Skip to main content
Compliance guideReviewed

PCI DSS compliance for a business that takes cards online: who decides it, and what a scan or a test can show

PCI DSS is the PCI Security Standards Council's standard for protecting payment account data.

This guide sets out what the Council's own open pages say about who the standard covers, who decides how a merchant validates, what a self-assessment questionnaire is, and what an ASV scan report does and does not show. It then sets out, in this library's own terms, how such a scan differs from a penetration test, marks where a website or web application test can contribute evidence and where that evidence stops, and comes with a responsibility matrix for the records the merchant, the scanning vendor, the assessor and whoever fixes the findings each hold.

SAQ-aware PCI DSS responsibility matrix for merchants, scanning vendors, assessors and remediation providers

Fourteen rows, R1 to R14, set out who holds each PCI compliance record, who passes it to whom, and who answers each question: the merchant, its compliance-accepting entity (typically an acquirer or a payment brand), the Approved Scanning Vendor, the Qualified Security Assessor or other assessor, the tester, and the remediation provider. Rows R1 to R5 cover validation, the SAQ, payment pages, scope and service providers; R6 and R7 the ASV scans; R8 to R10 penetration testing, for which none of the Council pages this guide cites names a requirement; R11 and R12 change; R13 and R14 the assessor's pack and the submission.

How to use it

The fourth column gives what a PCI Security Standards Council open page says, naming the page and any date it carries; where none covers the point, the cell says so and sends you to the standard's text in your own copy and to your assessor, and R12, this guide's own check, cites no Council page. The matrix never says which SAQ applies to you or that a requirement is met. R12 alone sends requests, with its limits written into the row: your own systems, a staging copy where one exists, test customers with invented orders, no card number and no payment, six steps with the two sign-ins counted, and no password, key, token or link that opens an order or file for whoever holds it in your notes. Each of the other rows only reads or writes records. In the Council's FAQs, your compliance-accepting entity determines how compliance is validated and evidenced, and the assessment is performed by you or your assessor and documented on the Council's forms; this matrix decides none of it.

14 rows

SAQ-aware PCI DSS responsibility matrix for merchants, scanning vendors, assessors and remediation providers

