Person using a laptop and a phone at the same time
Cybersecurity

What to Do After an Employee Clicks a Phishing Link

Someone in your office clicks a link in an email, lands on a page that looks like a Microsoft 365 sign-in, and types in a password. Two minutes later they realize something was off. They walk over to your desk, or they send a message that starts with "this is probably nothing, but."

What happens in the next stretch of time decides whether this is a password reset or a month of cleanup. Most small offices have no written plan for this moment, so people improvise. This post gives you a plan you can run with the tools you already have, in the order Microsoft's own incident guidance recommends.

The steps below assume Microsoft 365. Google Workspace follows the same logic, with different menus.

Why One Click Gets Expensive

The FBI's Internet Crime Complaint Center received 191,561 phishing and spoofing complaints in 2025, with reported losses of $215,843,126. Business email compromise, where an attacker uses a hijacked or spoofed mailbox to redirect payments, produced 24,768 complaints and $3,046,598,558 in reported losses. Total reported losses across all crime types passed $20.8 billion. These figures come from the FBI IC3 2025 Internet Crime Report.

Phishing complaints are about eight times as numerous as business email compromise complaints, but the reported loss per business email compromise complaint is far higher. A stolen password can give an attacker a mailbox, and the mailbox holds your invoices, your vendors and your customers.

What an Attacker Does With a Fresh Password

An attacker who signs in as your employee can read mail, search for words like "invoice" and "wire," and send messages that come from a real address inside your company. They can add an inbox rule or a forwarding address so they keep seeing mail after the password changes. Microsoft's guidance lists forwarders and inbox rules as specific places to check for this reason.

The First Five Minutes

Whoever clicked should tell someone right away. That sounds obvious, but it is the step most offices lose. People hide a click when they expect to be blamed for it. Say so out loud ahead of time: reporting a click quickly is the correct move, and nobody will be in trouble for it.

If the Person Only Clicked

If they opened the page and entered nothing, close the tab, delete the message and report it. CISA's guidance on phishing says to report the message using the report option in your email program, then delete it, and to avoid clicking any attachment or link in it. You can read the steps on CISA's recognize and report phishing page. Tell whoever manages your IT anyway, so they can check the device and watch the account.

If the Person Typed In a Password

Treat the account as compromised. The rest of this post is for that case. Write down the rough time of the click and the address of the sender, then hand the account to whoever has administrator rights in Microsoft 365.

Locking the Account Down

Microsoft's page Respond to a compromised cloud email account lays out the response as a sequence. These are the steps that matter most for a small office.

Disable the Account or Reset the Password

Microsoft calls disabling the account preferred and highly recommended until the investigation is finished. If you cannot disable it, the next best step is a password reset with a strong new password that is not one of the last five. Do not send the new password to the user by email, since the attacker may be reading that mailbox. Give it to them in person or by phone.

App passwords survive a password reset. If the person used any, they should delete the old ones and create new ones.

Revoke Active Sessions

A password reset does not necessarily end a session the attacker already has open. Revoking sign-in sessions invalidates refresh tokens, which cuts off access that was granted with the stolen credentials. An administrator can do this in PowerShell with the Microsoft Graph command Revoke-MgUserSignInSession.

Check the MFA Methods

An attacker can register their own phone or authenticator app to sign in again later. Review the multifactor authentication methods registered on the account and remove any device or method the employee does not recognize. If the account does not have MFA, turn it on before you re-enable it. Microsoft highly recommends enforcing MFA for the account.

Remove Hidden Forwarding and Inbox Rules

This is the step that gets skipped. Check the mailbox for forwarding addresses and for inbox rules, including hidden ones. Microsoft gives the commands: Get-Mailbox with the Forwarding address properties, and Get-InboxRule with the IncludeHidden switch. A rule that redirects or forwards mail to an outside address, or quietly moves messages with certain words into a folder, is a sign the attacker was there. Remove anything you did not create.

Review App Consent and Admin Roles

Microsoft also tells you to review applications the user has consented to and to remove any that should not be there. If the account holds an administrator role, check that it is still the role you gave it. For an admin account, escalate to an experienced person before you do anything else.

Finding Out What the Attacker Saw

Once the account is locked, the question changes from "can they get in" to "what did they do." Microsoft's guidance points to the Microsoft Entra sign-in logs and audit logs, which show IP addresses, sign-in locations and times, and whether each attempt succeeded. It recommends starting the date range just before the suspicious activity and not filtering by specific activity at first.

Two other places answer practical questions. The Sent Items folder and Microsoft's message trace show whether the account sent mail the employee did not write. If it did, those recipients need a short, plain note that the messages were not legitimate and should be deleted.

Questions to Answer in Writing

Write down the answers while they are fresh. When was the first suspicious sign-in? Which mailbox folders held financial or customer information? Did any outgoing message ask for money or a changed bank detail? Did the account have access to shared files? Those answers show whether you have a password problem or a data problem, and they give anyone you report to a clear timeline.

What to Do About Money and Customers

If the compromised mailbox touched payments, call your bank. Do it before you finish the technical cleanup. Tell the bank which payments went out and when, and ask what it can do. You can also file a complaint with the FBI at ic3.gov.

If customer or employee data was in the mailbox, the next question is whether you have a notice obligation. State law and some contracts set those rules, and they vary. Get advice from counsel before you decide, and keep your notes from the step above.

Making the Next Click Cheaper

Some phishing email will always get through. The work that limits the damage happens before anyone clicks.

  • Turn on MFA for every account. A stolen password alone then no longer opens the mailbox.
  • Add a report button. Make reporting a suspicious message one click, and tell staff reporting is always welcome.
  • Write the plan down. One page: who to tell, who holds admin rights, who calls the bank. Put it where people can find it without email.
  • Practice once a year. Walk through the steps above on a test account so the first time is not the real one.
  • Limit forwarding. Block automatic forwarding to outside addresses unless someone has a reason for it.

When to Bring In Help

A single mailbox with no admin rights and no financial mail is something a careful office administrator can handle with Microsoft's checklist. Bring in someone experienced if the account was an administrator, if money moved, if the attacker sent mail from the account, or if you cannot tell what they accessed. Canyon runs cybersecurity and recovery work for these situations, and the managed IT service covers the setup that makes them smaller. If you want to talk through a plan for your office, get in touch.

Reviewed by Blair Patterson

Sources & References

  1. [1] FBI IC3 2025 Internet Crime Report

    Phishing/spoofing: 191,561 complaints, $215,843,126 losses. BEC: 24,768 complaints, $3,046,598,558 losses. Total losses above $20.8 billion.

  2. [2] Respond to a compromised cloud email account (Microsoft Learn)

    Response sequence: disable account or reset password, revoke sessions, review MFA devices, app consent, admin roles, forwarders and inbox rules, then review logs.

  3. [3] Recognize and report phishing (CISA)

    Report the message, delete it, do not click links or attachments.

Keywords
clicked phishing linkphishing responsecompromised Microsoft 365 accountbusiness email compromisesmall business cybersecurityincident response checklist

Want help with this?

We can put this into practice for you and help keep your business protected.