Skip to content

Email Header Analyzer: Trace and Check an Email

The headers are read in your browser and never stored. The only request is to our own server, to look up the network and country of the public IP addresses in the Received lines. Private and reserved addresses are not looked up.

How do I get the headers?
  • Gmail: Open the message, click the three dots next to Reply, choose "Show original", then "Copy to clipboard".
  • Outlook (web): Open the message, three dots, View, "View message details" (or "View message source").
  • Outlook (desktop): Double-click the message, File, Properties, and copy the "Internet headers" box.
  • Apple Mail: Select the message, then View, Message, "All Headers" (or "Raw Source").
  • Thunderbird: Open the message, then View, Headers, "All" - or press Ctrl+U for the full source.

Headers can hold your own address and the names of other recipients. Paste only what you are comfortable analysing; nothing leaves this page except the IP lookups described above.

Learn more

Guide and answers to common questions about Email Header Analyzer

What email headers are

Every email carries two parts: the message you read and a block of technical lines above it, the headers. Your mail program hides most of them. They record who sent the message, which servers handled it on the way, when each step happened and what the receiving server concluded about the sender.

Headers are plain text, one per line, in the form Name: value. A long value continues on the next line when that line starts with a space or a tab. Some names appear many times, above all Received, because every server that handles the message adds its own.

How to get the headers

Each mail program has a menu entry for the raw message. Copy the whole block, not only the lines you find interesting.

  • Gmail: open the message, click the three dots next to Reply and choose Show original.
  • Outlook on the web: open the message, three dots, View, View message details.
  • Outlook for Windows: open the message in its own window, then File, Properties, and copy the Internet headers box.
  • Apple Mail: View, Message, All Headers, or Raw Source.
  • Thunderbird: View, Headers, All, or press Ctrl+U for the source.

How to read the Received lines

A Received line says: this server received the message from that server, using this protocol, at this time. Servers add new lines on top, so the first Received line in the file is the last step and the bottom one is the first. The tool turns them around and shows the path from the start to your inbox.

A typical line reads from mail.example.org (mail.example.org [203.0.113.7]) by mx.example.net with ESMTPS id abc123; Tue, 20 Oct 2026 09:15:05 +0000. The part after from is what the sending machine called itself plus, in brackets, the address the receiving server actually saw. The part after by is the receiver. The date after the semicolon is the receiver's clock.

The delay between two steps is the difference between their time stamps. Small numbers are normal. A delay of minutes often means greylisting, a queue or a busy filter. If a later step is stamped earlier than the one before it, the two servers' clocks disagree. The tool shows that as a clock difference, not as a travel time below zero, and then does not trust the total.

Finding the sender's IP address

The tool looks for the oldest step whose connecting address is public and does not belong to the recipient's own mail system. That is where the message first reached a server we can see, and it is the best candidate for the sender's origin. We then look up the network and country of that address in our own database, on our own server.

The result has real limits. Big webmail and app services such as Gmail, Outlook.com, Yahoo and iCloud do not copy the sender's device address into the message, so the first public address is the provider's server. In that case the tool says the sender's own address is not in these headers and does not guess. A VPN, a proxy or a company network shows its own address. And any Received line below the hop added by your own provider was written by some other machine and could be invented.

An address in X-Originating-IP or X-Sender-IP is shown separately, as a claim made by the sender's software.

SPF, DKIM and DMARC in plain words

Three checks help a receiving server decide whether a message really comes from the domain it claims. All three work with records the domain owner publishes in DNS.

  • SPF: the domain publishes a list of servers allowed to send its mail. The receiver compares the connecting address with that list. SPF looks at the hidden envelope sender, shown as Return-Path, not at the From line you see.
  • DKIM: the sending system signs parts of the message with a private key, and the domain publishes the matching public key. A valid signature shows the signed parts were not changed on the way and that the signing domain, named in d=, allowed it.
  • DMARC: the domain tells receivers what to do when a message fails, and requires that SPF or DKIM passes for the same domain as the visible From address. This is the check that ties the other two to what you read.

Results you will see

The receiving server writes its verdict in an Authentication-Results header, for example spf=pass, dkim=pass and dmarc=pass. Pass is good. Fail means the check was run and the message did not meet it. Softfail is a weak failure, the domain says the sender is probably not allowed but asks receivers to accept it with caution. Neutral and none mean the domain gave no answer. Temperror and permerror mean the check could not be completed.