Row and recordWho holds it, and the handoffWho answers the questionWhat the Council's open pages sayWhat to keepWhat a website or web application test can supply, and what it cannotLimits of this row
R1. How your PCI DSS validation is setThe merchant keeps it. Handoff: the merchant asks its acquirer (merchant bank) or the payment brand in writing, and the answer comes back to the merchant.The compliance-accepting entity, typically a payment brand or acquirer (FAQ 1473). FAQ 1126 sends merchants to the acquiring financial institution they hold a merchant agreement with.PCI DSS page: whether an entity is required to comply with or validate compliance is at the discretion of organisations that manage compliance programs, such as a payment brand, acquirer or other entity. FAQ 1473 (March 2023): compliance-accepting entities determine validation and reporting methods, including how compliance is to be evidenced, for example through a ROC or an SAQ.The written question and answer with their dates; who answered; whether validation is required; the form named (SAQ or ROC); any direction about which requirements to include; the PCI DSS version named; and the merchant agreement clause the answer relies on.Nothing. A test report does not decide how validation is set, and this row needs no test.Read-only row: correspondence only, with nothing sent to any system.
R2. Which SAQ, if any, and its versionThe merchant, from the answer in R1. Handoff: the merchant gives the questionnaire and its version to whoever completes or reviews it, including a QSA if one assists.The compliance-accepting entity, the entity the SAQ is submitted to (FAQ 1215). This matrix does not say which SAQ applies.FAQ 1215 (April 2024): there are several SAQs, each for a type of environment defined in its eligibility criteria; all the criteria for a particular SAQ must be met to use it; merchants should consult their compliance-accepting entity on whether they are eligible or required to submit an SAQ, and which one. FAQ 1316 (January 2015): the merchant or service provider, or their QSA, performs the Expected Testing each SAQ describes.The SAQ named in writing and its version; its eligibility section, read in your own copy, with a note against each criterion of the facts it rests on; who performs the Expected Testing; and the date.Nothing that settles eligibility. The payment page list in R3 is your own description of the store, to take to the conversation in R1.Read-only row: it reads documents and sends no requests.
R3. Payment pages and the route to the payment providerThe merchant, with its developer. Handoff: the merchant gives the list to the ASV for R6, to the tester for R8 and to the assessor for R13.The assessor validates that the scope of an assessment is accurately defined and documented (FAQ 1473).Glossary: a payment page is a web-based user interface with one or more form elements intended to capture account data from a consumer or submit captured account data, for purposes of processing and authorising payment transactions. FAQ 1332 (July 2015): where a merchant meets all the SAQ A criteria, its web server stays in scope so the assessor can examine its configuration and verify the redirection mechanism. FAQ 1604 (June 2026): SAQ A's ASV scanning reaches merchant webpages that redirect transactions to a third-party service provider or include its embedded iframe.One line for each website, subdomain and application: its address; who builds it; who hosts it; whether at the payment step it redirects customers to the provider, embeds the provider's page or form, or does something else; and the date of its last change. Scripts on those pages are recorded with the PCI payment script controls guide, not here.A website security check's report lists hostnames, including ones you may not know about, and a web application test's report lists the routes, roles and accounts tested. Either is a dated input to this list. Neither decides what is in scope.Read-only row: it is written from your own records and your developer's, and sends no requests.
R4. Scope boundary and segmentationThe merchant documents it. Handoff: to the assessor in R13.The assessor validates scope (FAQ 1473). FAQ 1135 encourages contacting a QSA to help scope the cardholder data environment.Glossary: scoping identifies all system components, people and processes to be included in a PCI DSS assessment; the CDE is the system components, people and processes that store, process or transmit cardholder data and/or sensitive authentication data, and system components with unrestricted connectivity to those; segmentation isolates what stores, processes or transmits cardholder data from what does not; the scoping and segmentation entries point to section 4 of the standard. FAQ 1135 (July 2012): in general, adequate network segmentation can reduce the scope of the assessment if it isolates those systems, and the Council says it is not able to offer an opinion on how an organisation achieves adequate segmentation.A diagram or list of the systems that store, process or transmit account data, the systems connected to them, and those you treat as outside the boundary with the control that separates them; who drew it and when; and what section 4 of the standard says, read in your own copy. The Council's scoping and segmentation supplement, announced on 9 September 2024, is guidance that the announcement says does not supersede or replace any PCI standard.Nothing about segmentation, which a Clever Ops test does not examine; Clever Ops refers internal networks to a network specialist. The hostname and route lists from R3 show what answered from the internet on the dates tested.Read-only row: it draws on your own records, sends no requests and asks for no test of segmentation.
R5. The payment provider and other service providersThe merchant. Handoff: each provider supplies its Attestation of Compliance and the agreement to the merchant, and the merchant passes them to the assessor in R13.The merchant, for the responsibilities FAQ 1092 lists; the compliance-accepting entity, for what it accepts.FAQ 1092 (June 2025): a merchant that outsources payment processing remains responsible for ensuring the provider is PCI DSS compliant for the services offered, maintaining written agreements that include acknowledgement of their responsibilities (Requirement 12.8.2), monitoring the provider's compliance status at least annually (Requirement 12.8.4), and clearly defining and understanding shared responsibilities. FAQ 1568 (April 2023): the AOC is intended to be shared externally to requesting entities, according to applicable Participating Payment Brand rules and as noted in the QSA Program Guide.For each provider: the service it provides; its AOC, with the date and the services it covers; the agreement clause on responsibilities; the date you last checked its compliance status; and your split of shared responsibilities, as you understand it after reading Requirements 12.8.2 and 12.8.4 in your own copy.Nothing about the provider: its own product and platform, hosted checkout and hosted pages included, are outside what Clever Ops tests. A test of your own store can show what your side sends to the provider and accepts back, on the build tested and where the provider's terms allow it.Read-only row: it reads agreements and attestations and sends no requests. Read each provider's terms before any test touches a page or address it runs, and ask the provider about anything they leave open.
R6. External ASV scans under Requirement 11.3.2The ASV produces the scan reports and the merchant keeps them. Handoff: the merchant gives the ASV the hosts in scope from R3, the ASV returns its reports, and the merchant passes them on where the compliance-accepting entity asks.The ASV performs the scan with its approved scan solution. Reporting: the compliance-accepting entity (FAQ 1234).ASV page: an ASV conducts external vulnerability scanning to validate adherence with the external scanning requirements of Requirement 11.3.2, with a scan solution the Council tests and approves. Resource guide post (10 July 2024): 11.3.2 requires evidence of passing external scans, performed by an ASV, at least once every three months; in SAQ A, the scans apply only to the merchant system that hosts the webpage that redirects to, or embeds a payment page or form from, a PCI DSS compliant provider. FAQ 1604 (June 2026): scans for 11.3.2 must be performed by a listed ASV using its scan solution. FAQ 1152 (January 2024): among its characteristics, a passing external scan generally shows no vulnerabilities scoring 4.0 or higher on CVSS. FAQ 1234 (June 2025) names Requirement 11.3.2.1 for quarterly external scans by an ASV, and FAQ 1604 says 11.3.2 and 11.3.2.1 were both added to SAQ A: read both in your copy.The ASV's name, the date you checked it on the Council's list, the hosts given to it, each scan date and result, the report on the ASV's official template, and any supplemental document it sent, marked as supplemental. FAQ 1234 says supplemental materials have not been endorsed by the Council and should not be considered replacements for its templates.Nothing that stands in for an ASV scan. Clever Ops is not an Approved Scanning Vendor, and a website check or web application test is not a scan under Requirement 11.3.2.Read-only row: the scans are the ASV's work under its own agreement with you, and this row sends nothing.
R7. ASV scan findings: remediation and rescansThe remediation provider (your developer, agency or host) fixes, the ASV rescans, and the merchant keeps both records. Handoff: merchant to remediation provider with the scan finding; remediation provider back to the merchant with the change record; merchant to the ASV for the rescan.The ASV, through its rescan result. The merchant decides who fixes each finding and by when.FAQ 1152 (January 2024): among the things that should occur to verify that the requirement to perform vulnerability scans at least once every three months has been met, rescans are performed as necessary and show that vulnerabilities identified in the initial scans have been remediated, for all affected systems, as part of that period's scanning process. FAQ 1087 (July 2023): to meet the scanning requirements in Requirement 11, an entity is required to complete its internal and external scans, and any required remediation, at least once every three months.For each finding the ASV reported: its owner, the fix and its date, the build or configuration change that carried it, and the rescan date and result.Nothing that replaces the rescan, which stays with the scanning vendor. Where a separate penetration test reported the same weakness, keep that finding and its retest status beside this record.Read-only row: it records fixes and results and sends no requests.
R8. Penetration testing: the scope given to the testerThe merchant writes the scope with the tester. Handoff: merchant to tester (scope, test accounts, environment and signed authorisation), then tester to merchant (the report, in R10).Your assessor or acquirer, on which requirement in your version of the standard covers penetration testing; your assessor, on whether the work meets it, once you have read that requirement there.Not on the open pages read for this guide: none names a penetration testing requirement or says what one asks. Read the requirement your assessor or acquirer names for penetration testing, with its testing procedures and guidance, in your copy of PCI DSS. FAQ 1333 (July 2015): compliance cannot be determined by testing only pre-production environments using test data.The requirement your assessor or acquirer named for penetration testing, with its number as they gave it; the version of the standard you read it in, and where its text sits in that copy; what your assessor said they expect to see; the scope agreed in writing, listing targets, roles, test accounts, environment, depth, exclusions and the testing window; each target owner's signed authorisation; the date the host's testing policy was read; and whether testing ran on production or on a staging copy.Dated evidence about one application on the build tested: the scope as tested and, from a web application test, the test cases that could not run and why. Not a finding that any PCI DSS requirement is met.Read-only row: it gathers documents before a test and sends no requests; the tester's signed authorisation and rules of engagement govern the test.
R9. The tester's relationship to the systemThe merchant, from the tester's written statement. Handoff: tester to merchant, then merchant to assessor in R13.Your assessor, on whether the tester is acceptable for the work. Whether PCI DSS says anything about who may perform penetration testing is not on the open pages read for this guide.The PCI DSS page and the QSA page describe QSA companies as independent security organisations qualified by the Council. Neither says anything about penetration testers. FAQ 1598 (June 2025): entities should follow their internal third-party due diligence processes when engaging an ASV or QSA company.The tester's name and company; whether it had any hand in building, hosting or running the system, or operates a control under test; any qualification or independence your copy of the standard, your assessor or your acquirer asks for, in their exact words; and who approved the choice, and when.The tester's own written statement. For a system Clever Ops built or hosts, the report identifies both the builder and the host, and the work is described as a security review, which is not independent testing.Read-only row: it notes who was chosen and why, and sends no requests.
R10. Penetration test report, remediation and retestThe tester issues the report and the retest, the remediation provider fixes, and the merchant keeps all three. Handoff: tester to merchant (report); merchant to remediation provider (findings); remediation provider to merchant (fixes on an identified build); merchant to tester (retest request); tester to merchant (retest result).The merchant decides what to fix, by when and who fixes it. Whether the record meets a PCI DSS requirement is for your assessor.None of the open pages read for this guide says anything about penetration test findings. Read the requirement your assessor or acquirer names for penetration testing in your copy of the standard for anything it says about findings and retesting.The report's version and issue date; which build was tested and on what dates; every finding with its ratings and the person who owns it; the fix, and the build it shipped in; when the retest ran; and the status each finding ended with.From a Clever Ops report: each finding with redacted evidence, its fix and the criteria the retest will apply, and, where the finding is rated, a CVSS v3.1 score and a business-risk rating; the checks that found nothing, limited to the build and accounts tested; and the fixes in priority order. Following the one consolidated retest, the report shows each finding as fixed, partially fixed, still reproducible or not retested. A change shipped after the retest is not covered.Read-only row: it files reports and decisions and sends no requests. The full report explains how to reproduce any finding still open, so keep its circulation to those who need it.
R11. Significant change logThe merchant. Handoff: the merchant tells the ASV (R6) and the tester (R8) of any change it evaluates as significant, and gives the log to the assessor in R13.FAQ 1317 says each listed activity must be considered and evaluated to determine whether a change is significant for that entity; it does not say who does the evaluating, so record who did.FAQ 1317 (April 2023), giving the significant change list in PCI DSS v4.0: new hardware, software or networking equipment added to the CDE; replacement or major upgrades of hardware and software in the CDE; changes in the flow or storage of account data; changes to the boundary of the CDE and/or to the scope of the assessment; changes to the underlying supporting infrastructure of the CDE, including, but not limited to, directory services, time servers, logging and monitoring; and changes to third party vendors or service providers (or services provided) that support the CDE or meet PCI DSS requirements on the entity's behalf. FAQ 1087 (July 2023): vulnerability scans are required after a significant change as well as at least once every three months.Each change to the store, its payment step, its hosting or its providers; its date; who evaluated it against the list, when, and the outcome; and which scans, tests or checks followed, with their dates.A dated test or check after a change, where one followed. Not the evaluation itself.Read-only row: it records changes and decisions and sends no requests.
R12. Access check after a change to the storeThe merchant's own team or developer runs it. Handoff: the results go into R11 against the change that prompted them.The merchant decides whether a request that was not refused means bringing the next test forward.None: this row is this guide's own narrow check between tests, not a step any Council page describes. It is not a scan under Requirement 11.3.2 or a penetration test.The change and its release date, then for each step the test account used (or none), the date and time, the status returned and whether any of test customer A's details came back.Nothing from a tester: your own team runs it. Refusals at steps 4, 5 and 6 show that those three requests were refused on that build, and nothing more.Limits for this row, the only one in the matrix that sends anything. Run it against a store or application that belongs to your business, on a staging copy wherever you have one. Use only test customers whose names, addresses and orders were invented for testing, never a real customer's account. No card number, real or test, goes into any request or any note, and no payment is submitted: apart from the two sign-ins, each step only reads an order, a page or a file. Take the six steps listed at the end of this cell, once each and one after another, and nothing more, and put no load on the store. Any email or text the store sends in response has to land in an inbox or on a number your business controls. Your notes hold no password, key or session token, and no link that would open an order or a file for anyone holding it; open a link of that kind only for a test order. If a vendor runs any part of the store, read its terms before you start, ask the vendor about anything the terms leave unclear, and stay within the terms and its written answer. If any response shows a real person's details, halt at that point, log the page address and the time, look at nothing more, and report it to the person who handles privacy and security incidents for your business. The six steps, the two sign-ins included, each done once. 1: sign in as test customer A. 2: as A, open the page of a test order that belongs to A on the staging copy (one your developer created, or one left from an earlier test) and note its order number, writing down the page address only if it carries no key or token. 3: in a separate private window, sign in as test customer B. 4: as B, open A's order page by that order number. 5: as B, request the invoice or receipt file for the same order. 6: in a new private window that holds no session, open A's order page by its order number, signed out.
R13. The assessor's evidence packThe merchant assembles it. Handoff: merchant to the QSA or other assessor, who returns the assessment documents it completes.The assessor, on scope, applicability and whether each requirement is in place; the compliance-accepting entity, on what it accepts.QSA page: QSA companies are independent security organisations qualified by the Council to validate an entity's adherence to PCI DSS, re-qualified each year; check the list when you engage one. FAQ 1473 (March 2023): the assessor validates that scope and applicability are accurately defined and documented, and must confirm through testing that a requirement is truly not applicable before reporting it as Not Applicable.An index of R1 to R12 for the period: the validation answer, the SAQ or ROC and its version, the payment page list, the scope and segmentation record, provider attestations, ASV scan reports and rescans, penetration test reports and retests, the tester's statement and the change log, with where each is kept.The test reports and retests filed in R10, for the assessor to weigh. Not an assessment, and not a decision on scope or applicability.Read-only row: it assembles records and sends no requests.
R14. Submission to the compliance-accepting entityThe merchant submits and the compliance-accepting entity receives. Handoff: merchant to acquirer or payment brand, on the Council's forms.The compliance-accepting entity, on how compliance is to be evidenced and reported (FAQ 1473).Glossary: the AOC is the official PCI SSC form for merchants and service providers to attest to the results of a PCI DSS assessment, as documented in an SAQ or ROC. Post of 2 September 2025, quoting FAQ 1220: only the official form documents from the Council's website are recognised for PCI DSS validation, and other certificates are not acceptable for evidencing compliance. FAQ 1234 (June 2025): acquirers and/or payment brands may also request quarterly ASV scan reports.What was submitted, on which official form and version, the date, to whom, and any reply.Nothing that goes into this submission as a validation document. A test report is not one of the Council's official forms.Read-only row: it logs what was lodged and sends nothing to any system.

