SOC 2 does not explicitly mandate penetration testing, but auditors routinely look for it as the strongest evidence that logical access and system monitoring controls are operating effectively. The relevant Trust Services Criteria are CC6.1, CC6.8 and CC7.2. For a Type II audit, a test within the 12-month audit period is expected, and findings must be triaged, remediated and re-tested, not just documented.
SOC 2 is often described as a flexible framework, and that is true in that it does not prescribe every security control an organisation must implement. But auditors are consistent practitioners with well-formed expectations, and penetration testing is near the top of the evidence list for the security category.
How does SOC 2 work?
SOC 2 reports on whether an organisation’s information security controls satisfy the AICPA Trust Services Criteria. The Security category, which is mandatory, is built around the Common Criteria (CC), a set of principles covering access controls, system operations, change management and risk assessment. The audit is either Type I (controls are suitably designed at a point in time) or Type II (controls operated effectively over a period, typically 12 months). Type II is the standard that customers and enterprise procurement teams request.
For Type II, the auditor tests whether controls operated as designed throughout the period, and that means examining evidence of real-world effectiveness, not just policy documents.
Which Trust Services Criteria does a SOC 2 penetration test address?
Penetration testing addresses multiple criteria, but three are central.
CC6.8 covers controls to prevent or detect, and act upon, the introduction of malicious software. This includes technical controls like antivirus and application control, but it also covers testing that those controls withstand real attack. A penetration test that attempts to execute malicious code, bypass endpoint controls and introduce persistence provides direct evidence for CC6.8.
CC7.2 covers procedures to monitor system components for anomalies that may indicate malicious acts. A penetration test that generates attack traffic and verifies that the monitoring and alerting systems detect it is direct evidence for CC7.2, specifically that detection controls operate as expected.
CC6.1 covers logical access controls, and a penetration test that attempts to bypass authentication, escalate privileges or access data beyond the tester’s authorised scope is evidence of how well the access controls hold up under attack. CC6.6, relating to access restrictions for external connections, is similarly addressed by an external penetration test.
What should be in scope for a SOC 2 penetration test?
Scope for a SOC 2 penetration test should match the SOC 2 boundary, which is the systems and services that are in scope for the audit. For a SaaS business, this typically means the customer-facing application, the underlying infrastructure (compute, storage, networking), the APIs, and supporting services such as authentication, identity management and data processing pipelines. Internal systems that process or store in-scope data may also be relevant.
Agree scope with the auditor during planning. Auditors may have a view on what constitutes adequate coverage for the specific criteria being tested, and mismatches between the audit scope and the pen test scope create gaps that show up in fieldwork.
The test should usually include:
- External penetration testing of internet-facing services
- Web application testing of the customer-facing application
- API security testing if APIs are a primary access mechanism
- Authentication and authorisation testing (privilege escalation, access control bypass)
- Cloud configuration review if the service runs on cloud infrastructure
Network-level internal testing is included when internal systems are in scope, particularly where the auditor is examining lateral movement risk.
How does a SOC 2 penetration test differ for Type I and Type II?
For Type I, a penetration test is still useful evidence but is less critical because the auditor is only asserting that controls are suitably designed at a single point in time. For Type II, the test needs to fall within the audit period, and the evidence needs to show the full cycle: the test was conducted, findings were triaged, critical issues were remediated, and that remediation was re-verified. A test done outside the audit period, or one where findings remain open, significantly weakens the evidence package.
Many organisations time the penetration test to fall midway through the audit period, giving time to remediate findings before the period closes.
What do auditors look for in SOC 2 penetration test results?
Auditors do not just want the penetration test report. They want to see the full cycle of response.
A good evidence package includes: the test report itself (scope, methodology, findings with risk ratings); the organisation’s management response to each finding, documenting the remediation plan and owner; evidence that critical and high findings were remediated within the committed timeframe; and a re-test report confirming that remediation was effective. Change management records tying the remediation to production changes are also useful.
A report where all findings are rated informational or low is sometimes treated with scepticism. Auditors understand that a thorough test often finds something, and a test that finds nothing may signal that the scope was narrow or the test was superficial. How findings were handled matters as much as the count.
How often should a SOC 2 penetration test run?
Annual penetration testing is the practical minimum for SOC 2. For organisations that deploy code frequently or operate on dynamic infrastructure, more frequent targeted testing is appropriate, particularly for high-risk components like authentication systems and data-handling APIs. A continuous vulnerability management programme running alongside the annual test is a strong evidence posture.
Remediation of critical and high findings within 30 days and medium findings within 90 days is a common expectation. These timeframes should be documented in the vulnerability management policy and evidenced in the audit artefacts.
How does a SOC 2 penetration test fit the broader security programme?
A penetration test for SOC 2 should not be a standalone exercise. The findings feed directly into the vulnerability management programme, the risk register, and the change management process. The remediation timelines and re-test results are evidence for CC4.1, the criterion covering ongoing monitoring activities, and they feed the quality of information used in risk decisions elsewhere in the Common Criteria. A penetration test done in isolation, reported and filed without being actioned, helps little.
Where an organisation is preparing for a first SOC 2 audit or an audit cycle renewal, Cybernion’s SOC 2 advisory service covers scoping, control design and pre-audit readiness, including penetration testing referrals and evidence management.
Frequently asked questions
Does SOC 2 require penetration testing?
SOC 2 does not contain an explicit requirement for penetration testing in the way PCI DSS does. However, the AICPA Trust Services Criteria, particularly CC6.8, CC7.2 and CC6.1, relate to preventing and detecting unauthorised access and evaluating whether controls are operating effectively. Auditors routinely look for penetration testing as the strongest evidence that these criteria are being met, especially for Type II reports covering a 12-month period. In practice, the absence of a penetration test is a common observation in SOC 2 audits.
Which SOC 2 criteria does penetration testing address?
Penetration testing primarily addresses CC6.8, which covers controls to prevent or detect and act upon the introduction of malicious software, and CC7.2, which covers system monitoring and response to anomalies. CC6.1, relating to logical access, and CC6.6, relating to access restrictions for external users, are also relevant. A penetration test provides evidence that these controls withstand real attack attempts, not just that they exist on paper.
How often should you run a penetration test for SOC 2?
For SOC 2 Type II, a penetration test conducted within the audit period, typically the 12 months covered by the report, is expected. An annual test is the practical minimum. If the audit period is shorter, the test should fall within it. Tests older than 12 months are typically not accepted as current evidence. For organisations with frequent code deployments or infrastructure changes, more frequent testing may be appropriate.
What should be in scope for a SOC 2 penetration test?
Scope typically includes the systems, applications and infrastructure that are in scope for the SOC 2 audit itself. For a SaaS business this usually means the customer-facing application, its underlying infrastructure, APIs, and supporting services like authentication and data storage. Internal systems may be in scope if they process, store or transmit cardholder, health or other sensitive in-scope data. The scope should be agreed with the auditor as part of the audit planning process.
What do auditors do with penetration test results?
Auditors review the penetration test report and look at how findings were handled. They want to see that findings were triaged, that critical and high findings were remediated promptly, that remediation was verified by a re-test, and that a management response was documented. A penetration test report full of open critical findings, or one where remediation was not verified, can lead to a qualified opinion. The test itself is not the evidence; the response to its findings is.
Sources:
- Trust Services Criteria, AICPA, 2022
- NIST SP 800-115, Technical Guide to Information Security Testing, NIST Computer Security Resource Center, 2008
Last updated: 30 August, 2026
