AegisAI Perspectives · 161
A one-person IT team triages reported phishing in native admin tools. Read the report, find and remove every copy, block the sender, then check sign-ins.
See it on your own Google Workspace or Microsoft 365 tenant. No MX change.
Or get the Q3 2026 State of the AI Attack report (free, instant download)
Not ready for a demo? Check your domain.

Work each report through 5 checks in tools you already pay for. In Google Workspace, the investigation tool can identify all users in your domain that have received the message20. AegisAI investigates every email for what it is trying to do and removes the ones that fail from every mailbox they reached19. In Microsoft 365, the search tool is Real-time detections, available in Defender for Office 365 Plan 19.
Each check in the table below has its tool, its time limit and the decision to make. What a product does after delivery is on the inbound email security page. How to pick a product for a small team is in how to choose email security for a lean IT team with no SOC.
Key takeaways
| Question | Answer |
|---|---|
| How does a one-person IT team at a startup triage reported phishing with no SOC? | By hand, with five checks in the native Google Workspace or Microsoft 365 admin tools, in the order of the checks table below |
| Where do user reports land? | Google: the alert center, where alert details show the total user reports4. Microsoft: the User reported tab on the Submissions page7 |
| How far back can I search? | Threat Explorer and Real-time detections go back up to 30 days11 |
| How fast does a block take effect? | A Tenant Allow/Block List entry should be active within 5 minutes15 |
| What if the reporter typed a password? | Reset the password and sign-in cookies6, then check sign-in logs18 |
Short on time? Watch the explainer, or book a demo.
Search by subject and sender in the investigation tool or Real-time detections, then delete every match in the same sitting. Steps 3 to 5 are Google only; steps 6 and 7 are Microsoft only.
Open the report. In Google, an admin with the Alert center administrator privilege opens the alert3. In Microsoft, open the User reported tab on the Submissions page.
Note the subject and sender, plus the time of the first report. The Google alert details show the message ID and subject hash, plus the recipient username, for sample reports.
In the Google investigation tool, choose Gmail log events as the data source and search by subject and date. The tool can identify all users in your domain that have received the message.
Delete the matching messages. Delete message and Restore message can each be used for 30 days after the event is logged5.
Check Google Vault for a retention rule or hold that covers the message. Erased messages are still subject to any retention rules and holds set in Vault2.
In Microsoft, filter Real-time detections or Threat Explorer by sender or subject, with dates up to 30 days ago. Remediate with move to junk, soft delete or hard delete12, up to 100 emails at a time when you select by hand.
Block the URL or sender in the Tenant Allow/Block List. An entry should be active within 5 minutes. Submit the message to Microsoft while it is under 30 days old and still in the mailbox14.
Ask the reporter one question: did you click or type a password? Then check the sign-in logs for that account.
Write one line per report with the subject, the recipients, the removed count and the account action.
Keep the one-line log in a shared sheet so the next report starts from the last one.
User reports land in the alert center or on the Submissions page, then five checks follow in this order. Each time limit is a window or delay the tool sets, and waiting past it loses data or options.
| Check | Google Workspace tool | Microsoft 365 tool | Time limit | Decision |
|---|---|---|---|---|
| 1. Read the report | Alert center alert details: total user reports and sample reports with message ID, subject hash and recipient | User reported tab on the Submissions page, fed by a reporting mailbox | None stated | Several reports of one subject: treat it as one campaign |
| 2. Find every copy | Investigation tool, data source Gmail log events (Frontline Plus, Enterprise Plus, Education Plus) | Real-time detections (Plan 1) or Threat Explorer (Plan 2) | Date filter up to 30 days ago (Microsoft) | List every recipient before you delete anything |
| 3. Remove the copies | Delete message from the alert; Vault retention and holds still apply | Move to junk, soft delete or hard delete; up to 100 emails selected by hand | Delete message usable for 30 days after the event is logged (Google) | Confirmed bad: delete. Unsure: move to junk |
| 4. Block and submit | Not covered by the Google pages used here | Tenant Allow/Block List entry; admin submission to Microsoft | Block active within 5 minutes; submissions up to 30 days old | Block the link or sender the same hour |
| 5. Check who signed in | Reset the password and sign-in cookies | Sign-in logs, then revoke sessions | Sign-in logs kept 7 days on Entra ID Free, 30 days on P1 and P2; tokens last 1 hour | Any sign-in after the click: contain the account |
Print this table and keep it next to the admin console. It is the working answer to the question: how does a one-person IT team at a startup triage reported phishing with no SOC?
Reset the password and end the sessions in the same sitting. In Google Workspace, resetting a user's sign-in cookies prevents them from signing in with an old password. In Microsoft 365, access tokens issued by Microsoft Entra ID last 1 hour by default, so a password reset alone leaves a live session for up to that hour.
Check the sign-in logs before you close the case. Microsoft Entra ID keeps sign-in logs for seven days on the Free tier and 30 days on P1 and P2. On Free, a report older than seven days has no sign-in history left to check.
Reset the password, revoke sessions, then read the sign-in log.
The search in check 2 needs a specific Google Workspace edition or a Defender for Office 365 plan. Gmail log events, the data source the investigation tool searches, are available only with the Frontline Plus, Enterprise Plus or Education Plus editions. Real-time detections is available in Defender for Office 365 Plan 1, and Threat Explorer in Plan 2.
If your plan has neither search tool, you cannot run check 2 with the tools in this runbook. Close that gap first, because check 2 decides how many mailboxes check 3 has to clean.
Open your admin console today and confirm which of these editions and plans you hold.
Automated Investigation and Response (AIR) needs Microsoft Defender for Office 365 Plan 2 licenses16. Without Plan 2, you run the five checks by hand.
AegisAI reports on its inbound email security page that 3 in 10 confirmed phishing emails cleared native Gmail and Microsoft filters in Q3 2026. How a team automates the queue itself is covered in how to automate triage of user-reported phishing.
Automate removal first, because check 3 is where the limit of 100 emails selected by hand applies.
One bad email is pulled from every inbox it reached, forwards included21. Every verdict comes with its reason in plain language, so your team can check it and share it. The inbound email security page states that the median verdict takes 1.9 seconds, and a convicted message is removed from every mailbox it reached within seconds.
Track three numbers from your one-line log.
To watch a reported email investigated and removed in your own tenant, talk to the team. On day 30, compare the minutes per report with the number of reports.
In native tools, read the report, find every copy, remove the copies, block the sender, then check who signed in. To take removal off your desk, AegisAI investigates every email for what it is trying to do and removes the ones that fail from every mailbox they reached.
No. Microsoft's guidance is: don't forward the original message to the reporting mailbox. Use the report button, so the message reaches the User reported tab on the Submissions page.
In Microsoft 365, Threat Explorer and Real-time detections take dates up to 30 days ago. Admin submissions accept messages up to 30 days old. In Google Workspace, Delete message can be used for 30 days after the event is logged.
No. Erased messages are still subject to any retention rules and holds set in Vault. The message leaves the user's inbox, and Vault rules decide how long a copy is kept.
Not on its own. In Microsoft 365, revoke the user's sessions as well. In Google Workspace, reset the user's sign-in cookies.