Row and record

Who holds it, and the handoff

  1. R1. How your PCI DSS validation is set
    Who holds it, and the handoff
    The merchant keeps it. Handoff: the merchant asks its acquirer (merchant bank) or the payment brand in writing, and the answer comes back to the merchant.
    Who answers the question
    The compliance-accepting entity, typically a payment brand or acquirer (FAQ 1473). FAQ 1126 sends merchants to the acquiring financial institution they hold a merchant agreement with.
    What the Council's open pages say
    PCI DSS page: whether an entity is required to comply with or validate compliance is at the discretion of organisations that manage compliance programs, such as a payment brand, acquirer or other entity. FAQ 1473 (March 2023): compliance-accepting entities determine validation and reporting methods, including how compliance is to be evidenced, for example through a ROC or an SAQ.
    What to keep
    The written question and answer with their dates; who answered; whether validation is required; the form named (SAQ or ROC); any direction about which requirements to include; the PCI DSS version named; and the merchant agreement clause the answer relies on.
    What a website or web application test can supply, and what it cannot
    Nothing. A test report does not decide how validation is set, and this row needs no test.
    Limits of this row
    Read-only row: correspondence only, with nothing sent to any system.
  2. R2. Which SAQ, if any, and its version
    Who holds it, and the handoff
    The merchant, from the answer in R1. Handoff: the merchant gives the questionnaire and its version to whoever completes or reviews it, including a QSA if one assists.
    Who answers the question
    The compliance-accepting entity, the entity the SAQ is submitted to (FAQ 1215). This matrix does not say which SAQ applies.
    What the Council's open pages say
    FAQ 1215 (April 2024): there are several SAQs, each for a type of environment defined in its eligibility criteria; all the criteria for a particular SAQ must be met to use it; merchants should consult their compliance-accepting entity on whether they are eligible or required to submit an SAQ, and which one. FAQ 1316 (January 2015): the merchant or service provider, or their QSA, performs the Expected Testing each SAQ describes.
    What to keep
    The SAQ named in writing and its version; its eligibility section, read in your own copy, with a note against each criterion of the facts it rests on; who performs the Expected Testing; and the date.
    What a website or web application test can supply, and what it cannot
    Nothing that settles eligibility. The payment page list in R3 is your own description of the store, to take to the conversation in R1.
    Limits of this row
    Read-only row: it reads documents and sends no requests.
  3. R3. Payment pages and the route to the payment provider
    Who holds it, and the handoff
    The merchant, with its developer. Handoff: the merchant gives the list to the ASV for R6, to the tester for R8 and to the assessor for R13.
    Who answers the question
    The assessor validates that the scope of an assessment is accurately defined and documented (FAQ 1473).
    What the Council's open pages say
    Glossary: a payment page is a web-based user interface with one or more form elements intended to capture account data from a consumer or submit captured account data, for purposes of processing and authorising payment transactions. FAQ 1332 (July 2015): where a merchant meets all the SAQ A criteria, its web server stays in scope so the assessor can examine its configuration and verify the redirection mechanism. FAQ 1604 (June 2026): SAQ A's ASV scanning reaches merchant webpages that redirect transactions to a third-party service provider or include its embedded iframe.
    What to keep
    One line for each website, subdomain and application: its address; who builds it; who hosts it; whether at the payment step it redirects customers to the provider, embeds the provider's page or form, or does something else; and the date of its last change. Scripts on those pages are recorded with the PCI payment script controls guide, not here.
    What a website or web application test can supply, and what it cannot
    A website security check's report lists hostnames, including ones you may not know about, and a web application test's report lists the routes, roles and accounts tested. Either is a dated input to this list. Neither decides what is in scope.
    Limits of this row
    Read-only row: it is written from your own records and your developer's, and sends no requests.
  4. R4. Scope boundary and segmentation
    Who holds it, and the handoff
    The merchant documents it. Handoff: to the assessor in R13.
    Who answers the question
    The assessor validates scope (FAQ 1473). FAQ 1135 encourages contacting a QSA to help scope the cardholder data environment.
    What the Council's open pages say
    Glossary: scoping identifies all system components, people and processes to be included in a PCI DSS assessment; the CDE is the system components, people and processes that store, process or transmit cardholder data and/or sensitive authentication data, and system components with unrestricted connectivity to those; segmentation isolates what stores, processes or transmits cardholder data from what does not; the scoping and segmentation entries point to section 4 of the standard. FAQ 1135 (July 2012): in general, adequate network segmentation can reduce the scope of the assessment if it isolates those systems, and the Council says it is not able to offer an opinion on how an organisation achieves adequate segmentation.
    What to keep
    A diagram or list of the systems that store, process or transmit account data, the systems connected to them, and those you treat as outside the boundary with the control that separates them; who drew it and when; and what section 4 of the standard says, read in your own copy. The Council's scoping and segmentation supplement, announced on 9 September 2024, is guidance that the announcement says does not supersede or replace any PCI standard.
    What a website or web application test can supply, and what it cannot
    Nothing about segmentation, which a Clever Ops test does not examine; Clever Ops refers internal networks to a network specialist. The hostname and route lists from R3 show what answered from the internet on the dates tested.
    Limits of this row
    Read-only row: it draws on your own records, sends no requests and asks for no test of segmentation.
  5. R5. The payment provider and other service providers
    Who holds it, and the handoff
    The merchant. Handoff: each provider supplies its Attestation of Compliance and the agreement to the merchant, and the merchant passes them to the assessor in R13.
    Who answers the question
    The merchant, for the responsibilities FAQ 1092 lists; the compliance-accepting entity, for what it accepts.
    What the Council's open pages say
    FAQ 1092 (June 2025): a merchant that outsources payment processing remains responsible for ensuring the provider is PCI DSS compliant for the services offered, maintaining written agreements that include acknowledgement of their responsibilities (Requirement 12.8.2), monitoring the provider's compliance status at least annually (Requirement 12.8.4), and clearly defining and understanding shared responsibilities. FAQ 1568 (April 2023): the AOC is intended to be shared externally to requesting entities, according to applicable Participating Payment Brand rules and as noted in the QSA Program Guide.
    What to keep
    For each provider: the service it provides; its AOC, with the date and the services it covers; the agreement clause on responsibilities; the date you last checked its compliance status; and your split of shared responsibilities, as you understand it after reading Requirements 12.8.2 and 12.8.4 in your own copy.
    What a website or web application test can supply, and what it cannot
    Nothing about the provider: its own product and platform, hosted checkout and hosted pages included, are outside what Clever Ops tests. A test of your own store can show what your side sends to the provider and accepts back, on the build tested and where the provider's terms allow it.
    Limits of this row
    Read-only row: it reads agreements and attestations and sends no requests. Read each provider's terms before any test touches a page or address it runs, and ask the provider about anything they leave open.
  6. R6. External ASV scans under Requirement 11.3.2
    Who holds it, and the handoff
    The ASV produces the scan reports and the merchant keeps them. Handoff: the merchant gives the ASV the hosts in scope from R3, the ASV returns its reports, and the merchant passes them on where the compliance-accepting entity asks.
    Who answers the question
    The ASV performs the scan with its approved scan solution. Reporting: the compliance-accepting entity (FAQ 1234).
    What the Council's open pages say
    ASV page: an ASV conducts external vulnerability scanning to validate adherence with the external scanning requirements of Requirement 11.3.2, with a scan solution the Council tests and approves. Resource guide post (10 July 2024): 11.3.2 requires evidence of passing external scans, performed by an ASV, at least once every three months; in SAQ A, the scans apply only to the merchant system that hosts the webpage that redirects to, or embeds a payment page or form from, a PCI DSS compliant provider. FAQ 1604 (June 2026): scans for 11.3.2 must be performed by a listed ASV using its scan solution. FAQ 1152 (January 2024): among its characteristics, a passing external scan generally shows no vulnerabilities scoring 4.0 or higher on CVSS. FAQ 1234 (June 2025) names Requirement 11.3.2.1 for quarterly external scans by an ASV, and FAQ 1604 says 11.3.2 and 11.3.2.1 were both added to SAQ A: read both in your copy.
    What to keep
    The ASV's name, the date you checked it on the Council's list, the hosts given to it, each scan date and result, the report on the ASV's official template, and any supplemental document it sent, marked as supplemental. FAQ 1234 says supplemental materials have not been endorsed by the Council and should not be considered replacements for its templates.
    What a website or web application test can supply, and what it cannot
    Nothing that stands in for an ASV scan. Clever Ops is not an Approved Scanning Vendor, and a website check or web application test is not a scan under Requirement 11.3.2.
    Limits of this row
    Read-only row: the scans are the ASV's work under its own agreement with you, and this row sends nothing.
  7. R7. ASV scan findings: remediation and rescans
    Who holds it, and the handoff
    The remediation provider (your developer, agency or host) fixes, the ASV rescans, and the merchant keeps both records. Handoff: merchant to remediation provider with the scan finding; remediation provider back to the merchant with the change record; merchant to the ASV for the rescan.
    Who answers the question
    The ASV, through its rescan result. The merchant decides who fixes each finding and by when.
    What the Council's open pages say
    FAQ 1152 (January 2024): among the things that should occur to verify that the requirement to perform vulnerability scans at least once every three months has been met, rescans are performed as necessary and show that vulnerabilities identified in the initial scans have been remediated, for all affected systems, as part of that period's scanning process. FAQ 1087 (July 2023): to meet the scanning requirements in Requirement 11, an entity is required to complete its internal and external scans, and any required remediation, at least once every three months.
    What to keep
    For each finding the ASV reported: its owner, the fix and its date, the build or configuration change that carried it, and the rescan date and result.
    What a website or web application test can supply, and what it cannot
    Nothing that replaces the rescan, which stays with the scanning vendor. Where a separate penetration test reported the same weakness, keep that finding and its retest status beside this record.
    Limits of this row
    Read-only row: it records fixes and results and sends no requests.
  8. R8. Penetration testing: the scope given to the tester
    Who holds it, and the handoff
