Email security for small businesses
Email security for small business owners comes down to five attacks: phishing, account takeover, spoofing your domain, look-alike addresses, and business email compromise, which usually arrives as a fake invoice or a change of bank details. Multi-factor authentication keeps a phished password from opening the mailbox. SPF, DKIM and a DMARC policy of reject let receiving servers refuse mail that fakes your exact domain, while look-alike addresses need inbox warnings and careful staff. The defenses against fake invoices are turning off forwarding to outside addresses, knowing where the mailbox audit log is, filtering, and an office rule that nobody changes bank details because an email said to.
Business email compromise is the second-largest source of reported losses in the FBI's numbers, behind investment fraud. The FBI Internet Crime Complaint Center's 2025 report counts 24,768 business email compromise complaints in 2025 with reported losses of $3,046,598,558 (pages 25 and 26). That works out to about $123,000 per complaint. The same report lists 191,561 phishing and spoofing complaints, more than any other category.
This guide takes each attack in turn, says which setting in Microsoft 365 or Google Workspace addresses it, and links to the posts that go step by step. It ends with a checklist.
How email gets used against a small office
| Attack | What it looks like | Main defense |
|---|---|---|
| Phishing | A message that gets someone to type a password into a fake sign-in page, open a file or approve a sign-in prompt | Multi-factor authentication, filtering, a habit of reporting |
| Account takeover | Someone signs in to a real staff mailbox with a stolen password | Multi-factor authentication, blocking old sign-in methods, the audit log |
| Spoofing your domain | Mail with your exact address in the From line, sent from a server you don't control | SPF, DKIM and DMARC at p=reject |
| Look-alike domains | Mail from a domain one letter off yours, or your name over a stranger's address | Warnings in the inbox, impersonation protection, staff habits |
| Business email compromise and invoice fraud | A request to pay, or to change bank details, from a real or faked mailbox | Checking payment changes by phone, forwarding controls, the audit log |
The attacks chain together. A phished password leads to a taken-over mailbox, and the taken-over mailbox sends the fake invoice. That's why no single setting covers email.
Phishing
Phishing is email written to get someone to hand over a password, open a malicious file or approve a sign-in. The joint CISA, NSA, FBI and MS-ISAC phishing guidance names training staff to spot and report suspicious messages, multi-factor authentication, and DMARC among its main defenses. It also points out that attackers spoof caller ID over internet phone services, so an incoming call that seems to confirm the email proves nothing. Call out, to a number you already have.
For a small office, three habits do most of the work:
- Report suspicious messages with the built-in report button in Outlook or Gmail instead of forwarding them around the office. The CISA guidance says built-in reporting helps providers spot new campaigns.
- Treat any page that asks for your email password after you click a link as fake until proven otherwise. Type the address of Microsoft 365 or Google yourself.
- Tell people that clicking by mistake is fixable if they say so right away, and that hiding it is what turns a click into a breach.
Filtering catches most phishing before anyone sees it, and the section on filtering below covers the settings. If someone has already typed a password into a fake page, what to do after clicking a phishing link goes through the steps in order. The cybersecurity checklist for small businesses puts email in context with the rest of the office's security.
Account takeover and multi-factor authentication
Account takeover is someone signing in to a real mailbox with a stolen or guessed password. Once in, they can read months of client mail, send from the account, and set up rules that hide what they're doing. The defense is multi-factor authentication (MFA), a code or approval from your phone on top of the password, so the password alone no longer opens the account.
In Microsoft 365, the simplest route is security defaults. Microsoft's security defaults page gives the steps: in the Microsoft Entra admin center, browse to Entra ID, Overview, Properties, select Manage security defaults, set Security defaults to Enabled and save. Security defaults require every user to register for MFA, require MFA for administrators, and block legacy authentication, the older sign-in methods that can't do MFA. Since July 29, 2024, users no longer get a 14-day grace period to register, according to the same page.
Blocking legacy authentication will stop old scanners and mail apps that sign in with only a password. Check those before you switch it on, so scan-to-email doesn't break on a Monday morning.
In Google Workspace, a super administrator goes to Menu, Security, Authentication, 2-step verification in the Admin console, checks Allow users to turn on 2-Step Verification, and sets Enforcement to On, per Google's 2-Step Verification guide. The optional new-user enrollment period lets new staff sign in with only a password until it ends, so keep it short.
Not all MFA is equal. CISA's fact sheet on phishing-resistant MFA calls FIDO-based methods such as security keys and passkeys the gold standard, and warns that codes by text message and push prompts without number matching can be defeated. Microsoft says number matching is enabled for all Authenticator push notifications, so a Microsoft 365 user has to type the number shown on the sign-in screen into the app. Start with any MFA for everyone, then move administrators and the people who approve payments to passkeys or security keys.
How to turn on MFA in Microsoft 365 walks through setup, including the accounts and devices that tend to break.
