Senderward

How to read DMARC aggregate (rua) reports

Updated

Once your DMARC record has rua=mailto:..., receivers that support reporting send you a compressed XML file, usually once a day. Each file says which IP addresses sent mail with your domain in the From: header, how many messages, and whether SPF, DKIM and DMARC passed.

What arrives

One email per reporting organisation per reporting period, with a .xml.gz or .zip attachment. The subject follows a fixed pattern:

Report Domain: example.com Submitter: example.net Report-ID: <1731628800.example.net>

The format was defined in RFC 7489. In May 2026 DMARC was republished as RFC 9989, with aggregate reporting in RFC 9990. The field names below are the ones from RFC 7489's schema, which the reports in your inbox use.

A short example

<feedback>
  <report_metadata>
    <org_name>example.net</org_name>
    <email>[email protected]</email>
    <report_id>1731628800.example.net</report_id>
    <date_range><begin>1731628800</begin><end>1731715199</end></date_range>
  </report_metadata>
  <policy_published>
    <domain>example.com</domain>
    <adkim>r</adkim><aspf>r</aspf>
    <p>none</p><sp>none</sp><pct>100</pct>
  </policy_published>
  <record>
    <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.com</header_from>
      <envelope_from>bounces.example.org</envelope_from>
    </identifiers>
    <auth_results>
      <dkim><domain>example.com</domain><selector>s1</selector><result>pass</result></dkim>
      <spf><domain>bounces.example.org</domain><scope>mfrom</scope><result>pass</result></spf>
    </auth_results>
  </record>
</feedback>

Read it as: example.net received 412 messages from 203.0.113.25 with From: [email protected] between the two timestamps (Unix seconds, UTC). DKIM passed and was aligned, so DMARC passed. SPF passed too, but for bounces.example.org, which is not your domain, so it does not count for DMARC.

The fields

FieldWhat it means
report_metadataWho sent the report (org_name, email), its ID and the period it covers (date_range).
policy_publishedYour DMARC record as the receiver saw it: p, sp, pct and the alignment modes adkim/aspf (r relaxed, s strict). If this is not what you published, the receiver cached an older record or found a different one.
row/source_ipThe IP address that connected to the receiver. One record per IP (and per combination of results).
row/countNumber of messages in this record.
policy_evaluated/dispositionWhat the receiver did: none, quarantine or reject.
policy_evaluated/dkim, /spfThe DMARC result: did DKIM or SPF pass and align with the From: domain. One pass is enough for DMARC to pass.
policy_evaluated/reasonWhy the receiver applied something other than your policy: forwarded, sampled_out, trusted_forwarder, mailing_list, local_policy or other.
identifiers/header_fromThe domain in the From: header. This is the domain DMARC protects.
identifiers/envelope_fromThe return-path (MAIL FROM) domain, which SPF checks.
auth_results/dkimEach DKIM signature: the signing domain (d=), selector and raw result. Not yet about alignment.
auth_results/spfThe SPF check: domain, scope (mfrom or helo) and raw result.

The key distinction: auth_results is the raw result, policy_evaluated is the result after alignment. "SPF pass" in auth_results with fail in policy_evaluated means SPF passed for some other domain, typically the bounce domain of an email service.

Typical surprises

Senders you did not know about

The first week of reports usually shows sources nobody listed: a CRM, the helpdesk, invoicing, a website contact form, a scanner in the office, a partner sending on your behalf. Look up each source_ip (reverse DNS, or the DKIM d= domain in auth_results) to find the service. If it is yours, set up DKIM with your domain at that service. If nobody recognises it and it fails both SPF and DKIM, it is probably spoofing, which is what DMARC is for.

Forwarding

When a recipient forwards your mail (an old address forwarding to Gmail, a university alias), the receiver sees the forwarder's IP. SPF fails, because that IP is not in your SPF record. DKIM usually survives if the message was not changed, so DMARC still passes on DKIM. This is why DKIM matters more than SPF for DMARC. Reports may mark these rows with reason forwarded, but often they just show a pass on DKIM and a fail on SPF from an IP you do not recognise.

Mailing lists

Lists that add a footer or a [list] tag to the subject break the DKIM signature. Then both SPF and DKIM fail and the row looks like spoofing. The source is a list server, and the count is usually small.

"Reject" in the record, "none" in the report

A disposition of none under p=reject does not always mean the policy was ignored. Look at reason: the receiver can override your policy (local_policy), trust a forwarder, or skip the message under pct (sampled_out).

No reports at all

If rua= points to an address on another domain, that domain must authorise it with a TXT record at example.com._report._dmarc.example.net containing v=DMARC1. Without it, receivers drop the reports. Report services set this up for you; your own second domain does not.

How to work through a report

  1. Add up count per source_ip and sort by volume. The top few sources are usually your mail platform and your main email service.
  2. For each source, check policy_evaluated. If DKIM or SPF passes there, that source is fine.
  3. For each failing source with real volume, find the service and fix its DKIM (or SPF, if it can use your domain as the return path).
  4. Leave small, scattered failures from unknown IPs alone. That is spoofing or forwarding, and it is what a stricter policy will block.

When all your own sources pass, you can move the policy up: see Moving DMARC from p=none to p=reject.

Check your domain

Free DMARC check by Lookproof. Opens on lookproof.dev.

Senderward checks your domain's DMARC, SPF, DKIM, MX, MTA-STS and TLS-RPT every day and emails you, with the fix, when something gets worse. $7 per domain per month, 14 days free. See pricing

Sources