Skip to content
Compliance Automation Tools

Home / FAQ

FAQ, read 12 Nov 2026

30 answers for teams choosing and running a compliance automation tool

How the scores workThe glossary

The answers

What is compliance automation software, and what does it not do?

It connects to your cloud, identity and HR systems, collects evidence and runs automated tests against your controls. It does not issue an attestation or certificate: an assessor still does that. NIST SP 800-137 (September 2011) frames continuous monitoring as visibility into whether deployed security controls are effective, and a platform only delivers that if its tests match controls you actually run. You still write policies, name owners and fix failing tests.

What is the difference between a framework and a control?

A framework is a published set of requirements, such as the 110 requirements of NIST SP 800-171 Rev. 2 that CMMC Level 2 uses. A control is one safeguard you run to meet them. NIST defines a security control as a safeguard or countermeasure that protects confidentiality, integrity and availability. One control, such as a quarterly access review, can answer requirements in several frameworks, which is why platforms keep a control layer between frameworks and evidence.

How many frameworks can one control set cover, and what limits it?

The plan often limits it before the control set does. Vanta's pricing page, read 12 Nov 2026, lists "One compliance framework" in its Essentials plan. Technical limits are scope and wording: CMMC Level 2 applies to systems handling controlled unclassified information, not your whole company, and one framework may want stricter evidence than another. Ask for the mapping export and count the controls that pass for framework A but need extra evidence for framework B.

How does a platform decide that a framework is supported, partial or none found?

On this site the call is made from the vendor's published pages and dated. Supported means the framework ships with controls and evidence mapped. Partial means it appears as a template, a mapping, a beta or a service without full control and evidence support. None found means no page read on 12 Nov 2026 showed it, and it scores 0 for that cell. The Framework Coverage Matrix at /matrix/ covers 20 frameworks this way.

What does continuous monitoring actually check?

It runs automated tests against the systems your connectors can see, for example whether multi-factor authentication is on for each identity provider account or whether a code repository requires review before merge. NIST's glossary defines continuous monitoring as maintaining ongoing awareness of information security, vulnerabilities and threats to support risk decisions. A passing test proves the setting, not the process around it, so keep manual evidence for judgment controls such as access review sign-off.

How often do automated tests run?

Each vendor sets its own schedule, and frequency is a choice, not a standard. NIST SP 800-137 ties monitoring to the organization's risk tolerance, so a daily test for a production access control and a monthly test for a policy review can both be correct. In a demo, ask for the last-run timestamp on three tests and for the interval per test type. A claim of "continuous" with no interval is a marketing phrase.

What is control drift?

Drift is a control that passed when you set it up and later stops meeting its test, for example a storage bucket made public after a deploy or an offboarded employee whose account stays active. Monitoring exists to catch drift between audits, not on the day the assessor arrives. NIST SP 800-137 describes the aim as ongoing assurance that implemented controls stay aligned with risk tolerance. Count how long a drift alert takes to reach a named owner.

Which integrations does a tool need for each category?

You need one for each system that holds evidence: cloud (AWS, Azure or Google Cloud), identity (Okta or similar), HR (the system that records joiners and leavers), code (GitHub or GitLab), ticketing (Jira) and endpoint (an MDM). Vanta's integrations page, read 12 Nov 2026, lists AWS with 40+ resources, Azure with 30+ and Google Cloud with 25+. Check depth, not count: what does the connector read from each resource? See the Integration Ledger.

Can an agent submit evidence to an auditor without a person?

Check the vendor's own words first. Vanta's pricing page, read 12 Nov 2026, lists "Agentic evidence collection" and an "Auditor API" but says nothing about unattended submission to an auditor. None found on 12 Nov 2026 is the honest reading. A control you can defend has a human approval step before agent output reaches an assessor and an audit trail of what the agent did. See human approval gate.

How do I test an agent claim in a demo?

Ask the presenter to run it on your data, not theirs. Pick one failing test, have the agent propose a fix, then ask four things: who approved the action, where the log entry is, what the agent could not see, and what happens when it is wrong. Vanta lists "Agentic issue management" and AI-generated code for failing tests in its plan table, so ask each vendor which plan tier includes the feature you watched.

Why do some vendors publish prices and others hide them?

Publishing a price invites comparison; hiding it lets a rep size the quote to your headcount. Secureframe's packages page, read 12 Nov 2026, shows Fundamentals as "Starting at $7,500/year" and sends Complete and Defense to "Get a quote". Vanta's page says to request a demo for "personalized pricing". On this site a published price figure is worth 6 of the 10 pricing points and published plan contents the other 4.

What is a trust center, and do I need one?

