Private digital intelligence bureau Lawful OSINT · Cyber monitoring · Due diligence
Analyst safely reviewing a suspicious message, sender details and timeline

Messaging and fraud guide · 11 min

Analyse a suspicious message without clicking, replying or disclosing more information.

A display name, logo or urgent tone proves nothing. Safe review separates the apparent identity, actual destination, requested action and expected context.

Published Updated Publisher and method: IIS

At a glance

Four rules for a useful review.

  • Do not interact to test it.
  • Compare the sender through a known channel.
  • Read the actual link destination.
  • Assess the request before the wording.

Step-by-step method

A reproducible path from question to conclusion.

  1. Laptop, phone and notebooks on a desk — AI-generated photograph
    1

    Keep the message at a distance

    Do not open the attachment or use supplied contact details. For an account or payment, use the official app or independently located website.

  2. 2

    Preserve context

    Retain the complete message, time, channel, apparent address or number and email headers where available. A screenshot often hides useful data.

  3. 3

    Compare claimed identity

    Check the domain after @, lookalike characters, the Reply-To address and consistency with previous exchanges. A display name can be chosen freely.

  4. 4

    Check the displayed address without opening it

    If your app lets you read or copy the address without opening or previewing it, check the visible domain. A short link or QR code can hide the final destination. If checking requires interaction, return through the official channel.

  5. A person comparing a printed document and references in a library — AI-generated photograph
    5

    Assess the request

    Classify what is sought: secret, login, payment, installation, document, changed bank details or urgent action. Higher consequence requires stronger independent confirmation.

  6. 6

    Confirm and report

    Call using an already known number, use the official channel and forward the item only through the approved internal or competent reporting route.

One signal is not a verdict

A typo, unusual phrase or new domain may prompt caution, but none proves fraud alone. A flawlessly written message can still be malicious.

Look for convergence between identity mismatch, sensitive request, manufactured urgency, unexpected destination and failure to confirm independently.

Return through a trusted channel

Useful verification does not reply within the suspect channel. Restart through an official app, known number, established contact or internal procedure.

At work, record time, targeted account and action taken so other recipients can be identified without spreading the message unnecessarily.

Report the message and act on what you shared

In the UK, the NCSC provides separate routes for suspicious emails and texts. In the US, follow the FTC’s reporting guidance. For a work account or device, use the internal reporting procedure. Keep the original message and record any actions you took.

If you entered a password, contact the provider through a known channel and secure the account from a trusted device. If you shared banking details or made a payment, contact your bank immediately.

Separate the message’s technical identities

From, Reply-To and the SPF-evaluated domain can differ. For DKIM, record the d= signing domain; an intermediary may sign the message. Have the trusted mail service interpret results. Technical authentication does not authorise document disclosure or a payment change.

Describe the action actually performed

Separate message receipt, link opening, submitted credentials and ordered payment. Preserve the original and headers: a screenshot of body text may omit needed fields. Use the designated reporting channel without circulating a suspicious attachment to everyone.

Reading technical evidence

Read Authentication-Results within its trust context

A result copied into an email is not automatically trustworthy. Its origin in the mail chain matters.

Field or situation and its interpretation
Observed itemWhat it means
authserv-idIdentify the service claiming to have performed the check.
spf=pass or dkim=passRead the mechanism and domain concerned, rather than just the word pass.
An upstream headerAsk the mail team which results originate from its trusted services.

Preserve the original message and headers. Even a trustworthy result does not authorise a requested bank-account change.

RFC 8601, § 7

Apply the method

Inspect a suspicious message without following its instructions

Before opening attachments or supplying a code, examine the actual sender address, requested action, reply destination and link hostname. Open the known application or type the usual address yourself. Report through your organisation’s designated channel rather than circulating the message.

Verify the action even when presentation and spelling are convincing. A bank-details change, one-time-code request or urgent access demand warrants an independent check using a previously known contact. If you already disclosed credentials or sent funds, tell the responsible team or bank immediately, preserve the original message and record actions and times.

Request
Identify payment, code or unusual urgency.
Verification
Use the known application or established contact.
Reporting
Use the designated channel before removing the message.

Fictional example: urgent invoice. An employee calls the established supplier contact about a bank-details change. The supplier denies the request and the employee reports the original message.

Does a familiar display name establish authenticity?

No. Confirm the requested action independently before acting.

Try the reasoning

The display name is your bank but replies go elsewhere. What next?

Reasoning: Do not reply. Contact the bank through its usual channel and report the discrepancy.