Spoofing your domain
Spoofing is mail that puts your exact domain in the From line but comes from a server you don't control. Nothing in the basic email system stops it. Three DNS records let receiving servers tell your real mail from a fake:
- SPF lists the servers allowed to send mail for your domain.
- DKIM adds a signature to each message that receivers check against a key in your DNS.
- DMARC tells receivers what to do when a message claiming to be from your domain fails both checks for your domain: nothing (
p=none), treat it as suspicious (p=quarantine), or refuse it (p=reject).
DMARC was first published in 2015 as RFC 7489. In May 2026 the IETF replaced it with RFC 9989, a standards-track version that keeps the same three policies. The CISA phishing guidance asks organizations to set DMARC to reject, because spoofed mail is then refused before it's delivered.
Most small offices haven't done this. On October 14, 2026, we checked the public DNS records of 316 small tax and accounting firms in Alabama, Tennessee and Georgia. Of the 262 whose domains receive email, 199 (76%) had nothing telling receivers to act on mail that fakes their domain: 122 had no DMARC record at all, 73 were at p=none, and 4 had broken records. How to set up DMARC for a small business has the full numbers and the steps, from turning on DKIM to moving to reject over a few weeks.
Two things to know before you start. A reject policy blocks your own mail too if a service that sends as you, such as invoicing software or a website contact form, isn't set up with your domain. And domains you own but don't use for email need records too, which Microsoft's Defender for Office 365 overview calls out for used and unused domains alike.
Look-alike domains and fake display names
DMARC protects your exact domain and nothing else. Section 2.2 of RFC 9989 says DMARC doesn't address visually similar domain names, or abuse of the display name, the human-readable name shown before the address. So DMARC does nothing about mail from yourfirrn.com, or a message showing your office manager's name over a free webmail address.
These need warnings in the inbox and staff habits:
- Show full addresses. The FBI's guidance on business email compromise says to check that the sender's address matches who it claims to be, especially on phones, and to turn on settings that show full email addresses.
- First contact safety tip in Microsoft 365. Every Microsoft 365 mailbox can show "You don't often get email from" a sender. Microsoft recommends turning it on in the anti-phishing policy, under Email & collaboration, Policies & rules, Threat policies, Anti-phishing in the Defender portal, according to its anti-phishing policy guide. The same policy can add a question mark to the sender's photo when a message fails authentication.
- Impersonation protection. Microsoft's protection against look-alike users and domains is only in Defender for Office 365, which is included in Microsoft 365 Business Premium as Plan 1 and sold as an add-on to other plans. Even with Business Premium, the people and domains to protect have to be added to the anti-phishing policy, since none are set there by default.
- Google Workspace. In the Admin console under Apps, Google Workspace, Gmail, Safety, the Spoofing and authentication section has "Protect against domain spoofing based on similar domain names" and "Protect against spoofing of employee names", per Google's guide to advanced phishing and malware protection. Once a protection is on, its default action keeps the email in the inbox with a warning. Moving it to spam or quarantine is stricter.
Microsoft's anti-phishing documentation notes that a look-alike domain can pass SPF, DKIM and DMARC, because the attacker owns it and publishes valid records. Authentication tells you who sent a message, not whether that sender is who they seem.
Business email compromise and invoice fraud
The FBI defines business email compromise as a scam aimed at businesses that pay by wire, carried out by compromising email accounts or other channels through social engineering or computer intrusion to make unauthorized transfers (2025 report, Appendix B). In a small office it usually takes one of two shapes:
- From inside a real mailbox. An attacker who took over a staff account waits for a real invoice or a closing, then replies in the same thread with new bank details. The message is real mail from your server, so it passes every authentication check.
- From a faked or look-alike address. The attacker never gets in. They spoof your domain, if DMARC isn't enforced, or register one close to it, and send your client an invoice or your bookkeeper a request from "the owner".
Filters have little to catch in the first kind, since the message is real mail from a real account. The strongest defense is a habit. The FBI's first piece of advice is to verify any request to change account information through a second channel. In practice:
- Call a number you already have, from your own records, before paying a new account or changing a vendor's bank details. Never the number in the email.
- Have a second person approve any change to payment details.
- Tell clients in writing, in your engagement letter or on your invoices, that your bank details won't change by email.
- Watch your accounts for irregularities such as missing deposits, as the FBI advises. A client payment that never arrives can mean the client paid someone else.
If money has already moved, the bank that sent it gets the first call, which is yours if your office paid and the client's if they did. The FBI says to ask the sending bank for a recall or reversal right away, then file a complaint at ic3.gov with the transaction details. What to do about fake invoices sent from your email address covers how to tell a taken-over account from a spoofed one, what to tell clients, and the calls to make.
Forwarding rules and the mailbox audit log
A forwarding rule is how an attacker keeps reading a mailbox after the password changes. They add a rule that copies every message to an outside address, or one that moves replies from the client into a folder nobody opens. Two controls deal with it.