A trust center is a public or gated page where customers read your security documents, certifications and policies without emailing you. Vanta says it hosts 6,000+ Trust Center pages and that its tool automates access approvals and NDA collection, figures it marks "results may vary". You need one when sales keeps answering the same security questions. Keep it accurate: stale documents on a trust center are worse than none.

What does a risk register add to a compliance tool?

A control list says what you do. A risk register says why: each risk has a likelihood, an impact, an owner and linked controls. NIST SP 800-53 Rev. 5 is a catalog of security and privacy controls meant to protect against threats and risks, and a register is where you record which risks each control reduces. Secureframe's packages page, read 12 Nov 2026, lists risk management in its entry package. See risk register.

What does a vendor risk module add beyond a questionnaire spreadsheet?

It keeps an inventory of vendors, their documents and their review dates in one place, so renewals and reassessments do not depend on someone's calendar. Secureframe's packages page, read 12 Nov 2026, lists Vendor Inventory, Vendor Risk Assessment, annual and one-time reviews and a vendor portal for requesting questionnaires. It matters most for DORA, which asks financial entities for a register of their ICT third-party arrangements. See vendor risk assessment.

What does HIPAA require of a compliance tool?

HIPAA does not certify tools. The Security Rule requires covered entities and business associates to keep reasonable and appropriate administrative, physical and technical safeguards, and 45 CFR 164.316 requires written policies and documented actions kept in written (which may be electronic) form. HHS proposed changes on 6 Jan 2025 that would add a written technology asset inventory and network map, so a tool should export both. See HIPAA.

What does PCI DSS v4.0.1 require of a compliance tool?

PCI DSS v4.0.1 is a limited revision of v4.0 (published March 2022) with no new or deleted requirements. The PCI Security Standards Council retired v4.0 on 31 December 2024, and the future-dated requirements became effective on 31 March 2025. A tool should therefore carry the v4.0.1 text, not v4.0, and track the requirements that became mandatory on that date. Ask which version a vendor's mapping cites. See PCI DSS.

What do CMMC 2.0 Levels 1, 2 and 3 require?

Level 1 is a self-assessment against the 15 requirements of FAR clause 52.204-21 for federal contract information. Level 2 covers controlled unclassified information with the 110 requirements of NIST SP 800-171 Rev. 2, by self-assessment or by a third-party assessor. Level 3 adds 24 requirements from NIST SP 800-172 and is assessed by DCMA DIBCAC. The 32 CFR part 170 rule took effect on 16 December 2024. See CMMC.

What does NIS2 require of a compliance tool?

NIS2 (Directive 2022/2555) applies to medium-sized and large entities in 18 critical sectors, and requires risk-management measures and notification of significant incidents to national authorities. Member States had until 17 October 2024 to transpose it, and it repealed NIS1 from 18 October 2024. A tool cannot make you compliant, but it should record measures, owners and incident timelines per member state you operate in. See NIS2.

What does DORA require of a compliance tool?

DORA entered into force on 16 January 2023 and applies from 17 January 2025 to financial entities supervised by the three European supervisory authorities. Its pillars are ICT risk management, ICT third-party risk, resilience testing and incident reporting, and the authorities published standard templates for the register of information on ICT providers. A tool needs a vendor module that can produce that register. See DORA.

What does GDPR require of a compliance tool?

GDPR entered into force on 24 May 2016 and has applied since 25 May 2018. Article 33(1) requires a controller to notify the supervisory authority "not later than 72 hours after having become aware" of a breach, unless it is unlikely to result in a risk, and to give reasons for any delay. A tool should timestamp when you became aware and keep the decision record. See GDPR.

How long must evidence be kept for common frameworks?

It depends on the framework, so set retention per control. HIPAA documentation under 45 CFR 164.316(b)(2) must be kept 6 years from creation or last effective date, whichever is later. For CMMC, DoD kept a six-year artifact retention period in the final rule and declined a request to cut it to one year. Where a source gives no period, say none found and ask your assessor. See evidence.

Who owns a control in a compliance tool?

One named person owns each control, usually the engineer or manager who runs the activity, not the GRC lead who tracks it. CMMC makes the principle concrete: its rule defines an Affirming Official, the senior representative with authority to affirm continuing compliance each year. Assign an owner for every control before you connect integrations, then use the tool's task assignment to chase failing tests. See control owners and RACI.

What happens to my evidence if I change tools?

Plan the exit before you sign. Retention clocks keep running: HIPAA documentation must be kept 6 years and CMMC artifacts six years, so you may need years of records outside the old tool. Ask for a full export of evidence, test history and task logs in a readable format, and for how long the vendor keeps your data after cancellation. Vanta lists an Auditor API, so ask whether an exit export exists. See evidence that survives an audit.

