Compare roles before drawing conclusions
From, Reply-To, and Return-Path: Three Different Roles
A message can name one address as its author, another for replies, and a third for delivery errors. That arrangement can support an ordinary support desk or notification service, but it can also deserve a closer look when a conversation changes unexpectedly. The useful question is not whether every address matches. It is which role each address declares, what changed, and what independent information you need before acting.
Open the Email Header Analyzer with raw delivery headers you are authorized to inspect. Paste them into the raw-header textarea and choose Analyze headers. Load example fills a public example only; it does not start analysis. Clear removes the working input and results. The separate address outputs describe the supplied text, not verified people, mailbox owners, or approved reply destinations.
Give each address the job it actually declares
The From field identifies the declared author or authors of the message. Separate its display name from its mailbox: a familiar business name can appear beside an unfamiliar address. For example, Accounts Team <billing@vendor.example> contains both a label and an address, and the label does not authenticate the mailbox. Read the complete domain rather than recognizing a word in the display name. In your investigation note, describe this as the declared From address. Calling it the verified sender would add a conclusion that neither the field itself nor this local parser can establish. Compare the mailbox with a previously established contact record, not merely another familiar-looking message in the same conversation. A repeated display name can conceal a changed underlying address. If a provider uses a shared service domain, ask about that specific relationship rather than assuming that the domain belongs to the business named on screen.
Reply-To gives the address or addresses suggested for replies. A company might send notifications from one mailbox and direct questions to a staffed support queue. That difference alone is not evidence of spoofing. If Reply-To is absent, the analyzer leaves it absent; it does not manufacture a value from From. Mail clients commonly use From as a default reply destination when Reply-To is missing, but client behavior is a separate matter from the existence of a header. Before sending a sensitive response, inspect the actual recipient in your mail application rather than assuming the analyzer has tested the reply action.
Return-Path records the delivery reverse-path used for error handling, not the human conversation destination. An automated service may use a dedicated bounce address while keeping the business address in From. The empty form <> can be appropriate for delivery notifications; it is not a hidden mailbox that the analyzer should fill in. An absent field and an explicitly empty reverse-path are different observations. Do not send an ordinary reply to Return-Path just because it looks more technical or authoritative. Its role explains why an address can differ, but does not prove that a particular message was legitimately sent.
Read complete fields rather than a familiar name
Obtain the full header block from the received message, not the compact sender line above its body. Gmail provides Show original in the message menu, while current Outlook variants expose message details or Internet headers. The Gmail headers guide and Outlook headers guide explain those starting points. Keep the exact source privately. A screenshot that shows only a display name cannot answer whether Reply-To was present, whether Return-Path was empty, or whether a second occurrence of a field was omitted from the image.
Address syntax can contain quoted names, encoded words, and folded continuation lines. A comma inside a quoted display name is not a reliable place to split an address by hand. The analyzer uses its bundled parser to present supported address declarations, while preserving the distinction between fields. Read the mailbox alongside the decoded name, and consult the original field if the presentation is unexpected. Several declared addresses do not automatically mean several authenticated people. Likewise, a separate Sender field can describe a transmitting agent without replacing the declared author. Do not collapse all these roles into one informal sender label. Keep folded lines attached to their field when collecting the source; flattening a screenshot into loosely transcribed lines can change the material another person is asked to interpret. If copying adds a page title or quotation markers, reacquire the raw block. Do not present a repaired transcription as an untouched original.
Keep duplicate fields and their source order when collecting evidence. If two Reply-To fields appear, do not quietly delete one and report the survivor as the original message. Repeated singleton fields may be a source-quality or message-format issue requiring administrator review; retaining an occurrence is not endorsing its validity. The same care applies when an address is absent or unsupported syntax prevents a neat interpretation. Record what the source contains and what remains unclear. A tidy three-address story is less useful than an accurate note that a field was missing, duplicated, or not understood, especially when another person must reproduce your reading.
Connect authentication declarations to the right identity
Authentication labels are not interchangeable endorsements of all three addresses. A reported SPF result generally concerns a transport identity, while a DKIM result concerns a signing domain. A reported DMARC result concerns domain alignment with the visible From domain, not ownership of a Reply-To mailbox. Use the SPF, DKIM, and DMARC results guide to keep those questions separate. If a header reports a pass for a service domain, write down the associated property rather than promoting that pass into approval of a different address or a request to send money.
Read every Authentication-Results field separately, with its declared authserv-id. The analyzer keeps all occurrences in source order; it does not merge contradictory results, vote between servers, or select a trusted authority automatically. The identifier names the service claiming to have assessed the message. Its familiar spelling or position near the top is not proof that the field came from your receiving provider. Establishing a trust boundary requires knowledge outside the pasted text. A useful note says which service declared which result for which property, leaving the administrator to compare that declaration with known receiving infrastructure.
A DKIM-Signature can declare d= for the signing domain, s= for the selector, and a= for the algorithm. Showing those tags is not cryptographic verification. The tool makes no API calls or DNS queries and cannot retrieve a key or validate a signature. ARC declarations are also separate from direct authentication, not a universal pass for the author or reply destination. Preserve a mismatch between a reported result and a signature tag as a question, not a solved fraud case. The analyzer cannot verify identity, determine spam status, or guarantee safety or deliverability from these declarations.
Keep the roles separate
Inspect the declared addresses
Read From, Reply-To, and Return-Path independently, then connect each question to the evidence actually present.
Investigate a changed reply destination proportionately
Imagine a supplier update declaring From: billing@vendor.example, Reply-To: cases@helpdesk.example, and a service-managed Return-Path. These public example domains illustrate roles, not an approved configuration. Begin with the event that prompted concern: perhaps previous invoices had no Reply-To, or the new message asks for bank details. State that concrete change. A different support domain might reflect an authorized ticketing workflow, but the headers do not demonstrate that business relationship. Equally, matching all three addresses would not prove the message harmless. The requested action and the established relationship still matter.
