SCSecurityGuide
Email Security for Microsoft 365 and Google Workspace
Most attacks still arrive by email: phishing links, malicious attachments, impersonated executives and fraudulent payment requests. This guide covers the configuration that matters in Microsoft 365 and Google Workspace, from domain authentication to user reporting.
- Published
- Updated
- Reading
- 3 minutes
- Prepared by
- EngineSpire Technologies
Key points
- Protect accounts first: MFA, blocked legacy authentication and sign-in monitoring.
- Publish SPF, DKIM and DMARC, and move DMARC to enforcement once legitimate senders are aligned.
- Turn on built-in protection for links, attachments and impersonation, and review quarantine regularly.
- Block automatic forwarding to external addresses and alert on suspicious inbox rules.
- Combine technology with an easy report button and a mandatory call-back for payment changes.
The threats
- Credential phishing pages that imitate sign-in screens.
- Malicious attachments and links to malware.
- Business email compromise: impersonated executives or suppliers requesting payments or bank detail changes.
- Taken-over mailboxes used to phish customers and partners.
- Spoofed messages that pretend to come from your own domain.
Protect the accounts
- Enforce MFA for every mailbox user, with phishing-resistant methods for administrators.
- Disable legacy authentication protocols that cannot use MFA.
- Use conditional access or context-aware access to restrict risky sign-ins.
- Monitor sign-in alerts and review them promptly.
- Keep administrator accounts separate from mailboxes used for daily email.
Authenticate your domain
| Record | What it does | Practical tips |
|---|---|---|
| SPF | Lists the servers allowed to send email for your domain | Include every legitimate sender, such as CRM, marketing and ticketing tools; stay within the ten DNS lookup limit |
| DKIM | Adds a cryptographic signature that receivers verify | Enable it in Microsoft 365 or Google Workspace and for each third-party sender; rotate keys periodically |
| DMARC | Tells receivers what to do when messages fail aligned SPF or DKIM, and sends you reports | Start with p=none, fix unaligned senders using the reports, then move to quarantine and reject |
Mailbox providers increasingly require authentication. Google's sender guidelines, for example, require SPF, DKIM and DMARC for organizations sending large volumes to personal Gmail accounts.
Microsoft 365
- Apply Microsoft's preset security policies (Standard or Strict) as a starting point.
- Where licensed, use Microsoft Defender for Office 365 features: Safe Links, Safe Attachments and impersonation protection for executives and key domains.
- Control automatic external forwarding through the outbound spam policy.
- Use alert policies for suspicious inbox rules and forwarding.
- Show external sender tags and make the built-in Report button available to users.
- Review quarantine policies so users know what is held and how to request release.
Google Workspace
- Enable the Gmail safety settings for attachments, links and spoofing, including protection against lookalike domains and employee name impersonation.
- Turn on enhanced pre-delivery message scanning.
- Use the security sandbox for attachments where your edition includes it.
- Show warnings for external recipients and restrict automatic forwarding.
- Monitor the alert center for suspicious sign-ins and forwarding changes.
Forwarding, rules and connected apps
Attackers who take over a mailbox often create rules that forward copies of mail or hide replies. Block automatic external forwarding by default, alert on new forwarding and deletion rules, and review third-party applications that users have allowed to read mail.
Stop payment fraud with process
- Require a call-back to a phone number already on file before changing any supplier's bank details.
- Use dual approval for payments above a set amount.
- Verify urgent or confidential payment requests from executives through a second channel.
- Control changes to supplier master data in the finance system.
Help people help you
- Run short, regular awareness sessions built around real examples.
- Use phishing simulations to teach, not to punish.
- Make reporting a suspicious message one click, and tell reporters what happened.
When a mailbox is compromised
- Reset the password, revoke sessions and review registered MFA methods.
- Remove malicious inbox rules, forwarding and connected applications.
- Check sent items and warn anyone who received phishing from the account.
- Assess what data the attacker could have accessed.
- If personal data was exposed, follow your breach management procedure.
Common questions
What DMARC policy should we use?
Start with p=none and review aggregate reports to find every legitimate service sending as your domain. Once they pass SPF or DKIM with alignment, move to p=quarantine and then p=reject to stop spoofing.
Do we need a third-party email security gateway?
Not necessarily. Microsoft 365 and Google Workspace include substantial protection when configured well, and higher-tier licences add more. Third-party tools can add value in some environments, but configure and measure the built-in features first.
How do we stop fake supplier payment requests?
Technical controls reduce spoofing, but a genuinely compromised supplier mailbox can still send convincing requests. Require call-back verification to a known phone number before changing bank details, and dual approval for payments.
Sources and further reading
- [1]DMARC overview — DMARC.org
- [2]RFC 7489: Domain-based Message Authentication, Reporting, and Conformance (DMARC) — IETF
- [3]Set up DMARC to validate the From address domain — Microsoft Learn
- [4]Preset security policies in EOP and Microsoft Defender for Office 365 — Microsoft Learn
- [5]Email sender guidelines — Google Workspace Admin Help
Need this applied to your environment?
We can turn this guide into a scoped assessment of your own sites, systems and constraints.
Keep reading
- SCFirewall High Availability: What a Real Failover Test Should ProveA firewall pair that reports "synchronized" has not proven anything yet. A real failover test checks traffic, sessions, VPNs, routing, logging and recovery under conditions that resemble an actual failure.3 min read
- SCZero Trust for Mid-Market Organizations: A Phased RoadmapZero trust replaces the idea of a trusted internal network with decisions based on identity, device and context for every access request. Mid-sized organizations can get there in phases, starting with identity and the most sensitive systems.4 min read
- SCCCTV Storage Sizing: Retention, Resolution and Real-World BitrateCCTV storage estimates built from datasheet bitrates are often wrong, because real bitrate depends on the scene, lighting, motion and camera settings. This guide shows how to size recording storage from measured values and agreed retention.4 min read