    The merchant writes the scope with the tester. Handoff: merchant to tester (scope, test accounts, environment and signed authorisation), then tester to merchant (the report, in R10).
    Who answers the question
    Your assessor or acquirer, on which requirement in your version of the standard covers penetration testing; your assessor, on whether the work meets it, once you have read that requirement there.
    What the Council's open pages say
    Not on the open pages read for this guide: none names a penetration testing requirement or says what one asks. Read the requirement your assessor or acquirer names for penetration testing, with its testing procedures and guidance, in your copy of PCI DSS. FAQ 1333 (July 2015): compliance cannot be determined by testing only pre-production environments using test data.
    What to keep
    The requirement your assessor or acquirer named for penetration testing, with its number as they gave it; the version of the standard you read it in, and where its text sits in that copy; what your assessor said they expect to see; the scope agreed in writing, listing targets, roles, test accounts, environment, depth, exclusions and the testing window; each target owner's signed authorisation; the date the host's testing policy was read; and whether testing ran on production or on a staging copy.
    What a website or web application test can supply, and what it cannot
    Dated evidence about one application on the build tested: the scope as tested and, from a web application test, the test cases that could not run and why. Not a finding that any PCI DSS requirement is met.
    Limits of this row
    Read-only row: it gathers documents before a test and sends no requests; the tester's signed authorisation and rules of engagement govern the test.
  9. R9. The tester's relationship to the system
    Who holds it, and the handoff
    The merchant, from the tester's written statement. Handoff: tester to merchant, then merchant to assessor in R13.
    Who answers the question
    Your assessor, on whether the tester is acceptable for the work. Whether PCI DSS says anything about who may perform penetration testing is not on the open pages read for this guide.
    What the Council's open pages say
    The PCI DSS page and the QSA page describe QSA companies as independent security organisations qualified by the Council. Neither says anything about penetration testers. FAQ 1598 (June 2025): entities should follow their internal third-party due diligence processes when engaging an ASV or QSA company.
    What to keep
    The tester's name and company; whether it had any hand in building, hosting or running the system, or operates a control under test; any qualification or independence your copy of the standard, your assessor or your acquirer asks for, in their exact words; and who approved the choice, and when.
    What a website or web application test can supply, and what it cannot
    The tester's own written statement. For a system Clever Ops built or hosts, the report identifies both the builder and the host, and the work is described as a security review, which is not independent testing.
    Limits of this row
    Read-only row: it notes who was chosen and why, and sends no requests.
  10. R10. Penetration test report, remediation and retest
    Who holds it, and the handoff
    The tester issues the report and the retest, the remediation provider fixes, and the merchant keeps all three. Handoff: tester to merchant (report); merchant to remediation provider (findings); remediation provider to merchant (fixes on an identified build); merchant to tester (retest request); tester to merchant (retest result).
    Who answers the question
    The merchant decides what to fix, by when and who fixes it. Whether the record meets a PCI DSS requirement is for your assessor.
    What the Council's open pages say
    None of the open pages read for this guide says anything about penetration test findings. Read the requirement your assessor or acquirer names for penetration testing in your copy of the standard for anything it says about findings and retesting.
    What to keep
    The report's version and issue date; which build was tested and on what dates; every finding with its ratings and the person who owns it; the fix, and the build it shipped in; when the retest ran; and the status each finding ended with.
    What a website or web application test can supply, and what it cannot
    From a Clever Ops report: each finding with redacted evidence, its fix and the criteria the retest will apply, and, where the finding is rated, a CVSS v3.1 score and a business-risk rating; the checks that found nothing, limited to the build and accounts tested; and the fixes in priority order. Following the one consolidated retest, the report shows each finding as fixed, partially fixed, still reproducible or not retested. A change shipped after the retest is not covered.
    Limits of this row
    Read-only row: it files reports and decisions and sends no requests. The full report explains how to reproduce any finding still open, so keep its circulation to those who need it.
  11. R11. Significant change log
    Who holds it, and the handoff
    The merchant. Handoff: the merchant tells the ASV (R6) and the tester (R8) of any change it evaluates as significant, and gives the log to the assessor in R13.
    Who answers the question
    FAQ 1317 says each listed activity must be considered and evaluated to determine whether a change is significant for that entity; it does not say who does the evaluating, so record who did.
    What the Council's open pages say
    FAQ 1317 (April 2023), giving the significant change list in PCI DSS v4.0: new hardware, software or networking equipment added to the CDE; replacement or major upgrades of hardware and software in the CDE; changes in the flow or storage of account data; changes to the boundary of the CDE and/or to the scope of the assessment; changes to the underlying supporting infrastructure of the CDE, including, but not limited to, directory services, time servers, logging and monitoring; and changes to third party vendors or service providers (or services provided) that support the CDE or meet PCI DSS requirements on the entity's behalf. FAQ 1087 (July 2023): vulnerability scans are required after a significant change as well as at least once every three months.
    What to keep
    Each change to the store, its payment step, its hosting or its providers; its date; who evaluated it against the list, when, and the outcome; and which scans, tests or checks followed, with their dates.
    What a website or web application test can supply, and what it cannot
    A dated test or check after a change, where one followed. Not the evaluation itself.
    Limits of this row
    Read-only row: it records changes and decisions and sends no requests.
  12. R12. Access check after a change to the store
    Who holds it, and the handoff
    The merchant's own team or developer runs it. Handoff: the results go into R11 against the change that prompted them.
    Who answers the question
    The merchant decides whether a request that was not refused means bringing the next test forward.
    What the Council's open pages say
    None: this row is this guide's own narrow check between tests, not a step any Council page describes. It is not a scan under Requirement 11.3.2 or a penetration test.
    What to keep
    The change and its release date, then for each step the test account used (or none), the date and time, the status returned and whether any of test customer A's details came back.
    What a website or web application test can supply, and what it cannot
    Nothing from a tester: your own team runs it. Refusals at steps 4, 5 and 6 show that those three requests were refused on that build, and nothing more.
    Limits of this row
    Limits for this row, the only one in the matrix that sends anything. Run it against a store or application that belongs to your business, on a staging copy wherever you have one. Use only test customers whose names, addresses and orders were invented for testing, never a real customer's account. No card number, real or test, goes into any request or any note, and no payment is submitted: apart from the two sign-ins, each step only reads an order, a page or a file. Take the six steps listed at the end of this cell, once each and one after another, and nothing more, and put no load on the store. Any email or text the store sends in response has to land in an inbox or on a number your business controls. Your notes hold no password, key or session token, and no link that would open an order or a file for anyone holding it; open a link of that kind only for a test order. If a vendor runs any part of the store, read its terms before you start, ask the vendor about anything the terms leave unclear, and stay within the terms and its written answer. If any response shows a real person's details, halt at that point, log the page address and the time, look at nothing more, and report it to the person who handles privacy and security incidents for your business. The six steps, the two sign-ins included, each done once. 1: sign in as test customer A. 2: as A, open the page of a test order that belongs to A on the staging copy (one your developer created, or one left from an earlier test) and note its order number, writing down the page address only if it carries no key or token. 3: in a separate private window, sign in as test customer B. 4: as B, open A's order page by that order number. 5: as B, request the invoice or receipt file for the same order. 6: in a new private window that holds no session, open A's order page by its order number, signed out.
  13. R13. The assessor's evidence pack
    Who holds it, and the handoff
    The merchant assembles it. Handoff: merchant to the QSA or other assessor, who returns the assessment documents it completes.
    Who answers the question
    The assessor, on scope, applicability and whether each requirement is in place; the compliance-accepting entity, on what it accepts.
    What the Council's open pages say
    QSA page: QSA companies are independent security organisations qualified by the Council to validate an entity's adherence to PCI DSS, re-qualified each year; check the list when you engage one. FAQ 1473 (March 2023): the assessor validates that scope and applicability are accurately defined and documented, and must confirm through testing that a requirement is truly not applicable before reporting it as Not Applicable.
    What to keep
    An index of R1 to R12 for the period: the validation answer, the SAQ or ROC and its version, the payment page list, the scope and segmentation record, provider attestations, ASV scan reports and rescans, penetration test reports and retests, the tester's statement and the change log, with where each is kept.
    What a website or web application test can supply, and what it cannot
    The test reports and retests filed in R10, for the assessor to weigh. Not an assessment, and not a decision on scope or applicability.
    Limits of this row
    Read-only row: it assembles records and sends no requests.
  14. R14. Submission to the compliance-accepting entity
    Who holds it, and the handoff
    The merchant submits and the compliance-accepting entity receives. Handoff: merchant to acquirer or payment brand, on the Council's forms.
    Who answers the question
    The compliance-accepting entity, on how compliance is to be evidenced and reported (FAQ 1473).
    What the Council's open pages say
    Glossary: the AOC is the official PCI SSC form for merchants and service providers to attest to the results of a PCI DSS assessment, as documented in an SAQ or ROC. Post of 2 September 2025, quoting FAQ 1220: only the official form documents from the Council's website are recognised for PCI DSS validation, and other certificates are not acceptable for evidencing compliance. FAQ 1234 (June 2025): acquirers and/or payment brands may also request quarterly ASV scan reports.
    What to keep
    What was submitted, on which official form and version, the date, to whom, and any reply.
    What a website or web application test can supply, and what it cannot
    Nothing that goes into this submission as a validation document. A test report is not one of the Council's official forms.
    Limits of this row
    Read-only row: it logs what was lodged and sends nothing to any system.

