· 4 min read

Understanding Your DMARC Reports: What Those Daily Emails Actually Mean

DMARC aggregate reports are the most underrated deliverability diagnostic you have. Here is how to read them without wading through raw XML.

If you publish a DMARC record — even a permissive p=none one — the major mailbox providers start sending you daily XML reports about every message that claims to come from your domain. Most businesses never read them. The XML is dense, ugly and about as welcoming as a stack of invoices.

Which is a shame, because buried inside those reports is the clearest picture you will ever get of who is sending email as you, whether your legitimate mail is passing authentication, and where your reputation is quietly bleeding.

What is in an aggregate report

A DMARC aggregate report (the technical name is RUA, “reporting aggregate”) lists, per message or grouped batches:

  • The sending IP address — where the message actually originated
  • The results of SPF and DKIM checks — pass or fail, for each
  • The alignment results — whether the authenticated domains match your From domain
  • The From domain and envelope details — what the recipient saw
  • Message counts — how much mail this source sent

Notice what is not in there: message content, subject lines or recipient addresses. Aggregate reports are metadata only, which is also why they can be shared with a third-party processor without exposing anything sensitive.

The four findings that matter

Read enough of these reports and the same patterns repeat.

1. Legitimate mail that is failing

Your own CRM, invoicing system or marketing platform sends mail that fails SPF or DKIM — usually because someone connected a tool without adding its DNS records. Every failed message from a legitimate source is a deliverability problem you can actually fix.

2. Senders you forgot about

An old ESP you migrated away from, a department running a mailing list on a dusty server, a SaaS tool sending notifications under your domain. DMARC reports surface all of it. Before you tighten your policy, every one of these needs to be identified and either fixed or decommissioned.

3. Pure spoofing

Sources you do not recognise, sending volume you cannot explain. This is impersonation — and the reason DMARC exists. At p=none you merely observe it; at p=quarantine or p=reject, receiving providers act on it.

4. Alignment failures

The subtle one. A message can pass SPF and pass DKIM and still fail DMARC, because neither authenticated domain aligns with the domain in the From address. This happens constantly with forwarded mail, third-party senders and careless setups, and it is invisible without reading reports.

The workflow, start to finish

The intended sequence looks like this:

  1. Publish DMARC at p=none and collect reports
  2. Read the reports; identify every legitimate sender
  3. Fix authentication for each one until clean reports come back
  4. Move to p=quarantine, watch again
  5. Move to p=reject once nothing legitimate is caught

Businesses that skip the reading step and jump straight to p=reject are the ones who lock their own invoices out of customer inboxes. The reports exist precisely so you do not have to guess.

Why nobody actually does this

Because the raw files arrive as gzipped XML attachments with names like example.com!101293840182!5739.json.gz, and opening one in a text editor rewards you with a wall of namespace declarations. The industry’s answer is usually to pay a vendor to parse them for you.

That works, but you can also just parse them yourself — and for independent verification, it is worth knowing what your data looks like before it reaches anyone else’s dashboard.

Read your own reports

Paste an aggregate report file into the tool below and it parses the XML directly in your browser — plain-English verdicts per source, pass/fail breakdowns, and alignment results. Nothing is uploaded; the file never leaves your machine.

View your DMARC reports →

One honest caveat: a single day’s report is a snapshot. The value is in the trend — the same senders appearing week after week, failures that never get fixed. Bring the reports back once a month and read the patterns, not just the incidents.

Share:
Back to Blog

Related Posts

View All Posts »