Introduction
The best free DMARC checker is any tool that does a live DNS lookup and validates the record against RFC 9989, the current DMARC standard. For a one-off check that is enough. What matters more is knowing what to look at in the result and what a lookup cannot tell you: who actually sends as your domain, and whether you are ready for p=reject.
This article is published by DMARC AI, a DMARC platform provider for MSPs. The evaluation criteria are designed to be useful when comparing our platform, another vendor, open-source tooling, or an internal build.
This article gives you a checklist for judging any free checker, explains what a single lookup shows and misses, and sets out a check, fix, monitor and enforce workflow, with notes for MSPs who check whole client portfolios. For the protocol itself, start with the DMARC guide.
Key takeaways
- A good free checker queries DNS live, validates against RFC 9989 and shows you the raw record it found.
- It should accept the new
np,psdandttags and flagpct,rfandrias removed, not throw errors on modern records. - A lookup shows configuration. It cannot show who sends as your domain, pass rates or forwarding problems; only aggregate reports over time can.
- Use checkers to verify DNS changes, and aggregate report monitoring to decide when to enforce.
What makes a free DMARC checker good?
A good checker reads the live record from DNS, parses it against the current standard and tells you plainly what receivers will do with it. Most free tools look similar on the surface; the differences are in whether they cache results, which version of the standard they follow and whether they check the related records that make DMARC work.
Use this table to judge any free checker, including ours:
| Criterion | What good looks like | Why it matters |
|---|---|---|
| Live DNS lookup | Queries authoritative or public resolvers at the time you check, not a stored result | A cached answer hides the change you just published |
| RFC 9989 syntax | Validates v=DMARC1 as the first tag and knows the full tag list in Section 4.7, including np, psd and t |
Older parsers report valid modern records as broken |
| Removed tags | Flags pct, rf and ri as historic (Section 9.3) rather than treating them as live settings |
A leftover pct tag gives a false sense of partial enforcement |
| Organizational domain | Follows the DNS tree walk (Section 4.10) when a subdomain has no record of its own | "No record found" on a subdomain can be wrong if the parent policy covers it |
| External report destinations | Checks the _report._dmarc authorisation for rua (and ruf) addresses on another domain, per RFC 9990, Section 4 |
Without it, compliant receivers send no reports |
| SPF lookup count | Counts DNS-querying terms against the limit of 10 and void lookups against the limit of 2, per RFC 7208, Section 4.6.4 | Exceeding either produces permerror, so SPF fails for every message |
| DKIM | Asks for a selector, or says which selectors it guessed | DKIM keys cannot be discovered without the selector |
| Raw record | Shows the exact TXT string returned | You can see quoting, stray spaces and duplicates yourself |
If a checker fails several of these, its verdict is not wrong because it is free; it is wrong because it is out of date or incomplete.
Why does RFC 9989 support matter in a checker?
Because the standard changed in 2026. RFC 9989 obsoletes RFC 7489, adds the np, psd and t tags and lists pct, rf and ri as historic. Section 4.7 also says unknown tags must be ignored, so a record with np=reject stays valid, and receivers still on the old specification simply ignore the tag. A checker that reports np or t as an error is following the old specification.
The reverse matters too. A checker should tell you that a pct percentage no longer stages enforcement, so you do not rely on it. Rollout is now p=none, then p=quarantine, then p=reject, optionally rehearsed with t=y, which asks receivers to apply the policy one level softer than published. The new tags np, psd and t covers each one.
Why should a checker follow the DNS tree walk?
Because a subdomain without its own _dmarc record is usually still covered. Under RFC 9989, Section 4.10, a receiver that finds no record at the author domain walks up the DNS tree, label by label, to find the organizational domain's record and applies its sp or np value. A checker that only queries _dmarc.mail.example.com and stops will tell you the subdomain is unprotected when _dmarc.example.com says p=reject.
What does a one-off DMARC lookup actually show?