Confirm unexpected sensitive instructions through a contact route you already know independently, such as a previously verified supplier record. Do not use the new Reply-To address to ask whether that same address is legitimate. Do not follow a phone number or link from the questionable message as the sole confirmation route. The analyzer renders supplied values as text rather than active input-derived links, but it cannot make your later correspondence trustworthy. For a low-risk scheduling question, the appropriate response may differ from a bank-account change. Keep the decision tied to the actual consequence instead of treating every address difference as the same incident. If the action is time-sensitive, tell the responsible colleague that confirmation is pending without forwarding the instructions as approved. Record which established contact route was used and what was confirmed, without publishing private contact details. Urgency does not supply the missing evidence about the new address or authorize sending confidential information there.
If you compare an earlier message, use its own full headers and retain each original separately. Avoid comparing a received message with a colleague’s manually forwarded body, which may show only a typed From label. Note whether a different sending service, list, alias, or forwarding arrangement could explain the change; then ask the authorized mail administrator to check that hypothesis. The forwarded email authentication guide explains why another delivery path can change the available evidence. The tool provides one current analysis, not a saved comparison or a history of approved supplier addresses.
Use route and source details without inventing evidence
Received fields can add context when a forwarding service or gateway appears relevant. The analyzer displays them in reversed source order, commonly described as oldest-to-newest, rather than sorting timestamps into a plausible journey. Adjacent signed timestamp differences compare declarations in that displayed sequence. A negative interval can indicate clock or order inconsistency; it is not negative delivery time. If a date cannot be interpreted, the interval remains unknown. The Received headers and delays guide develops that distinction. Neither the route labels nor their placement independently prove where the message traveled.
The first blank line ends the header block. Text after that separator is ignored as body content, even if it contains a line beginning with From or Reply-To. A forwarded conversation printed in the body is therefore not another delivery-header set for this parser. Start with raw headers from the correct message, and do not remove a legitimate separator to force quoted body text into the analysis. The tool does not connect to a mailbox, read attachments, render the body, or open an .eml file. Copy only the header text through your authorized mail-client workflow.
Input is bounded to 128 KiB of UTF-8 source, 2,000 lines, 500 fields, 128 Received fields, and 32 fields across Authentication-Results and ARC-Authentication-Results combined. A value longer than 8,192 UTF-16 units is also rejected; retained folding whitespace and line breaks are part of that value. Oversized or structurally malformed source is rejected as a whole, including an orphan continuation or a non-header line before the separator. Rejection is not evidence that an address is fraudulent. Reacquire a clean header block rather than deleting inconvenient fields until it passes. Unsupported authentication-clause syntax can remain labeled as unsupported instead of becoming an invented result. Keep the distinction between a rejected source, an unparsed declaration, and a reported authentication failure clear in any handoff.
Finish with a narrow finding and a private original
A practical handoff might say that Reply-To names a different domain, the source contains two distinct authentication reports, and the relationship between the reply service and supplier is unconfirmed. That gives the next person specific work without declaring identity theft from a difference alone. Keep the original header block private, and share only what an authorized recipient needs. Follow the safe header-sharing guide to make a manually redacted copy outside this tool and record every changed field. The analyzer does not automatically redact data or create a shareable report. Separate observations from questions in that note: field present is an observation; service authorized by supplier is an unresolved relationship unless independently confirmed. Name who can answer that question rather than silently assigning the analyzer authority to decide it. This prevents later readers from mistaking a working hypothesis for a completed check.
Analysis runs locally in the isolated standalone PHP panel, with an opaque iframe origin, no-store, and connect-src 'none'. An explicit analysis starts a terminable local worker with a six-second deadline. Clear, a hidden page, pagehide, pageshow, and freeze wipe the input and derived result DOM and stop work. Switching away is not a way to save unfinished evidence. These measures do not defeat browser extensions, a compromised device, or a hostile parent replacing the interface, and they do not promise secure memory erasure. Keep your private source in your own approved workflow.
There is no account connection, project, storage, history, clipboard control, export, or PDF in this analyzer. It does not send email, change a mailbox rule, or certify an address for future correspondence. Clear the session once the question is answered, then make any operational decision through the appropriate verified channel. Return to the Email Header Analyzer overview for another header-reading question, or browse Tools for a separate task. The useful outcome is a precise distinction between declared roles, reported checks, and unresolved business context, not an unsupported safe-sender verdict.
Questions about From, Reply-To, Return-Path
Do different From, Reply-To, and Return-Path addresses prove spoofing?
No. They have different roles, and support desks, notification services, and forwarding arrangements can use different addresses legitimately. An unexpected difference can justify independent confirmation, especially before a sensitive action, but the analyzer cannot verify that relationship or establish sender identity.
What does a missing Reply-To mean in the result?
It means no Reply-To field was supplied in the analyzed header block. The analyzer does not invent one from From. Your mail client may use From when composing a reply, so inspect the actual destination there before sending sensitive information.
Does a DKIM or DMARC pass approve the reply address?
No. The tool presents reported authentication declarations, not independently verified checks. A signing domain or aligned From-domain result is not a mailbox-ownership check for Reply-To and does not guarantee that the message, its instructions, or its reply destination is safe.
Continue with another email-header guide
Interpret a supplied header in the Email Header Analyzer or choose a related task below. Explore more practical utilities in all free tools.
A precise reading, not an identity verdict
Review one header block deliberately
Analyze authorized raw headers locally, preserve uncertainty, and clear the working session when finished.