Guide
The cyber insurance questionnaire, question by question
Every insurer's application is different, and forms change from year to year. But the security section of most applications checks the same handful of controls. For each one below: what the question is really checking, what evidence shows it, where to find that evidence if you run Microsoft 365, and where small firms get caught out.
Read your own form's wording closely. Words like "all users", "remote access" and "privileged accounts" decide what a true answer is. Questions about what your policy covers belong with your broker.
1. Multi-factor authentication
What it checks: whether a stolen password alone can get someone into your email, remote access and administrator accounts.
Evidence: the policy that requires it, and a report of each person's registered sign-in methods.
In Microsoft 365: a Conditional Access policy (or security defaults) in Microsoft Entra ID, and the user registration details report.
Where firms get caught: one old account without it, a shared mailbox people sign in to directly, or older email apps that skip the second step.
2. Administrator accounts
What it checks: whether the accounts that can change everything are few, protected and kept apart from everyday work.
Evidence: a list of everyone holding an administrator role.
In Microsoft 365: the role assignments in Microsoft Entra ID.
Where firms get caught: the owner's everyday email account is also the top administrator.
3. Protection on every computer
What it checks: whether every device that touches firm data runs protection software that can detect and stop an attack.
Evidence: the device list matched against protection status, device by device.
In Microsoft 365: Microsoft Defender for Business, which comes with Business Premium, and the device list in Intune.
Where firms get caught: the laptop someone works from at home that was never enrolled.
4. Updates and unsupported software
What it checks: whether security updates are installed promptly, and whether anything runs on software its maker no longer supports.
Evidence: the update policy, update status per device, and a list of operating system versions.
In Microsoft 365: Windows update policies and reports in Intune.
Where firms get caught: a Windows 10 computer still in use (Microsoft ended support on October 14, 2025), or an old server nobody thinks about.
5. Encryption
What it checks: whether a lost or stolen laptop gives away client data.
Evidence: encryption status per device.
In Microsoft 365: BitLocker, managed and reported through Intune. Windows Pro includes it.
Where firms get caught: laptops set up before anyone turned encryption on.
6. Email security
What it checks: whether dangerous links and attachments are screened, and whether others can easily send mail that looks like it came from you.
Evidence: your filtering policies and your domain's email authentication records (SPF, DKIM and DMARC).
In Microsoft 365: the Defender for Office 365 policies that come with Business Premium, and your domain's DNS records.
Where firms get caught: a DMARC record that only monitors and was never moved on to block or quarantine.
7. Backups
What it checks: whether you could get your data back after ransomware, including a copy the attacker can't reach, and whether anyone has proved a restore works.
Evidence: what the backup covers, how often it runs, where the copies are kept, and a dated record of a restore.
In Microsoft 365: keeping deleted items for a while is not the same as a separate backup. Most firms need a backup service for Microsoft 365 data and for their computers.
Where firms get caught: backups exist, but nobody has ever restored anything from them.
8. Remote access
What it checks: whether anyone can reach your systems from the internet without strong sign-in.
Evidence: how staff connect from outside, and an outside look at what your systems expose.
Where firms get caught: a remote desktop connection opened for a vendor years ago and never closed.
9. Training
What it checks: whether staff can spot a phishing email or a fake invoice, and whether you can show it.
Evidence: dated training records: who attended and what was covered. Practice phishing emails, if you run them.
In Microsoft 365: Microsoft's practice phishing tool needs an add-on that Business Premium doesn't include.
Where firms get caught: "we train every year" with no record of who, when or what.
10. Payment and wire changes
What it checks: whether someone confirms a change to bank details by calling a number they already know, not one in the email.
Evidence: a written rule, and staff confirming they know it.
Where firms get caught: the rule exists only in one person's head.
11. Incident response plan
What it checks: whether you know who to call and what to do first when something goes wrong.
Evidence: a written plan with current names and numbers, including how to report to your insurer, and the date it was last reviewed.
Where firms get caught: the plan lists people who left two years ago.
12. Outside access
What it checks: who outside the firm can get into your systems: your IT provider, software vendors, guests.
Evidence: a list of external accounts, guest users and connected apps.
In Microsoft 365: guest users and enterprise applications in Microsoft Entra ID.
Where firms get caught: a former IT provider that still has an administrator account.
Yes, no, or yes with exceptions
A control that's in place for most people but not all is not a plain yes. Write down what's true, including the exceptions, and talk to your broker before you sign. Fixing a gap first is often quicker than people expect.
Want each question checked against your systems?
We review the security questions on your application against your Microsoft 365 and Windows computers, fix agreed gaps, and give you dated evidence. Done remotely, from $995.
