MongooseWeb

File

Edit

View

History

Bookmarks

Profiles

Tabs

Window

Help

Mon 15 Oct

13:37

PCI DSS Penetration Testing: What Tick-Box Tests Miss Under Requirement 11.4

Introduction

Every year, organisations that take card payments commission a penetration test because PCI DSS says they must. The report arrives, it goes into the evidence folder for the Qualified Security Assessor (QSA), and everyone moves on. The box is ticked.

The trouble is that Requirement 11.4 asks for far more than a report with "penetration test" on the cover. It asks for a defined methodology, coverage of the whole cardholder data environment (CDE) perimeter, testing from inside and outside the network, proof that segmentation actually works, application-layer testing against specific attack types, and evidence that what was found has been fixed and retested. A cheap, narrowly scoped test can produce a tidy document and still leave most of that untouched.

This article explains what PCI DSS v4.0.1 actually requires from penetration testing, where narrowly scoped tests tend to fall short, and how to commission a test that gives your QSA, your acquirer and your board real evidence.

Where things stand: PCI DSS v4.0.1

PCI DSS v4.0.1 is the only active version of the standard. The PCI Security Standards Council (PCI SSC) published it on 11 June 2024 as a limited revision of v4.0, with clarifications but no new or deleted requirements, and retired v4.0 on 31 December 2024. The requirements in v4.0 that had been "future-dated" became mandatory on 31 March 2025.

In other words, any assessment you face now is against v4.0.1 in full. If your testing approach was designed around v3.2.1 and has not been revisited, it is worth checking it against the current wording.

What Requirement 11.4 actually asks for

Requirement 11.4 sits within Requirement 11, "Test Security of Systems and Networks Regularly". Its headline is simple: external and internal penetration testing is regularly performed, and exploitable vulnerabilities and security weaknesses are corrected. The detail is where the work is.

11.4.1: A defined, documented methodology

The organisation must have a penetration testing methodology that is defined, documented and implemented. According to the standard, it must include:

  • Industry-accepted approaches. The guidance names examples such as OSSTMM and OWASP testing programmes.
  • Coverage of the entire CDE perimeter and critical systems. Not a sample of IP addresses chosen because they were easy to list.
  • Testing from both inside and outside the network.
  • Testing to validate any segmentation and scope-reduction controls.
  • Application-layer testing to identify, at a minimum, the vulnerability types listed in Requirement 6.2.4.
  • Network-layer testing covering all components that support network functions, as well as operating systems.
  • Review of threats and vulnerabilities experienced in the last 12 months.
  • A documented approach to assessing and addressing the risk posed by exploitable vulnerabilities found during testing.
  • Retention of test results and remediation results for at least 12 months.

Note who owns this. The methodology belongs to the entity being assessed, not just to the tester. Your QSA will expect to see that your organisation has defined how testing is done, and that the test you commissioned followed it.

11.4.2 and 11.4.3: Internal and external testing

Internal and external penetration testing must each be performed at least once every 12 months, and after any significant infrastructure or application upgrade or change. The test must follow the entity's defined methodology and be carried out by a qualified internal resource or qualified external third party, with organisational independence of the tester. The tester does not need to be a QSA or an Approved Scanning Vendor (ASV).

The applicability notes are specific about what "internal" means: testing both from inside the CDE and into the CDE from trusted and untrusted internal networks. A test run only from inside the payment network does not show whether the office network can reach it.

11.4.4: Fix it, then prove it

Exploitable vulnerabilities and security weaknesses found during testing must be corrected in line with the entity's own risk assessment (Requirement 6.3.1), and penetration testing must be repeated to verify the corrections. A report followed by a patching ticket is not enough. The standard expects a retest.

11.4.5 to 11.4.7: Segmentation and service providers

If you use segmentation to reduce the scope of your CDE, Requirement 11.4.5 says those controls must be penetration tested at least once every 12 months and after any change to segmentation controls or methods. The test must cover all segmentation methods in use and confirm that they are operational and effective and that they isolate the CDE from all out-of-scope systems.