Turn off automatic forwarding to outside addresses. In Microsoft 365, the setting is Automatic forwarding rules in the outbound spam policy, under Email & collaboration, Policies & rules, Threat policies, Anti-spam in the Defender portal. Microsoft's forwarding guide says the default automatic, system-controlled setting behaves differently from one organization to another, and recommends choosing On or Off explicitly. Off disables inbox rules and mailbox forwarding that send mail to external addresses, returns a bounce to the sender, and leaves forwarding between your own staff alone. If someone needs to forward outside the office, make a custom policy for that person instead of leaving it open for everyone.
In Google Workspace, the setting is under Apps, Google Workspace, Gmail, End User Access, Automatic forwarding. The box Allow users to automatically forward incoming email is on by default, according to Google's help page, and clearing it removes the forwarding option from Gmail settings.
Know where the audit log is. Microsoft says mailbox audit logging is turned on by default in all organizations. By default it records when an inbox rule is added, removed or changed (UpdateInboxRules), when the owner sends mail, and when mail is accessed, among other actions. Audit (Standard) keeps records for 180 days, per Microsoft Purview's audit overview, and an admin searches them with the Audit tool in the Microsoft Purview portal. On Business Basic, Standard and Premium, though, auditing isn't turned on by default, so the first visit to Audit shows a banner to start recording user and admin activity; select it once, and records become searchable within a few hours. Then you need someone who knows the log exists and checks it when an account looks wrong.
Microsoft 365 also watches outgoing mail. Its outbound spam guide says outbound spam from a user usually means a compromised account, and the default alert policies email global administrators when someone is restricted from sending. Make sure those alerts go to a mailbox someone reads.
Forwarding left behind by former staff is the same problem without an attacker. The employee offboarding IT checklist covers what to do with a departing person's mail.
Filtering in Microsoft 365 and Google Workspace
Filtering is one of several settings worth checking; Microsoft 365 security settings for small businesses lists the rest in order.
Every Microsoft 365 mailbox gets built-in protection against malware, spam and spoofing. The easiest way to set it up well is the Standard preset security policy. In the Defender portal, go to Email & collaboration, Policies & rules, Threat policies, Preset Security Policies, and turn on Standard protection for everyone. Microsoft's preset security policy guide says nearly all of its settings aren't configurable and are based on what Microsoft sees in its datacenters. Strict protection is tighter. Spam that Standard puts in the Junk Email folder goes to quarantine instead, so most small offices start with Standard and use Strict only for the people who approve payments.
If you have Microsoft 365 Business Premium, which includes Defender for Office 365 Plan 1, the Standard and Strict presets also cover its Safe Links and Safe Attachments features. Anyone with a Defender for Office 365 license who isn't in either preset gets Microsoft's Built-in protection policy, a basic version of both.
In Google Workspace, the Gmail Safety settings under Apps, Google Workspace, Gmail, Safety hold Gmail's advanced phishing and malware protections, including the spoofing settings above. The same page has Apply future recommended settings automatically, which turns on new settings Google recommends as it adds them.
Filtering is the layer that fails quietly. A rule someone added to let one vendor's mail through can let a lot of other mail through with it. Look at your allow lists once a year.
When something gets through
If someone typed a password into a fake page, approved a prompt they didn't start, or opened a strange attachment, work through what to do after clicking a phishing link: new password from another device, sign out every session, check the sign-in methods and the inbox rules.
If a client got an invoice that looks like it came from you, first work out whether it came from inside your mailbox or only pretends to. The answer decides everything else, and what to do about fake invoices sent from your email address shows how to check.
If money moved, call the bank before anything else, then report it at ic3.gov.
Email security checklist for a small office
Copy this and put a name next to each line.
- [ ] MFA on every mailbox, with no grace period for new staff. Administrators and whoever approves payments on passkeys or security keys.
- [ ] Legacy authentication blocked, after checking scanners and old mail apps.
- [ ] SPF and DKIM set up for every service that sends as your domain.
- [ ] DMARC at
p=rejecton your mail domain and on every domain you own that doesn't send mail. - [ ] Automatic forwarding to outside addresses turned off, with a named exception for anyone who needs it.
- [ ] Standard preset security policy on in Microsoft 365, or Gmail's Safety settings reviewed in Google Workspace.
- [ ] First contact safety tip on, and impersonation protection set up if you have Defender for Office 365.
- [ ] A written rule: payment detail changes are confirmed by phone to a number on file, and approved by a second person.
- [ ] Clients told in writing that your bank details won't change by email.
- [ ] Someone named who reads security alerts and knows where the audit log is.
- [ ] Staff know how to report a suspicious message and that reporting a mistake fast is what you want.
- [ ] Once a year, review inbox rules, allow lists and who has admin rights.
More in this topic
Crashcart runs IT and security for small offices for $35 a computer a month. Start a 14-day free trial.