A single lookup shows the configuration published in DNS at that moment. That is genuinely useful: many configuration problems, from a missing record to a reporting address that will never receive anything, are visible in DNS. A complete check covers:
- Record present: a TXT record at
_dmarc.example.comthat starts withv=DMARC1. If none exists, see how to fix "no DMARC record found". - Exactly one record: RFC 9989, Section 4.10, says that if multiple DMARC records are returned for one name, they are all discarded.
- Syntax: valid tag values, semicolon separators and no historic tags.
- Policy:
p,spandnpvalues, plust=yif test mode is on. - Reporting addresses:
rua(andruf) asmailto:URIs, with external destinations authorised. - SPF validity: one SPF record, within the lookup and void lookup limits, ending in
-allor~all. - DKIM key for a known selector: a key at
selector._domainkey.example.com, as defined in RFC 6376, Section 3.6.2.1. An emptyp=means the key is revoked (Section 3.6.1). - MX and MTA-STS presence: whether the domain receives mail at all, and whether it publishes an
_mta-stsTXT record.
Here is what a healthy result looks like for a client domain that sends reports to an MSP's reporting domain:
_dmarc.contoso.example. TXT "v=DMARC1; p=quarantine; rua=mailto:contoso@reports.msp.example"
contoso.example._report._dmarc.reports.msp.example. TXT "v=DMARC1"
contoso.example. TXT "v=spf1 include:spf.protection.outlook.com -all"
selector1._domainkey.contoso.example. CNAME selector1-contoso-example._domainkey.contoso.onmicrosoft.com.
And here is a record a good checker should question:
_dmarc.contoso.example. TXT "v=DMARC1; p=quarantine; ri=3600; rf=afrf; rua=mailto:dmarc@reports.othervendor.example"
Three findings: ri and rf are historic and have no effect under RFC 9989, and unless reports.othervendor.example publishes contoso.example._report._dmarc.reports.othervendor.example, receivers will drop the rua address. The record "works", but it reports to nobody.
What can a DMARC checker not tell you?
A checker cannot tell you how mail actually behaves, because DNS only describes intent. Whether real messages pass SPF and DKIM, align with the From domain and survive forwarding is only visible in the aggregate reports receivers send back. A lookup is a photograph of the record; reports are the record of what happened.
| Question | Answered by a lookup? | Where the answer comes from |
|---|---|---|
| Is the record published and valid? | Yes | DNS |
| Will reports be delivered? | Mostly (authorisation record) | DNS, then confirmed by reports arriving |
| Who sends mail as this domain? | No | Aggregate reports |
| What share of mail passes DMARC? | No | Aggregate reports over several weeks |
| Is forwarding or a mailing list breaking SPF or DKIM? | No | Aggregate reports and message headers |
| Is someone spoofing the domain? | No | Aggregate reports showing unknown sources |
Is the domain ready for p=reject? |
No | Aggregate reports showing every legitimate source passing |
RFC 9990 describes aggregate reports as typically covering a single UTC day per receiver. One day of data is a start, not a verdict. The NCSC's advice on monitoring and updating your DNS records is to monitor at p=none for at least two weeks and expect several rounds of fixes before moving to quarantine. DMARC aggregate reports explained shows what the reports contain.
Can a checker tell me why one specific message failed?
Not from DNS alone. For a single message, the evidence is in its headers: the Authentication-Results line shows the SPF, DKIM and DMARC results the receiver recorded. For example:
Authentication-Results: mx.receiver.example;
spf=pass smtp.mailfrom=bounce.mailer.example;
dkim=pass header.d=mailer.example;
dmarc=fail header.from=contoso.example
Both SPF and DKIM pass here, but for mailer.example, not contoso.example, so DMARC fails on alignment. A record checker would show nothing wrong. Paste headers like these into an email header analyzer to read them, and see why your DMARC record is not working for the common causes.
Which free tools does DMARC AI offer?
DMARC AI publishes a set of free tools, each for one job. Use them for one-off checks and DNS changes; they are listed together on the free email security tools page:
- DMARC checker: look up and check a domain's DMARC record.