How does this site score platforms?

The weights were fixed on 12 Nov 2026, before any score existed, and do not change afterward. Seven criteria add to 100: framework coverage 25, monitoring 20, agents 15, human expertise 15, risk and vendor modules 10, pricing 10 and proof 5. A vendor's own claim scores at most 7 of 10 until a second source confirms it. The full rules are on Method. Pricing is scored from the vendor's own pages, and none found scores 0.

Are the weights different for different situations?

Yes, and all five weight sets are on Best for. The index uses one fixed set of seven weights. Each situation page applies its own weights to the same evidence and, on framework pages, replaces the coverage criterion with that framework's own fit score. The evidence stays the same; only the weights move. Your situation may match none of the five, so read the criterion scores, not only the total.

What does none found mean in the Framework Coverage Matrix?

None found means no page read on the stated date showed that framework, and the cell is dated, for example 12 Nov 2026. It scores 0 for that element. It does not mean the framework is absent from the product, because a customer-only page or a sales call may cover it. If a framework you need shows none found, ask the vendor for a link and tell us. Read the grid at /matrix/.

Where do I download the data behind the scores?

The data downloads are on /datasets/. They are published under the Creative Commons Attribution 4.0 license, which lets you copy and adapt them for any purpose, even commercially, if you give appropriate credit. Cite the page and the read date of 12 Nov 2026, because scores change only when a sourced fact changes and each change is logged with that fact.

How many of the 20 matrix frameworks should I weigh when choosing?

Count only the frameworks you will be audited against in the next 24 months, then check each as supported, partial or none found. The matrix lists 20, from SOC 2 and ISO 27001 to CMMC 2.0, NIS2, DORA and TISAX, but the headline count carries at most 1 of the 10 coverage points here. Coverage that matches your audit calendar beats a large number. See frameworks.

What should I ask for in a demo to check evidence handling?

Ask to see one control from raw evidence to auditor view: the connector that pulled it, the timestamp, who reviewed it and what the auditor sees. NIST's definition of a security control is a safeguard, so evidence must show that the safeguard operated, not just that it exists. Request the audit trail of changes to the control. If evidence cannot be exported with its dates, your assessor may not accept it. See evidence collection.

Does a platform replace a GRC lead or an auditor?

No. It replaces collection and chasing, not judgment. The assessor still tests and signs, as CMMC Level 2 shows: a certified third-party assessor, not the software, performs the certification assessment. Your GRC lead still scopes the audit, decides risk appetite and answers the assessor's questions. A realistic goal is fewer hours collecting screenshots, so measure hours before and after. See measuring compliance work in hours.

Sources

  1. csrc.nist.gov/pubs/sp/800/137/final, read 12 Nov 2026
  2. csrc.nist.gov/glossary/term/continuous_monitoring, read 12 Nov 2026
  3. csrc.nist.gov/glossary/term/security_control, read 12 Nov 2026
  4. federalregister.gov/documents/2024/10/15/2024-22905/cybersecurity-maturity-model-certification-cmmc-program, read 12 Nov 2026
  5. vanta.com/pricing, read 12 Nov 2026
  6. nist.gov/cyberframework, read 12 Nov 2026
  7. vanta.com/integrations, read 12 Nov 2026
  8. secureframe.com/pricing, read 12 Nov 2026
  9. vanta.com/products/trust-center, read 12 Nov 2026
  10. csrc.nist.gov/pubs/sp/800/53/r5/upd1/final, read 12 Nov 2026
  11. esma.europa.eu/esmas-activities/digital-finance-and-innovation/digital-operational-resilience-act-dora, read 12 Nov 2026
  12. ecfr.gov/api/renderer/v1/content/enhanced/current/title-45?part=164&section=164.316, read 12 Nov 2026
  13. federalregister.gov/documents/2025/01/06/2024-30983/hipaa-security-rule-to-strengthen-the-cybersecurity-of-electronic-protected-health-information, read 12 Nov 2026
  14. blog.pcisecuritystandards.org/just-published-pci-dss-v4-0-1, read 12 Nov 2026
  15. digital-strategy.ec.europa.eu/en/policies/nis2-directive, read 12 Nov 2026
  16. commission.europa.eu/law/law-topic/data-protection/legal-framework-eu-data-protection_en, read 12 Nov 2026
  17. edpb.europa.eu/system/files/documents/2023-04/edpb_guidelines_202209_personal_data_breach_notification_v2.0_en.pdf, read 12 Nov 2026
  18. creativecommons.org/licenses/by/4.0, read 12 Nov 2026

Last reviewed 12 Nov 2026.