2.3.1. Threat Policies and Rules
💡 First Principle: Microsoft publishes recommended settings for a reason: most tenants get compromised through misconfigured or missing policies, not exotic attacks. Preset policies apply those recommendations as a bundle that Microsoft keeps current — which is why they outrank anything you build yourself.
Where: Microsoft Defender portal > Email & collaboration > Policies & rules > Threat policies.
Preset security policies are the starting point:
- Built-in protection — on by default, giving every recipient basic Safe Links and Safe Attachments protection.
- Standard and Strict — both off by default. Standard is Microsoft's recommended baseline for most users; Strict is for high-value targets such as executives and finance. You can't edit their settings — only add impersonation-protection entries and exceptions.
Order of precedence is the most-tested fact here. A recipient is evaluated against policies in this order, and processing stops at the first one that applies:
So a user included in the Standard preset ignores your custom Safe Links and Safe Attachments settings. If you need different behavior for that user, exclude them from the preset — raising the custom policy's priority won't help.
The individual policy types:
| Policy | What it stops | Settings worth knowing |
|---|---|---|
| Anti-phishing | Impersonation and spoofing | User impersonation protection for up to 350 users per policy; presets protect all your accepted domains from domain impersonation; mailbox intelligence on by default; spoof intelligence on in every profile |
| Safe Attachments | Malicious files, by detonating them in a sandbox | Actions: Off, Monitor, Block (default, used by Standard and Strict), or Dynamic Delivery (deliver the body now, attachments after scanning — Exchange Online mailboxes only) |
| Safe Links | Malicious URLs, checked at time of click | Rewrites URLs in email; in Teams and Office apps, checks at click time without rewriting |
| Anti-spam | Spam and bulk mail | Bulk complaint level (BCL) threshold 1–9, default 7; zero-hour auto purge (ZAP) moves mail after delivery |
| Anti-malware | Known malware | Common attachments filter for risky file types |
The Tenant Allow/Block List (Policies & rules > Threat policies > Tenant Allow/Block Lists) handles exceptions for domains, email addresses, files, URLs and spoofed senders. Block entries expire after 30 days by default (configurable to 90 days or never), while spoof blocks never expire. Quarantine policies decide what users may do with quarantined mail. Malware and high-confidence phishing use AdminOnlyAccessPolicy by default: users can't release them (at most request release) and get no notifications.
⚠️ Exam Trap: "A user in the Standard preset still gets a custom Safe Attachments setting you configured" is false — presets win. And "Safe Links rewrites URLs in Teams messages" is false; Teams gets time-of-click checks without rewriting.
Reflection Question: Your CFO is the target of repeated impersonation attempts. Which preset would you put the CFO in, which anti-phishing setting adds protection specifically for their name, and why won't a high-priority custom policy override the preset?