For service providers, Requirement 11.4.6 raises that frequency to at least once every six months. Requirement 11.4.7 adds that multi-tenant service providers must support their customers' external penetration testing, either by giving customers evidence that comparable testing has been done on their subscribed infrastructure or by giving them prompt access to test it themselves.

Where tick-box tests fall short

None of the following is exotic. Each one is a gap between what a narrowly scoped test produces and what the requirement describes.

The scope is a list of IP addresses, not the CDE

Requirement 11.4.1 asks for coverage of the entire CDE perimeter and critical systems. A test scoped from a spreadsheet of external IP addresses, with no conversation about how card data flows, will miss systems that connect to the CDE, provide security services to it, or can affect its security. Jump hosts, management interfaces, logging servers, identity infrastructure and payment page hosting all deserve a place in the scoping discussion.

Getting scope right is the foundation of everything else. If the CDE is defined too narrowly, the test can pass while the real exposure sits just outside the line.

Segmentation is assumed, not tested

Segmentation is often the single biggest lever on PCI DSS scope. It is also a control that can quietly fail. A firewall rule added for a one-off project, a new VLAN, a cloud security group copied from another environment: any of these can open a path from an out-of-scope network into the CDE.

The standard's own explanation for Requirement 11.4.5 notes that many attacks have involved an attacker moving laterally from what an entity deemed an isolated network into the CDE. Segmentation testing means testing from each out-of-scope segment towards the CDE, covering every segmentation method in use, not one representative network. If your segmentation test is a short paragraph saying "no access observed" with no detail of where testing was performed from, ask for more.

Application testing does not reach the payment logic

Requirement 11.4.1 links application-layer testing to the attack types in Requirement 6.2.4. Those include injection attacks, attacks on data and data structures, attacks on cryptography, attacks on business logic (including abuse of APIs, client-side functionality, cross-site scripting and cross-site request forgery), and attacks on access control mechanisms.

Business logic and access control flaws are exactly the issues automated scanners struggle to find. Can a customer change the amount after the payment is authorised? Can one merchant account see another's transactions through the API? Can a refund be issued twice? These questions need a tester who understands the application and has time to work through it manually. We covered why in Why Most Web App Penetration Tests Miss Business Logic Flaws and API Penetration Testing: What Scans and UI Tests Miss.

The test ignores the last 12 months

Requirement 11.4.1 asks the methodology to include review and consideration of threats and vulnerabilities experienced in the last 12 months. If your organisation patched a critical flaw in a remote access appliance, suffered a phishing incident, or migrated a payment service to a new platform, the next test should reflect that. A test that runs the same scope and checklist as last year, regardless of what has happened since, does not meet that intent.

Significant change does not trigger a test

Annual testing is the minimum. Requirements 11.4.2 and 11.4.3 also require testing after any significant infrastructure or application upgrade or change. A new checkout flow, a re-platformed e-commerce site, a move to a new cloud provider or a new connection to a payment processor are all candidates. Organisations should define what "significant" means for them, write it down, and make sure change management triggers a test when the threshold is met.

Findings are fixed but never retested

Requirement 11.4.4 requires testing to be repeated to verify corrections. Without a retest, you have a list of vulnerabilities and a belief that they have been fixed. With a retest, you have evidence. Your QSA will usually want to see the second.

Vulnerability scans are mistaken for a penetration test

PCI DSS has separate requirements for vulnerability scanning. Requirement 11.3.1 covers internal scans at least once every three months, and since 31 March 2025 Requirement 11.3.1.2 has required those internal scans to be authenticated. Requirement 11.3.2 covers external scans by a PCI SSC Approved Scanning Vendor at least once every three months.