PCI DSS compliance at a glance: the standard, its publisher and its version

PCI DSS, the PCI Data Security Standard, is published by the PCI Security Standards Council.

The Council's page for the standard says it was developed to encourage and enhance payment card account data security and to facilitate the broad adoption of consistent data security measures globally, and that it provides a baseline of technical and operational requirements designed to protect payment account data. That page also sends readers to the Council's Document Library for the standard and its supporting documents.

Which version a record refers to matters. In a post dated 11 June 2024, the Council announced PCI DSS v4.0.1, a limited revision that corrects formatting and typographical errors and clarifies the focus and intent of some requirements and guidance, with no additional or deleted requirements.

The post said v4.0 would be retired on 31 December 2024, after which v4.0.1 would be the only active version the Council supports, and that the revision did not change the 31 March 2025 effective date of the new requirements.

FAQ 1585, dated October 2024, says that from a new requirement's effective date, all requirements applicable to an entity's assessment, the newly effective ones included, must be fully considered as part of its PCI DSS assessment.

On 3 June 2026 the Council posted a request for comments on the published v4.0.1, open to eligible stakeholders from 3 June to 20 July, and said it was beginning the next iteration of PCI DSS.

FAQ 1328, dated October 2024, says the Document Library holds the standard and keeps retired versions there as archived documents, and sends compliance questions, including whether it is acceptable to submit an assessment report for a retired version, to the organisations that manage compliance programs, for example payment brands and acquirers. Check the version in the Document Library before you rely on any requirement number on this page.

This guide does not quote the standard or its questionnaires. Where it says what a requirement covers, it repeats what one of the Council's open pages says about that requirement, and credits that page. The standard and the SAQs are the Council's own documents: its PCI DSS page sends readers to its Document Library for the standard, and its post on compliance certificates calls its website the official source for the reporting templates and forms, the SAQs among them.

The Intellectual Property Rights Policy on the Council's policies page, last updated 18 January 2023, says the copyright for all Standards and Other Work Product shall belong to the Council, and the policy's Exhibit A is the text of a licence agreement that names the Council as owner of the copyright in the standards and documents reached through it.

This guide accepted no licence and opened none of those documents, and it was written without the text of the standard.

Apart from the Council's privacy policy, which says the Council conducts penetration tests against its own infrastructure, none of the Council pages this guide cites mentions penetration testing.

Obtain the standard and its SAQs from the Council and read them under the licence terms it sets. A requirement recalled from memory, or copied from a secondary website, is an easy way to file the wrong evidence against it. Where this page needs to say what a requirement asks and none of the Council's open pages says it, the page says so and asks you to read the text in your own copy and put the question to your assessor. The matrix at the end has a column for that answer.

The page is written for the owner, operations manager or technical lead of a business that takes card payments through a website or web application, and that has met PCI DSS through a merchant agreement, a payment provider's request, a customer's questionnaire or an insurer's form.

It reports what the Council's open pages say and stops there. It offers no legal opinion and performs no assessment, and it cannot tell you which questionnaire is yours or whether your business meets the standard.

Who PCI DSS covers, and who decides what a merchant has to do

The Council's PCI DSS page gives the standard's intended audience as entities that store, process or transmit cardholder data and/or sensitive authentication data, or could impact the security of the cardholder data environment.