Received-SPF is an older header with the SPF result alone. DKIM-Signature shows who signed, with which selector and algorithm. ARC headers appear when a message was forwarded or passed through a mailing list, so that the next receiver can see what the first one concluded.

How phishing shows up in headers

Forged mail leaves traces, though none of them alone proves anything. Look for these combinations.

  • DMARC fail, or SPF or DKIM fail for the domain in the From line. The message did not come from where it says.
  • A display name that names a bank, a shop or a payment service while the address is on an unrelated domain, or a display name that contains an email address of its own.
  • A Reply-To on a different domain than From, so that your answer goes to the attacker. Support desks and newsletters also do this, so ask yourself whether it fits.
  • A look-alike domain such as paypa1 or a long name with the brand at the front. These domains often pass SPF, DKIM and DMARC, because the attacker owns them. A pass proves the domain is genuine, not that the sender is honest.
  • A Message-ID, mailer program or origin country that does not fit the claimed sender, or a Date far from the time stamps of the first servers.
  • No authentication results at all from your own provider.

What this tool can and cannot tell you

The tool reads the text you paste and explains what it says. It does not look up DNS records and does not check any signature again, so it reports the receiving server's result and nothing more. Only that server's own Authentication-Results line deserves trust. A header added further down, or in the message of a forwarder, can be made up by anyone.

Mailing lists, forwarding and security gateways change a message on its way and can turn a correct message into a DKIM or SPF failure. A mismatch of domains is a reason to look closer, not a verdict. The verdict at the top is a summary of the points listed below it: bad for authentication failures and sender-name mismatches, warn for weak or partial results, good when DMARC passed, unknown when no results are in the headers.

Be careful with the conclusion in the other direction too. A clean result does not make a link or an attachment safe. If a message asks for money, a password or a code, contact the sender by another route that you already know.

Privacy

The headers are parsed by JavaScript on this page, in your browser. They are not uploaded and not stored, and we keep no copy. The one network request is a lookup of the public IP addresses from the Received lines at our own server, which answers from a database it holds itself, so no outside service sees them. Private and reserved addresses, such as 10.x.x.x, 192.168.x.x and documentation ranges, are not looked up.

Headers can still contain your address, other recipients and internal server names. Share an analysis only after you have checked what it shows.

Frequently asked questions

Is it safe to paste my email headers here?

The text stays in your browser and is not sent or stored. Only the public IP addresses from the Received lines are sent to our own server for a lookup. Headers can still show your address and other recipients, so avoid sharing screenshots of the result.

Can this show who really sent an email?

Sometimes it shows the network the message first came from. It cannot name a person. Webmail services hide the sender's device address, and VPNs and proxies show their own. When the sender's address is not in the headers, the tool says so.

Why does it say the sender's IP is not in the headers?

Services such as Gmail and Outlook.com accept the message over a login or an app and put their own server address in the first Received line. The sender's address stays inside the provider.

What does DMARC pass mean?

The receiving server found that SPF or DKIM passed for the same domain as the visible From address, as the domain owner requires. It means the domain is genuine. It does not mean that the content is safe.

The result says SPF softfail. Is the mail fake?

Not necessarily. Softfail means the domain thinks the sending server is probably not allowed but wants receivers to be careful. Forwarded mail and misconfigured domains often get it. Look at DKIM and DMARC and at the From address before you judge.

Why is the Reply-To different from the From address?

Newsletters, ticket systems and mailing lists use a separate reply address. Scammers use it to receive your answer. Check whether the domain makes sense for the sender you expected.

What is a clock difference?

Each server stamps the message with its own clock. When a later step shows an earlier time than the step before, the clocks differ. The message did not travel back in time, so the tool shows a clock difference and does not calculate a delay.

Can headers be forged?

Yes. A sender can add fake Received lines and fake authentication lines at the bottom. The lines added by your own receiving server, at the top, are the ones to trust.

Does the tool check DNS records or signatures itself?

No. It does not do any DNS lookup and does not validate signatures again. It reports what the receiving server wrote into the headers.

I have no Authentication-Results header. What now?

Some small servers and old clients do not add one. Make sure you pasted the complete header block. If it is truly missing, the headers cannot show whether the sender is genuine, and the tool says unknown.