Scans and penetration tests do different jobs. The PCI SSC's Information Supplement: Penetration Testing Guidance (version 1.1, September 2017) explains how a penetration test differs from a vulnerability scan. A scan identifies potential weaknesses. A penetration test attempts to exploit them, chain them together and show what an attacker could actually reach. You need both.

What this means for the business

PCI DSS compliance is a condition of taking card payments, but the business case goes beyond the assessment. The CDE is where your customers' card data lives. A test that only satisfies the paperwork leaves the organisation exposed in the place it can least afford it.

A well-scoped test also saves money. Proving that segmentation works can keep large parts of the estate out of PCI DSS scope. Clear evidence of methodology, independence and retesting can make the QSA's job quicker. And finding a path into the CDE during a test is far cheaper than finding it during a forensic investigation.

How to commission a PCI DSS test that produces evidence

Before you ask for quotes, gather the following:

  • Your current CDE definition and data flow diagrams. Include systems that connect to the CDE or can affect its security, not just those that store card data.
  • Your segmentation design. List every segmentation method in use, such as firewalls, VLANs, cloud security groups and access control lists, and every out-of-scope network that should not reach the CDE.
  • Your penetration testing methodology. If you do not have one written down, ask your testing provider to help you document it against Requirement 11.4.1.
  • A list of significant changes and incidents in the last 12 months. This shapes what the test should focus on.
  • Your applications in scope. Payment pages, customer portals, APIs and back-office tools that handle card data, along with test accounts at each privilege level.
  • Your entity type. Whether you are assessed as a merchant or a service provider changes the segmentation testing frequency.

Then ask any provider these questions:

  • How will you test segmentation, and from where? Expect a clear answer covering each out-of-scope segment.
  • How will application testing address the attack types in Requirement 6.2.4? Look for manual testing of business logic and access control, not just a scanner run.
  • What does the report include? It should describe the methodology, scope, testing performed (including unsuccessful attempts), findings with evidence, and risk ratings your team can map to its own risk assessment.
  • Is a retest included, and how quickly can it be scheduled? Plan for remediation and retest before your assessment deadline, not after.
  • How is independence demonstrated? The tester should be organisationally independent from the people who manage the systems being tested.

As a CREST-accredited penetration testing company, Mongoose provides independent, qualified testing with a methodology and report designed for your QSA. You can read more on our PCI DSS penetration testing page.

Practical Insights

  1. Requirement 11.4 is more than an annual test. It covers methodology, scope, internal and external testing, segmentation, application testing, remediation, retesting and record keeping.
  2. The methodology is yours. Your organisation must define, document and implement it. Your tester should follow it.
  3. Scope the CDE, not a list of IP addresses. Include systems that connect to the CDE or can affect its security.
  4. Test segmentation from every out-of-scope network. Annually and after changes for merchants, every six months for service providers.
  5. Make sure application testing covers business logic and access control. These are named in Requirement 6.2.4 and are rarely found by scanners.
  6. Define "significant change" and let it trigger testing. Annual testing is a minimum, not a schedule.
  7. Always retest. Requirement 11.4.4 expects verified corrections, and your QSA will want the evidence.
  8. Keep scans and penetration tests separate in your mind. Quarterly ASV and authenticated internal scans do not replace a manual test.

Conclusion

PCI DSS v4.0.1 does not ask for a penetration test as a formality. It describes a programme: a defined methodology, coverage of the whole CDE, proof that segmentation holds, application testing against real attack types, and verified remediation.

A tick-box test can produce a report that satisfies a filing requirement. A properly scoped test tells you whether an attacker can reach your customers' card data, and gives your QSA, your acquirer and your board evidence they can rely on.

Don't wait for a real adversary to find the pathway. Contact our team today to discuss a tailored manual assessment of your cardholder data environment.

Sources

Cobras strike without warning. The mongoose strikes with intent.
‍
Don’t sit back and wait for the bite, it’s time to take the fight to the cobra.

Ready to flip the script? Let’s begin the hunt.

Request a confidential Consultation