Microsoft 365’s built-in protection is good at stopping bulk spam, known malware, and obvious phishing. It’s less effective against the targeted attacks that actually cost businesses money: email from a compromised vendor account, a payment request with no link or attachment, or a sign-in that looks legitimate but isn’t. Those attacks are designed and tested to get past default filtering, and someone needs to be watching when they do.
The same is broadly true of Google Workspace. This isn’t a knock on either platform. It’s the reason a second layer exists.
What you get out of the box
Every Microsoft 365 mailbox includes Exchange Online Protection, which handles spam, known malware, and basic anti-phishing. Higher-tier plans add Microsoft Defender for Office 365, with features like link rewriting, attachment sandboxing, and impersonation protection. Configured well, that’s a solid baseline.
The catch is in “configured well.” Many organizations run with default policies, have advanced features licensed but switched off, or have exceptions and allow-lists that have quietly piled up over the years.
Where the gaps are
1. Attackers test against it
Microsoft 365 is the most widely used business email platform, so it’s the one attackers optimize for. Phishing kits are tuned to get through common default configurations before a campaign ever launches.
2. Trusted senders that aren’t
When a real vendor’s mailbox is compromised, the attacker’s email comes from a legitimate, authenticated domain with a real sending history. Filters built around sender reputation and authentication have little to object to. The same goes for lookalike domains with their own valid authentication.
3. No payload to scan
Business email compromise often contains no link and no attachment. It’s a short, polite request to pay an invoice or update bank details. There’s nothing malicious to detonate, so detection depends on spotting impersonation and unusual requests.
4. Attacks after a successful login
Once an attacker has valid credentials, the threat moves from the inbox to the account: new forwarding rules, malicious app consents, and internal phishing sent from a real colleague’s mailbox. These are account takeover signals, and they live in sign-in and audit logs that someone has to watch.
5. Nobody acting on the alerts
Even when Microsoft flags something, an alert in a portal nobody checks doesn’t stop anything. Most small and mid-sized IT teams don’t have someone whose job is to triage email security alerts as they arrive.
What to add
- Tune what you already own. Review your anti-phishing, safe-links, and attachment policies, remove stale allow-list entries, and turn on features you’re already paying for.
- Add a detection layer built for targeted attacks, one that looks for impersonation, BEC language, and anomalous account behavior, not just known-bad files and links. API-based tools connect without changing your mail routing.
- Watch account behavior, not just messages: unusual sign-ins, new inbox rules, and new app consents.
- Make sure someone acts. Detection only helps if a person triages and responds while it matters.
- Watch outside your tenant. Lookalike domains are registered and staged on the open internet before any email is sent. Brand protection monitoring catches them there.
How FirmGuardian helps
Email & Workspace Protection adds a monitored layer over Microsoft 365 or Google Workspace, built on Check Point Harmony. It connects by API, with no MX record changes, and covers phishing and BEC across incoming, outgoing, and internal email, account takeover, malicious links and attachments, and shadow IT. Our team operates it and tells you what we found and what we did, in plain language.
Related reading: Account takeover warning signs and Business email compromise.
Close the gaps in your inbox
Talk to us about Email & Workspace Protection for Microsoft 365 or Google Workspace.