Every carrier writes its own application, and the forms run anywhere from four pages to fifteen. Read enough of them and a pattern shows up: the same two dozen questions appear in different words, grouped under headings like “access controls” or “data protection.” Each one maps to a common way small firms get breached.

Below are the 25 that matter most. For each, you’ll find what the underwriter is testing, what a strong answer looks like, and the evidence to keep on file. The six that open the SabiSafe self-check (MFA, offline backups, EDR, exposed Remote Desktop, patching within 14 days and a rehearsed response plan) are marked, because those six decide more applications than the other nineteen put together.

Identity and access#

1. Do you require multi-factor authentication for email? (one of the six)#

The underwriter wants to know whether a stolen password alone can open a mailbox. Most business email compromise starts there. A strong answer is that MFA is enforced for every user through policy, with no exceptions or with each exception documented. Keep a screenshot of the conditional access or enforcement policy.

2. Do you require MFA for remote access?#

This covers VPNs, remote desktop gateways, remote management tools and cloud portals. Any door into your network from outside should need a second factor. List each remote access method and confirm MFA on each.

3. Do you require MFA for privileged and administrator accounts?#

Admin accounts can turn off security tools and delete backups, so insurers want the strongest protection on them. Some forms ask whether you use phishing-resistant MFA, such as hardware security keys, for these accounts.

4. Are administrator accounts separate from everyday user accounts?#

If a staff member reads email while logged in as an administrator, one malicious attachment gets admin rights. The right answer is that admins use a standard account for daily work and a separate account for administration.

5. How do you manage passwords?#

Expect questions about minimum length, reuse and whether you use a password manager. A firm-wide password manager and a policy against reuse is a good answer. Shared logins for the practice management system are a common weak spot.

6. How quickly do you remove access when someone leaves?#

Former staff with working accounts are a real exposure. Describe your offboarding steps and how long they take, ideally the same day.

Endpoints and network#

7. Do you use endpoint detection and response on all devices? (one of the six)#

EDR watches behavior and can isolate a device during an attack, which signature-based antivirus can’t do. The underwriter wants coverage on every workstation and server. Compare the EDR console’s device count against your inventory before you answer.

8. Are your EDR alerts monitored, and by whom?#

A tool that raises alerts nobody reads at 2 a.m. doesn’t stop ransomware. State who watches alerts and during which hours. For most small firms, round-the-clock coverage means a managed detection and response provider.

9. Is Remote Desktop Protocol exposed to the internet? (one of the six)#

The only good answer is no. RDP should be closed to the internet or reachable only through a VPN or gateway that requires MFA. A quick external scan of your public IP addresses confirms it.

10. How quickly do you apply critical security patches? (one of the six)#

Fourteen days for critical patches is a common benchmark. Internet-facing systems matter most: Verizon’s 2025 Data Breach Investigations Report found exploitation of vulnerabilities was the initial way in for 20 percent of breaches, and VPNs and other edge devices made up 22 percent of those exploitation cases. Keep a recent patch report.

11. Do you run unsupported software or operating systems?#

End-of-life systems no longer get security fixes. If you have one (an old server running a legacy accounting package is typical), say so and explain how it is isolated from the rest of the network.

12. Is your network segmented?#

Underwriters want to know whether one infected laptop can reach everything. Even basic separation helps, such as keeping guest Wi-Fi, printers and servers on different network segments.

13. Do you encrypt laptops and mobile devices?#

Full-disk encryption turns a lost laptop from a reportable breach into a replacement cost. BitLocker on Windows and FileVault on macOS are standard. Keep a report from your device management tool.

Email and web#

14. Do you filter incoming email for malicious attachments and links?#

Most firms on Microsoft 365 or Google Workspace have some filtering. Know what you have and whether attachments are scanned in a sandbox before delivery.

15. Have you set up SPF, DKIM and DMARC?#

These DNS records make it harder for criminals to send email that looks like it came from your domain. A DMARC policy at “quarantine” or “reject” is a stronger answer than “none.”