It says this includes all entities involved in payment card processing, including merchants, processors, acquirers, issuers and service providers. The Council's glossary defines a merchant, for PCI DSS purposes, as any entity that accepts payment cards bearing the logos of any PCI SSC Participating Payment Brand as payment for goods and/or services.

The FAQ on its Merchant Resources page adds that PCI DSS is intended for all entities involved in payment processing, including merchants, regardless of their size or transaction volume.

FAQ 1092, dated June 2025, takes up the merchant who outsources all payment processing operations and never stores, processes or transmits cardholder data, and answers yes: PCI DSS applies. It says many requirements may not apply directly to that merchant's environment, but that this does not remove the merchant's responsibility to ensure account data is properly protected by the third party.

It lists what the merchant remains responsible for: ensuring the provider is PCI DSS compliant for the services offered, maintaining written agreements with the provider that include acknowledgement of their responsibilities (Requirement 12.8.2), monitoring the provider's compliance status at least annually (Requirement 12.8.4), and clearly defining and understanding any shared responsibilities.

It adds that merchants are still required to validate PCI DSS compliance, typically through a Self-Assessment Questionnaire such as SAQ A, and should confirm their compliance obligations with the organisations that manage their compliance program, such as their acquirer or payment brand.

In its post of 10 March 2025 on payment page security, the Council says it does not enforce compliance or determine whether specific implementations are compliant. Its PCI DSS page says whether an entity is required to comply with or validate compliance to a PCI SSC standard is at the discretion of organisations that manage compliance programs, such as a payment brand, acquirer or other entity.

FAQ 1092 says those organisations are also referred to as compliance-accepting entities, and FAQ 1473, dated March 2023, says compliance-accepting entities (typically, payment brands and acquirers) are responsible for determining the validation and reporting methods of their merchants and service providers, including how compliance is to be evidenced, for example through a Report on Compliance or a Self-Assessment Questionnaire.

The same FAQ gives assessors a separate job: validating that the scope of the assessment and the applicability of PCI DSS requirements are accurately defined and documented.

FAQ 1126, dated July 2012, tells merchants to contact the acquiring financial institutions with whom they have merchant agreements, such as their merchant bank, to determine whether they must validate compliance and the specific requirements for performing and reporting that validation. The acquirer you hold that agreement with is therefore the first party to ask.

Whether validation applies to your business, in what form and over what scope, is for you to confirm with your acquirer or the payment brand, as the Council's FAQs put it, and with your assessor where you have one.

The Merchant Resources FAQ also encourages entities to be aware of potential nuances in local laws and regulations that could affect applicability of the PCI standards; for an Australian business, take that question to your own adviser.

Self-assessment questionnaires, the Report on Compliance and the attestation

The Council's glossary defines the documents that carry validation results.

A Self-Assessment Questionnaire (SAQ) is a reporting tool used to document self-assessment results from an entity's PCI DSS assessment. A Report on Compliance (ROC) is a reporting tool used to document detailed results from an entity's PCI DSS assessment.

An Attestation of Compliance (AOC) is the official PCI SSC form for merchants and service providers to attest to the results of a PCI DSS assessment, as documented in an SAQ or a ROC.

FAQ 1568, dated April 2023, says the AOC is intended to be shared externally to requesting entities, according to applicable Participating Payment Brand rules and as noted in the QSA Program Guide.

FAQ 1215, dated April 2024, says SAQs are validation tools for SAQ-eligible merchants and service providers, and that there are several, developed for specific types of environments as defined in each SAQ's eligibility criteria. It says each SAQ has a section headed "Completing the Self-Assessment Questionnaire" that outlines the type of environment it is intended for, and all the eligibility criteria for a particular SAQ must be met to use it.

The FAQ then says merchants should consult their compliance-accepting entity, the entity the SAQ will be submitted to (typically an acquirer or a payment brand), to determine whether they are eligible or required to submit an SAQ and, if so, which one is appropriate for their environment.

It also says SAQ D for Service Providers is the only SAQ for SAQ-eligible service providers, and that all other SAQs are for merchant use only.

A questionnaire involves testing as well as answering. FAQ 1316, dated January 2015, says the Expected Testing column of each SAQ gives a high-level description of the testing activities that should be performed to determine whether a requirement is in place, and that the people responsible for the self-assessment, that is, the merchant or service provider, or their QSA, are expected to perform them.

The Council recognises only its own forms as validation documents. In a post dated 2 September 2025, it quotes its FAQ 1220: the only documentation recognised for PCI DSS validation is the official form documents from the PCI SSC website, and any other certificate or documentation issued to document compliance is not authorised or validated by the Council and is not acceptable for evidencing compliance.

The post says the official templates include the ROC, the AOC, the SAQs and the Attestations of Scan Compliance for ASV scans, says organisations receiving compliance certificates are under no obligation to accept them, and says the Council does not provide compliance certificates or compliance logos. A penetration test report is not one of the Council's official forms, whoever writes it.

What the Council's FAQs say about PCI DSS SAQ A

The Council's FAQs discuss SAQ A in connection with online merchants whose pages redirect customers to a payment provider or include a provider's embedded iframe. This guide does not say whether it is yours. On the open pages, FAQ 1092 names it as an example of the questionnaire through which a merchant that outsources all payment processing typically validates.

FAQ 1604, dated June 2026, says SAQ A for PCI DSS v4.x includes requirements for external vulnerability scanning by an Approved Scanning Vendor for merchant e-commerce webpages, even where payment processing is fully outsourced to a third party, and that Requirements 11.3.2 and 11.3.2.1 were added to SAQ A to address risks where a merchant's webpage could be compromised and thereby result in compromise of the payment process.

It says merchants with e-commerce webpages that complete SAQ A still have responsibility for the requirements included in it, the ASV scanning requirements among them, and that this includes merchants whose webpages redirect transactions to a third-party service provider (or to another third-party redirection server that then redirects to one) or include a provider's embedded iframe (or an iframe that includes another provider's iframe).

FAQ 1332, dated July 2015, says a merchant website is still in scope for PCI DSS even when it meets all the criteria for SAQ A: the merchant web server must be included in scope so the assessor can examine its configuration and verify the redirection mechanism used.

It adds that if the merchant environment and web server redirection meet all criteria for SAQ A, the minimum applicable requirements can be considered as those within that SAQ. Scripts that run on payment pages, and PCI DSS v4.x Requirements 6.4.3 and 11.6.1, which the Council's post of 10 March 2025 says focus on payment page scripts being properly authorised, checked for integrity and monitored for tampering, are the subject of this library's PCI payment script controls guide.

A PCI ASV scan and a penetration test are different evidence

An Approved Scanning Vendor (ASV) holds a role the Council defines.

Its ASV page says an ASV conducts external vulnerability scanning to validate adherence with the external scanning requirements of PCI DSS Requirement 11.3.2, and that the vendor's scan solution is tested and approved by the Council before the vendor is added to its list.

In its resource guide post of 10 July 2024, the Council describes Requirement 11.3.2 as requiring evidence of passing external scans, performed by an ASV, at least once every three months.

FAQ 1604 says scans for the purposes of satisfying Requirement 11.3.2 must be performed by an ASV listed on the Council's website, using that vendor's ASV scan solution.

FAQ 1152, dated January 2024, says a clean or passing scan generally has these characteristics: no configuration or software detected that results in an automatic failure; for external scans, no vulnerabilities scoring 4.0 or higher on the Common Vulnerability Scoring System (CVSS); and, for internal scans, vulnerabilities resolved by the entity according to Requirement 11.3.1.

Among the things it says should occur to verify that the requirement for scans at least once every three months has been met, it lists rescans performed as necessary that show vulnerabilities identified in the initial scans have been remediated, for all affected systems, as part of that period's scanning process.

FAQ 1087, dated July 2023, notes that "quarterly" is the term PCI DSS v3.2.1 defines and "at least once every three months" the term v4.0 defines. It says that to meet the vulnerability scanning requirements in Requirement 11 an entity is required to complete its internal and external scans, and perform any required remediation, at least once every three months.

It treats 90 days as the most time that should pass between quarterly scans, and says PCI DSS also requires vulnerability scans after significant changes.

