rather than the inbox. Watch reports and your helpdesk for a week, then raise pct to 50, then 100. If something breaks, lower pct or return to p=none in minutes.
Step 6: Move to p=reject
Once 100 per cent quarantine has run quietly for a week or two, switch to rejection. This is the setting that actually stops spoofed mail being delivered.
"v=DMARC1; p=reject; rua=mailto:[email protected]"Step 7: Add the subdomain policy
Attackers often spoof subdomains that were never used for mail, such as billing.example.com. The sp tag covers them:
"v=DMARC1; p=reject; sp=reject; adkim=s; rua=mailto:[email protected]"adkim=s requires strict DKIM alignment, so the signing domain must match exactly. Add it only if your senders sign with your own domain rather than a subdomain. For a parked domain that sends nothing, publish v=DMARC1; p=reject; sp=reject; plus an SPF record of v=spf1 -all.
Third-party senders and the SPF ten-lookup limit
SPF allows a maximum of ten DNS lookups per evaluation. Every include, a, mx, ptr, exists and redirect counts, and nested includes inside a vendor's record count too. Exceed ten and the result is permerror, which most receivers treat as a failure, so DMARC fails with it.
Four fixes, in the order you should try them:
- Remove includes for tools you no longer use; old CRMs and form plugins linger for years
- Use the vendor's narrower include where one is offered
- Send bulk mail from a subdomain such as
news.example.comwith its own SPF record - Rely on DKIM for that sender, since DMARC passes if either SPF or DKIM aligns
Avoid permanently flattening includes into raw IP lists. Vendors change IPs without telling you, and flattened records break quietly.
How to check your own site
After each step, verify the record actually resolves rather than trusting the control panel. The free check your DMARC record free tool reads your live SPF, DKIM and DMARC records, counts SPF lookups and flags a missing or malformed _dmarc entry. No account is needed.
Scantinel (scantinel.ai) is an automated website security assessment platform that runs OSINT, network and OWASP Top 10 web vulnerability tests against ownership-verified assets and delivers actionable, business-owner-legible reports. The Essential scan includes a full DNS records audit and your SPF, DKIM and DMARC posture alongside TLS configuration, security headers and cookie flags, so email authentication sits in the same report as the rest of your domain hygiene. It is an automated assessment, not a penetration test. Scan options are listed on the Scantinel pricing page.
Key takeaways
- Inventory your senders before touching DNS; an unknown sender is the main cause of broken rollouts
- Enable DKIM in both Google Workspace and Microsoft 365; it survives forwarding when SPF does not
- Always start at
p=nonewith aruaaddress, and never jump straight top=reject - Use
pctto ramp quarantine from 25 to 100 per cent, and keep SPF under ten DNS lookups - Finish the job with
sp=rejectso unused subdomains cannot be spoofed
Frequently asked questions
How long should I stay on p=none?
Two to four weeks is enough for most small businesses, because aggregate reports arrive daily and low-volume senders show up within a fortnight. Stay longer only if you have seasonal senders, such as an annual mailing, that have not run yet. Move on once every sender you recognise is passing SPF or DKIM with alignment.
Will DMARC stop my newsletter from being delivered?
Only if the newsletter tool is not authenticated for your domain. Enable the vendor's DKIM signing and add its SPF include, confirm passes in your reports, and delivery is unaffected when you tighten the policy. That is precisely why the rollout uses p=none and then pct rather than an immediate p=reject.
Do I need DMARC for a domain that does not send email?
Yes, and it is the easiest case. Publish v=spf1 -all and a DMARC record of v=DMARC1; p=reject; sp=reject; so receivers reject anything claiming to come from that domain. Parked and redirect domains are common spoofing targets precisely because owners forget them.
Published November 2025 by the Scantinel Security Research Team. Verify details against primary sources such as OWASP, the NVD and WPScan before you act.
Comments
No comments yet.