Read the message without following its path
ItemPossible observationIndependent check
Displayed identityNames and logos can be copied.Locate the established contact.
DestinationReply, link or payment destination.Compare with previously known details.
RequestUrgency, secrecy or unexpected change.Confirm through another channel.
From doubt to verification
  1. PauseDo not execute the request.
  2. FindUse an independent channel.
  3. DocumentRetain necessary evidence without amplifying it.

NIST — Reconnaître et signaler le phishing

Worked example

An authenticated-looking message requests an unusual document

Fictional teaching case, not a client investigation.

Is the request authorised even if it appears to come from the usual account?

Observation, interpretation and limit
ObservationWhat it establishesWhat remains unproven
The message appears in a known conversation and quotes a real file.The apparent sender has conversation context.Account access or a copied conversation could explain that knowledge.
The form destination uses a different hostname from the usual service.The document would reach a different destination.A legitimate third-party platform remains possible; a change does not prove fraud.
The contact reached through a previously retained number does not recognise the request.The request is unconfirmed through the independent channel used.The technical origin still requires examination by the relevant team.

Conclusion to record

Withhold the document until the destination and authorisation are confirmed. Preserve the original message and refer the discrepancy to the file owner.

What would change the conclusion?

Review after confirmation by an authorised contact through the known channel, clarification of the platform’s role and the data actually needed.

Check your reasoning

Replace the displayed name with the actual hostname in your reasoning. Would the decision change? Checking a message does not replace checking the requested action.

· Open the verification log

Your working sheet

Turn the method into a documented review.

For each point, record what the material actually supports. “Supported” applies to the specific observation, not to the entire case. A contradiction needs explanation; missing material remains an open question.

  1. Define the question
  2. Record the source
  3. Test the observation
  4. State the limits

Entries stay in this page until you leave or reload it. Nothing is saved or submitted by this tool. Use neutral references instead of personal data or credentials.

1. Sender

Full address, Reply-To and original message headers.

Limit: The display name and conversation thread can be imitated.

2. Destination

Link host read without opening or uploading sensitive messages.

Limit: A redirect can change the final destination.

3. Authentication

SPF, DKIM and DMARC results interpreted by the responsible team.

Limit: Passing checks does not exclude a compromised legitimate account.

How to write a useful conclusion
Editorial examples, not findings about your case
MaterialUseful wordingNext check
One dated sourceThe consulted source stated X on the recorded date.Check its origin and whether it still applies.
Multiple copies of one claimSeveral pages repeat X; independent corroboration is missing.Find the original and a source with independent evidence.
Conflicting sourcesA states X while B states Y; the difference is unresolved.Compare dates, scope, identifiers and definitions.
No matching resultNo result was found in the recorded scope.Describe queries, coverage and access limitations.

What to check and what to record

Use this grid to turn the method into a reviewable case file. A missing item remains an open question. The interpretation limit prevents a finding from becoming an unsupported conclusion.

Topic-specific checks
CheckUseful recordInterpretation limit
SenderFull address, Reply-To and original message headers.The display name and conversation thread can be imitated.
DestinationLink host read without opening or uploading sensitive messages.A redirect can change the final destination.
AuthenticationSPF, DKIM and DMARC results interpreted by the responsible team.Passing checks does not exclude a compromised legitimate account.

Common pitfalls

Four shortcuts that weaken the result.

Clicking to look

Opening the link may signal an interaction, redirect you or trigger a download.

Two people checking a request in a direct conversation — AI-generated photograph

Replying to the sender

This may confirm an active address and keeps the conversation in the sender’s channel.

Trusting the display name

It is neither the technical address nor proof of identity.

Forwarding widely

Uncontrolled circulation increases click risk and data exposure.

Practical questions

Frequently asked questions.

Is a well-written message reliable?

No. Writing quality proves neither sender identity nor legitimacy.

Can a search engine verify a link?

Search adds context but does not replace identifying the actual domain and using the official channel.

What if I already entered a password?

Change it through the official service, revoke sessions, enable multifactor authentication and alert the appropriate support team.

Public references

These primary references support the checks described here. The sequence and working sheet are an IIS editorial method; a cited organisation does not endorse this guide. Read each reference within its stated date, jurisdiction and scope.

UK NCSC — Report a scam email. UK reporting instructions, with separate guidance for suspicious texts and for people who have already responded.

US FTC — Recognise and avoid phishing scams. US guidance on independent checks, first actions after responding and reporting suspicious messages.

RFC 5322. Reference for the technical fields and interpretation boundaries described in this guide.

RFC 6376 — DKIM. Reference for the technical fields and interpretation boundaries described in this guide.

Editorial scope

Published by Internet Intelligence Service on 15 September 2026. Last content update: 21 November 2026. This educational guide describes a lawful, defensive method. It is not legal advice, an emergency service or authority instruction.

Find primary portals and their limitations

Editorial responsibility, examples and corrections