FAQ 1234, dated June 2025, says the ASV produces a scan report detailing the results of the vulnerability scan, and that this report is not an indication that any other PCI DSS requirements have been reviewed or are in place. It says ASVs are required to provide scan reports on official templates defined in the ASV Program Guide, that any certificates, letters or other extra documents an ASV supplies should be identified as supplemental materials, which the Council has not endorsed and which should not be considered replacements for its official templates and forms, and that acquirers and/or payment brands may also request quarterly ASV scan reports.

The Council's pages do not all give the same requirement number here.

FAQ 1234 opens by saying Requirement 11.3.2.1 addresses the need for quarterly external vulnerability scans by an ASV, while the ASV page, the resource guide post and FAQ 1604 tie ASV scans to Requirement 11.3.2, and FAQ 1604 says both 11.3.2 and 11.3.2.1 were added to SAQ A.

Read both requirements in your copy of the standard, and ask your assessor which one a scan report is filed against.

None of the Council's open pages this guide cites names a penetration testing requirement in PCI DSS. None of them says what such a requirement covers, how often it applies, who may perform it or whether it appears in a particular SAQ. Ask your assessor or your compliance-accepting entity which requirement in the version of the standard they name covers penetration testing; you may see it cited as Requirement 11.4.

Read that requirement with its testing procedures in your copy of the standard, and ask them what they expect to see before a test is scoped against it.

A scan and a penetration test also do different work. As this library uses the words, a vulnerability scan is an unattended, automated match of what it can reach against known issues, and a penetration test has a person use the application in each of its roles, attempt to get past its controls and keep evidence of what got through; the penetration testing versus vulnerability scanning guide compares the two at length.

FAQ 1234 says an ASV scan report is not an indication that any other PCI DSS requirement has been reviewed or is in place. FAQ 1604 says scans for Requirement 11.3.2 must be performed by a listed ASV using that vendor's own scan solution.

This guide therefore treats them as separate evidence: a scan report does not answer for penetration testing, and a web application test, which records how the application behaved, is not a Requirement 11.3.2 scan.

Scope and segmentation: where a merchant website sits

Scope decides which systems the requirements reach.

The Council's glossary defines scoping as the process of identifying all system components, people and processes to be included in a PCI DSS assessment. It describes the cardholder data environment (CDE) as the system components, people and processes that store, process or transmit cardholder data and/or sensitive authentication data, together with system components that may not store, process or transmit that data but have unrestricted connectivity to ones that do.

It describes segmentation as isolating system components that store, process or transmit cardholder data from systems that do not.

Its scoping and segmentation entries each point to section 4 of the standard, Scope of PCI DSS Requirements.

For a website, FAQ 1332, set out above, keeps the web server of a merchant that meets all the SAQ A criteria in scope so the assessor can examine its configuration and verify the redirection mechanism. The ASV resource guide post of July 2024 says the ASV scan requirements in SAQ A apply only to an e-commerce merchant system that hosts the webpage that either redirects payment transactions to a PCI DSS compliant third-party service provider or includes an embedded payment page or form from one.

It says the intent is for merchants to minimise the risk of compromise by scanning for and resolving identified vulnerabilities that could potentially expose their link to the provider's payment page.

On segmentation, FAQ 1135, dated July 2012, says adequate network segmentation can in general reduce the scope of the PCI DSS assessment if it isolates systems that store, process or transmit cardholder data from other systems, and that it can be implemented with, for example, properly configured internal firewalls, routers with strong access control lists, VLANs or other technologies that restrict access to a particular network segment.

It says the Council cannot offer an opinion on how an organisation achieves adequate segmentation, because that needs an understanding of the security features and controls in its environment, and it encourages contacting a Qualified Security Assessor to help scope the cardholder data environment.

The Council's information supplement on scoping and segmentation for modern network architectures, announced on 9 September 2024, includes guidance on specific requirements for verifying scope and segmentation controls, and the announcement says that guidance is supplemental and does not supersede or replace any PCI standard.

A test of a website or web application does not settle scope. It can supply inputs: a website security check's hostname inventory and a web application test's list of routes, roles and accounts tested show what answered on the dates of the test, which is dated material for whoever draws the scope.

Clever Ops works on what customers reach in a browser or through an API and refers internal networks to a network specialist, so segmentation, which the glossary also calls network segmentation, is not something a Clever Ops test examines.

Significant change: the trigger that brings scans and tests round again

Some PCI DSS activity is tied to change rather than to the calendar.

FAQ 1317, dated April 2023, says several PCI DSS requirements specify performance upon a significant change in an entity's environment, and that what constitutes a significant change is highly dependent on the configuration of a given environment. It lists these activities as included under "Significant Change" in the section of PCI DSS v4.0 that describes the timeframes used in its requirements:

  • New hardware, software or networking equipment added to the CDE.
  • Any replacement or major upgrades of hardware and software in the CDE.
  • Any changes in the flow or storage of account data.
  • Any changes to the boundary of the CDE and/or to the scope of the PCI DSS assessment.
  • Any changes to the underlying supporting infrastructure of the CDE, including, but not limited to, changes to directory services, time servers, logging and monitoring.
  • Any changes to third party vendors or service providers (or services provided) that support the CDE or meet PCI DSS requirements on behalf of the entity.

Deciding whether a change counts

The FAQ says each of those activities, at a minimum, has potential impacts on the security of an entity's cardholder data environment and must be considered and evaluated to determine whether a change is significant for that entity and in the context of related PCI DSS requirements.

For scans, FAQ 1087 says PCI DSS requires vulnerability scans after significant changes, in addition to the scans conducted at least once every three months. The list FAQ 1317 gives is the one in PCI DSS v4.0, so read the same section in the version your assessor names.

Whether PCI DSS ties penetration testing to significant change is in the standard's text, which the open pages read for this guide do not reproduce.

For a merchant's website, this guide reads a new payment provider, a change to how the checkout hands customers to the provider, and a new host or other service provider for the store as candidates to evaluate against that list, because the FAQ names changes in the flow of account data, to the boundary of the CDE and to the providers that support it.

Whether any of them is significant for your environment is not this guide's call. Row R11 of the matrix logs each change, who evaluated it, when, and what followed.

Who may scan, assess and test, and the question of independence

The Council qualifies the companies that scan and assess.

Its PCI DSS page says Qualified Security Assessors (QSAs) are independent security organisations that have been qualified and trained by the Council to perform PCI DSS assessments, and that ASVs are qualified and trained by the Council to conduct external vulnerability scanning services in accordance with the applicable PCI DSS requirement.

Its QSA page describes a Council program under which security companies are qualified as QSAs and re-qualified each year, and links to the Council's list of them. The ASV and QSA pages each say the list is updated frequently and advise a client to check it each time it engages an ASV or a QSA.

FAQ 1598, dated June 2025, adds that whatever relationship these companies have with the Council, entities should follow their internal third-party due diligence processes when engaging an ASV or QSA company.

Whether PCI DSS says who may carry out penetration testing, or expects any independence of them, is a question for its text: none of the open pages read for this guide covers it. Ask your assessor or acquirer which requirement in your version of the standard covers penetration testing, read it in your copy, and ask your assessor how they will judge the tester.

Row R9 then records how the tester stands in relation to the system (whether it had a hand in building, hosting or running it, or operates a control under test), so that whoever weighs independence has those facts.

When Clever Ops is the builder or the host of the store or application, what it delivers on that system is a security review rather than an independent test, and the report records the builder and the host by name. Where an assessor, an acquirer or your own policy asks for a tester who had no part in building or running the system, engage a provider that did neither.

PCI DSS penetration testing: what a website or application test can add, and what it cannot

What a website or application test yields is a dated record of one build: what was tried, from which accounts, inside the scope and window agreed in writing.

For a merchant, a signed-in test, such as a WordPress security audit of a WooCommerce store or a web application and API penetration test, can show how the store's own side of the payment behaved: whether the amount comes from the server or can be changed in the checkout request, and whether an order is marked paid only once the payment provider confirms the payment.

A web application and API penetration test also checks whether the route that receives the provider's notifications turns away a message without a valid signature. Either can show whether one customer's orders and account details stay out of another customer's reach, and how sign-in, sessions and staff roles behave.

Those areas are set out in the web application penetration testing guide, and the ecommerce website security guide traces them along the order path.

Where a test ran matters for PCI DSS purposes. FAQ 1333, dated July 2015, says PCI DSS compliance cannot be determined by testing only pre-production environments using test data, because there are many tests an assessor would be unable to perform there.