16. Do you tag or warn on email from outside the organization?#

A visible “external sender” banner helps staff catch messages that pretend to come from a colleague. It costs nothing to enable.

Backup and recovery#

17. Do you keep backups offline or immutable, and have you tested a restore? (one of the six)#

This is the question that decides ransomware coverage. Insurers want a copy of your data that an attacker can’t delete or encrypt, plus proof that you have restored from it. Keep a dated restore test showing what you restored and how long it took.

18. Are backup credentials separate from your main network accounts?#

If the same admin password that runs your network also manages your backups, an attacker who gets one gets both. Backup consoles should use separate credentials with MFA.

19. How long would it take to restore critical systems?#

The underwriter is estimating your business interruption exposure. Give a realistic number based on your restore test, rather than the vendor’s estimate.

People and process#

20. Do staff receive security awareness training?#

Annual training is the minimum on most forms. Keep completion records with names and dates.

21. Do you run phishing simulations?#

Many carriers now ask for regular simulated phishing in addition to training. Keep the results, including how click rates changed over time.

22. Do you verify changes to payment or wire instructions?#

This question protects the funds transfer fraud coverage. The right answer is a written procedure: any change to payment details is confirmed by phone, using a number already on file, before money moves. Accountants, title companies and law firms get the most scrutiny here.

23. Do you have a written incident response plan, and have you rehearsed it? (one of the six)#

Insurers ask whether the plan exists, who owns it and whether you have tested it with a tabletop exercise. Put the insurer’s breach hotline number in the plan. Many policies require you to call it before hiring your own forensic firm.

Data and vendors#

24. What sensitive data do you hold, and how much?#

Expect questions about records counts for personal data, health information, payment cards and financial account details. The numbers drive notification cost estimates. A rough inventory by system is enough.

25. Do you assess the security of vendors with access to your data?#

Verizon’s 2025 report found that breaches involving a third party doubled to 30 percent. Underwriters want to know whether your IT provider, cloud apps and other vendors with access use MFA and whether your contracts address security. A simple vendor list with each one’s access and MFA status is a good start.

How to answer when you aren’t there yet#

Most small firms have a few “no” or “partial” answers. That is normal, and underwriters expect it. What hurts you is an answer that turns out to be wrong after a claim.

When a control is incomplete, describe what is in place, what is left and the date it will be finished. For example: “MFA is enforced for all 22 users on email. The two remote access tools will require MFA by January 15.” That kind of answer often gets you a quote with a condition attached, where an unsupported yes risks the claim itself.

Keep one evidence folder

Save every screenshot, report and restore log you use to answer the form in a single folder named with the renewal date. Next year’s renewal, client security questionnaires and any forensic team will ask for the same material.

Where to start if the list feels long#

Not every question carries the same weight. If you can only fix a few things before renewal, work in this order: MFA on email and remote access, an immutable backup with a tested restore, EDR on every device, then closing any exposed Remote Desktop. Those four address the routes that most ransomware and email fraud cases follow, and they are the answers an underwriter reads first.

Questions people ask

Can I answer "partially" on a cyber insurance application?

Many forms allow it, and where they don't, you can usually add a note. A partial answer with a dated plan is safer than a yes you can't support, because inaccurate answers can give the insurer grounds to dispute a claim.

How long before renewal should I start?

Six to eight weeks. Some controls, such as rolling out EDR or setting up an immutable backup, take a few weeks to put in place and longer to show evidence for.

Do these questions differ for law firms, accountants and clinics?

The core 25 are the same. Law firms usually get extra questions about wire transfer verification and client data, accountants about tax season and client portals, and clinics about health records and HIPAA.

Sources

  1. SabiSafe, six questions insurers ask (self-check)
  2. Cyber Insurance in 2026, The Controls Underwriters Expect (Cyber Advisors, August 2026)
  3. Cyber Insurance Requirements, A 2026 Readiness Checklist (Consilien, July 2026)
  4. Verizon 2025 Data Breach Investigations Report, summary of findings

Keep reading

This guide is general information for small firms. It is not legal, insurance or compliance advice for your situation.