- SPF analyzer: analyse a domain's SPF record.
- DKIM validator: validate a DKIM key for a domain and selector.
- DNS lookup: query the DNS records behind any of the above.
- Email header analyzer: read the authentication results in a message's headers.
- DMARC generator: build a DMARC record to publish.
Judge them by the same table as any other tool.
What is the right workflow after a DMARC check?
Treat the checker as step one of four. The lookup finds configuration problems; the reports tell you whether the fixes worked and when enforcement is safe:
- Check. Look up DMARC, SPF and DKIM for the domain. Note the policy, the reporting addresses and any errors.
- Fix. Publish a valid record with a working
ruaaddress, correct SPF lookup counts and confirm DKIM for each sending platform. Removepct,rfandri. Re-check after the DNS change. - Monitor reports. Collect aggregate reports at
p=noneuntil you can name every legitimate source and see it passing with alignment. Fix the stragglers, usually third-party senders. - Enforce. Move to
p=quarantine, thenp=reject, in full steps. Optionally publisht=yfirst to rehearse. Keep monitoring afterwards, because new senders appear.
A clean result at step 2 means the DNS is right, not that the mail is: "the checker says it is fine but email is failing" is what stopping there looks like. To confirm the policy receivers actually apply, see how to check your DMARC policy with headers and reports.
How should MSPs use free DMARC checkers?
MSPs get the most value from checkers at the edges of a client relationship: auditing a prospect, onboarding a new client and verifying a change. Between those points, a portfolio needs continuous report monitoring, because client environments change without the MSP being told.
Practical uses:
- Prospect audits. A lookup of a prospect's domains is a fair, factual starting point for a conversation: no record,
p=nonewith norua, or SPF over the lookup limit. Keep it factual and avoid fear marketing; see how MSPs should audit client domains. - Bulk portfolio checks. Checking dozens of client domains one by one in a web form does not scale. Script DNS queries against your domain list, or use a platform that tracks every tenant's records, and recheck after every change window.
- Change verification. After any DNS change under change control, look up the record and keep the result with the change ticket.
- Onboarding baselines. Record each new client's starting configuration, including parked domains and subdomains, before you touch anything.
What a checker does not replace is monitoring. A client adding a new marketing platform, a DNS host migration that drops a TXT record or a spoofing campaign against the domain will not show up in last month's lookup. DMARC monitoring vs managed DMARC compares the ways to cover that gap, and DMARC for MSPs describes how DMARC AI handles aggregate reports across client tenants.
Frequently asked questions
What is the best free DMARC checker?
Any checker that performs a live DNS lookup, validates against RFC 9989, follows the organizational domain tree walk and shows the raw record will do for a one-off check. The bigger differences are in what you check alongside DMARC, such as SPF limits and report authorisation, and in knowing that a lookup cannot measure real mail.
Why does my DMARC checker show a different result from another one?
Usually because one tool cached an older answer or follows the old standard. A recent DNS change may not be visible everywhere until the previous record's TTL expires. A tool built on RFC 7489 may also flag np, psd or t as errors, while an RFC 9989 tool accepts them and flags pct instead.
Can a DMARC checker tell me if I am ready for p=reject?
No. Readiness depends on whether every legitimate sender passes DMARC with alignment, and only aggregate reports collected over several weeks show that. A checker can confirm the record is valid and reports will be delivered. The decision to enforce needs the report data, ideally reviewed per sending source.
Do I need a selector to check DKIM?
Yes. DKIM keys are published at selector._domainkey.example.com, and RFC 6376 provides no way to list selectors from DNS. Find the selector in your mail platform's settings or in the s= value of a DKIM-Signature header on a message you sent, then check that exact name.
Is a free DMARC checker enough for an MSP?
For prospect audits, onboarding baselines and verifying changes, yes. For ongoing protection across a client portfolio, no. Records drift, new senders appear and spoofing attempts happen between checks. Those only show up in aggregate reports, so MSPs pair occasional lookups with continuous report monitoring per client.