It says reviewing a pre-production environment before it goes live may help the assessor understand how it will function, but that the assessor could not complete an assessment, or state that all applicable requirements are in place, until the environment is in use.

In a Clever Ops web application and API penetration test, the payment checks use a staging copy connected to a payment account that is not live, wherever one exists, so that no real card is billed; a WordPress security audit does its checkout testing either on a staging copy or with the gateway set to its test mode.

Record which environment each test used, and ask your assessor how they will treat a result from staging.

Against the matrix on this page, a report from a web application test or a WordPress security audit can supply these records, and nothing beyond them:

  • For the payment page and scope rows (R3 and R4): the targets, routes, roles and accounts tested, as a dated input for the people who set scope. It does not set the scope, and it says nothing about network segmentation.
  • For the penetration testing rows (R8 to R10): the scope as it was actually tested; each finding with redacted evidence, scored on CVSS v3.1 and given a business-risk rating where it is rated; and the checks that found nothing on the build and accounts tested. A web application test report also lists the test cases that could not run. The one consolidated retest then gives every finding a status: fixed, partially fixed, still reproducible or not retested. Whether that record meets a PCI DSS requirement is for your assessor.
  • For the change log (R11): a dated record of what was tested after a change, where a test followed one.
  • For the assessor's pack (R13): the report and the retest, as documents the assessor may weigh.

What it cannot supply

It is not an ASV scan and cannot stand in for one, because FAQ 1604 ties Requirement 11.3.2 scans to a listed ASV using its own scan solution. It is not an SAQ, a ROC or an AOC, and the Council's post on compliance certificates says only its official forms are recognised for PCI DSS validation.

It does not decide SAQ eligibility, scope or whether a requirement applies: the Council's FAQs leave eligibility to the compliance-accepting entity, and scope and applicability to the assessor to validate.

It covers nothing of the payment provider's own platform, hosted checkout or hosted pages, nothing of the merchant's internal networks, and nothing released after the build it tested. And no test makes anyone PCI DSS compliant.

How to use the SAQ-aware responsibility matrix

The matrix records who holds which document and who answers which question, row by row.

It does not state what PCI DSS requires of you. Its fourth column gives what a Council open page says, naming the page and any date the page carries; where no open page covers the point, the cell says so and sends you to the standard's text and to your assessor, and in R12, the guide's own check, it cites no Council page at all.

"SAQ-aware" means only that the matrix has a place to record which questionnaire your compliance-accepting entity named and its version; the matrix itself never says which SAQ applies.

  • Copy it into a spreadsheet and keep one copy for each business that accepts cards, with one line under R3, R6 and R8 for each website or application that takes customers to the payment step.
  • Fill in R1 and R2 first, from your acquirer's or payment brand's written answer, and write the PCI DSS version they name beside every requirement number you record.
  • Rows R1 to R11, R13 and R14 read or write records and send nothing to any system. Only R12 sends requests, and it carries its own limits in its last column.
  • The second column names who passes each record to whom, among them the merchant, the ASV, the QSA or other assessor, the tester, and the remediation provider, whether that is your developer, your agency, your host or a security fix sprint.
  • Keep the records each row names even when the result is clean. A dated pass is what lets a later row, or your assessor, see that something was checked.
  • Where a row says a decision belongs to someone, write down who made it and when.
  • The stop rule in R12's limits holds for every row: if a real person's details appear, halt and report it as that cell describes.

Where Clever Ops fits, and the PCI work it refers on

Clever Ops works on what a customer can reach in a browser or through an API.

For a business that takes cards online, that means an external website security check; a WordPress security audit, which on a WooCommerce store brings customer accounts, orders and the checkout into scope; or a web application and API penetration test of a custom-built store or application that customers or staff sign in to.

A website security check's hostname inventory can go into row R3. A WordPress security audit or a web application and API penetration test produces the report and the retest that rows R8, R10 and R13 file, with the contents listed earlier on this page.

Who fixes the findings is a separate choice. It can be your developers or your agency, or a Clever Ops security fix sprint, which has its own statement of work and can take findings from a Clever Ops report or from another tester's.

Clever Ops holds neither PCI role. It is not an Approved Scanning Vendor or a Qualified Security Assessor, offers no ASV scan and no QSA assessment, and refers a merchant that needs either to a company named on the Council's ASV or QSA list.

A Clever Ops report is evidence that the merchant, its assessor or its acquirer may weigh; it is not an Attestation of Compliance, a scan report under Requirement 11.3.2 or a certificate, and it gives no opinion on whether anyone meets PCI DSS.

Clever Ops holds no security certification or accreditation, and because it is not CREST accredited, work for which an acquirer, contract or tender specifies a CREST-accredited tester goes to a provider that is.

A payment provider's own product or platform, its hosted checkout and hosted pages included, is not something Clever Ops tests: the payment work in a Clever Ops test stays on the merchant's side of the payment. Clever Ops refers internal networks, servers, Active Directory and staff devices to a network specialist, offers no denial-of-service or load testing, and does not take on live Shopify storefronts or Laravel applications.

No testing starts until each target's owner has signed the authorisation and the host's testing policy has been read; the methodology guide covers the rest of an engagement, from scope to retest, and the website security audit guide covers the external check.

Frequently asked questions

The Council's FAQ 1234 takes up this question. It says the scan report an ASV produces details the results of the vulnerability scan and is not an indication that any other PCI DSS requirements have been reviewed or are in place. The same FAQ describes quarterly ASV scan reports as providing evidence of meeting a PCI DSS requirement. How your compliance is validated, and with which documents, is determined by your compliance-accepting entity, typically a payment brand or your acquirer, as FAQ 1473 puts it.

The Council's FAQ 1092 answers yes for merchants who outsource all payment processing operations and never store, process or transmit cardholder data. It says many requirements may not apply directly to such a merchant's environment, but that the merchant remains responsible for, among other things, ensuring the provider is PCI DSS compliant for the services offered and monitoring the provider's compliance status at least annually, and is still required to validate, typically through a Self-Assessment Questionnaire such as SAQ A. FAQ 1604 adds that SAQ A for PCI DSS v4.x includes ASV scanning of merchant e-commerce webpages that redirect to a provider or include its embedded iframe. How this applies to your business is for your acquirer or payment brand to confirm.

Not this page, and not a test. FAQ 1215 says merchants should consult their compliance-accepting entity, the entity the SAQ will be submitted to (typically an acquirer or a payment brand), to determine whether they are eligible or required to submit an SAQ and which one fits their environment. FAQ 1473 says those entities determine how compliance is evidenced, for example through a Report on Compliance or an SAQ, and FAQ 1126 sends merchants to the acquiring financial institution they hold a merchant agreement with. Put the question to them in writing and keep the answer in rows R1 and R2 of the matrix.

The one your compliance-accepting entity and your assessor name. In a post dated 11 June 2024, the Council announced PCI DSS v4.0.1, a limited revision with no additional or deleted requirements, and said v4.0 would be retired on 31 December 2024. On 3 June 2026 it opened a request for comments on the published v4.0.1 and said it was beginning the next iteration of the standard. FAQ 1328 points to the Document Library for the standard and sends compliance questions, including whether a report may be submitted for a retired version, to organisations that manage compliance programs, for example payment brands and acquirers. Write the version beside every requirement number you record.

The Council's FAQ 1333 says PCI DSS compliance cannot be determined by testing only pre-production environments using test data, because there are many tests an assessor would be unable to perform there, and that an assessor could not state that all applicable requirements are in place until the environment is in use. A staging test is still evidence of how that build behaved. Whether it counts towards any PCI DSS requirement is a question for your assessor; ask them or your acquirer which requirement in your version of the standard covers penetration testing, and read it there. Row R8 of the matrix records the environment each test used.

The Council recognises only its official forms for PCI DSS validation. In its post of 2 September 2025 it quotes FAQ 1220: any other certificate or documentation issued to document compliance is not authorised or validated by the Council and is not acceptable for evidencing compliance. The post adds that organisations receiving compliance certificates are under no obligation to accept them. A Clever Ops report is a test report, not a certificate, and Clever Ops is not a Qualified Security Assessor or an Approved Scanning Vendor.

Scope the right test for your application

Tell us what you have built, who uses it and when it needs to launch. You will get a plain-English recommendation: a website audit, a web application test or a referral to a specialist.