Join our Newsletter — 33% off our NHI Course
Home› Guides› Banking Cybersecurity Programme Guide: One Control Set Across…
Guide Regulations & Frameworks

Banking Cybersecurity Programme Guide: One Control Set Across FFIEC, DORA, SWIFT and PCI

PDF for newsletter subscribers · free to read online

← All guides
By NHI Mgmt Group, edited by Lalit Choda Updated 27 October 2026 15 min read
On this page

Most banks run four or five separate compliance programmes for cyber risk, and the gaps between them are where incidents and findings land. A US or EU bank with a SWIFT connection and a card business answers to supervisory guidance (FFIEC in the US), an operational resilience regulation (DORA in the EU), the SWIFT Customer Security Controls Framework and PCI DSS, each with its own evidence cycle, scope and deadline. This guide shows how to run them as one programme: a single control library, one asset and third-party inventory, one incident clock and one testing calendar. It is written for CISOs, heads of technology risk and second-line teams in banks and credit institutions. It links to the detailed NHI Mgmt Group guides for each regime rather than repeating them.

Key takeaways

  • Build one control library, map it out to each regime rather than maintaining parallel spreadsheets for FFIEC, DORA, SWIFT CSCF and PCI DSS. NIST CSF 2.0 is a practical common spine because FFIEC points to it and it maps cleanly to the others.
  • Accountability sits with the board under DORA Article 5(2), which states the management body bears "the ultimate responsibility for managing the financial entity's ICT risk". Design governance and reporting for that standard even outside the EU.
  • Run one incident classification with several clocks: 36 hours to the primary federal regulator under the US rule, DORA major incident timelines, SWIFT and card scheme notification, plus privacy breach law. Classify once, notify many.
  • Third parties and machine identities are the shared weak point. Core banking providers, cloud platforms, payment gateways and the service accounts and API keys that connect them appear in every regime's scope.
  • Evidence is the product. SWIFT asks for annual attestation with independent assessment, PCI DSS for annual assessment, DORA for a register of information and testing records. Collect evidence continuously and reuse it.

The regulatory stack a bank actually faces

Which obligations apply depends on charter, geography and business lines. The table lists the common layers and what each one asks for at programme level. Check the current text of each before relying on a date: these regimes change on annual or multi-year cycles.

RegimeWho it bindsWhat it asks forEvidence rhythm
FFIEC IT Examination Handbook and statements (US)Banks and credit unions examined by the FFIEC member agenciesRisk-based information security programme, authentication and access guidance (2021), cloud security statement (2020); examiners assess against the handbookExamination cycle
Computer-security incident notification rule (US, 86 FR 66424)Banking organisations and their bank service providersNotify the primary federal regulator within 36 hours of determining a notification incident; providers notify customers as soon as possible after a disruption of four or more hoursPer incident
DORA, Regulation (EU) 2022/2554EU financial entities including credit institutions, and critical ICT third-party providersICT risk management framework, major ICT incident reporting, digital operational resilience testing including TLPT for selected entities, ICT third-party risk and register of information, information sharingContinuous, with annual reviews and periodic testing
SWIFT Customer Security Programme (CSP)Every SWIFT userImplement CSCF controls (mandatory and advisory) under three objectives: secure your environment, know and limit access, detect and respondAnnual attestation via KYC-SA, 1 July to 31 December, with independent assessment
PCI DSS v4.0.1Card issuers, acquirers and anyone storing, processing or transmitting cardholder dataTwelve requirement families covering network controls, secure configuration, data protection, vulnerability management, access, logging and testingAnnual assessment, quarterly scans
Privacy and breach law (GDPR, US state laws)Any bank handling personal dataSecurity of processing, breach notificationPer incident and ongoing

Three status points matter in 2026. The FFIEC published a sunset statement for its Cybersecurity Assessment Tool in August 2024 and lists the NIST Cybersecurity Framework and CISA resources among alternatives on its cybersecurity page, so banks that relied on the CAT need another self-assessment method. PCI DSS v4.0 was retired on 31 December 2024 in favour of v4.0.1, and the future-dated requirements took effect on 31 March 2025. SWIFT publishes an updated CSCF each July, a year before it takes effect, so the control set you attest against changes annually. For regime detail see the FFIEC Cybersecurity Guidance Guide, the DORA Guide, the SWIFT Customer Security Programme (CSP) Guide and the PCI DSS 4.0 Guide.

Designing one control library

The single biggest saving is a unified control set. Pick a spine, write each control once with an owner, a test procedure and an evidence source, then map it outward. NIST CSF 2.0 works well as the spine because its six functions (Govern, Identify, Protect, Detect, Respond, Recover) are regime-neutral and the FFIEC points to it, and NIST SP 800-53 Rev. 5 supplies control-level detail where a regime needs it. Some banks use the CRI Profile, which the FFIEC also lists, as a sector-tuned version.

Programme domainNIST CSF 2.0SP 800-53 Rev. 5DORASWIFT CSCF objectivePCI DSS v4.0.1
Governance, risk appetite, board reportingGV.OC, GV.RMPM familyArt. 5 and 6 (ICT risk framework)All, via attestation sign-offReq. 12
Asset and data inventoryID.AMCM-8ICT asset identificationSecure your environment (scope by architecture type)Req. 12.5 scope, Req. 1 network diagrams
Identity, MFA, least privilegePR.AAIA-2, IA-5, AC-6Access and authentication policiesKnow and limit accessReq. 7 and 8
Segmentation and secure configurationPR.PS, PR.IRSC-7, CM-6, CM-7ICT systems protectionSecure your environment (secure zone)Req. 1 and 2
Monitoring and detectionDE.CMAU-6, SI-4Detection of anomalous activityDetect and respondReq. 10 and 11
Incident response and reportingRS.MA, RS.COIR-4, IR-6Chapter III incident management and reportingDetect and respondReq. 12.10
Resilience, backup, recoveryRC.RPCP-9, CP-10Business continuity, backup and recoveryDetect and respond (recovery)Req. 12.10 (response plan)
Third-party and supply chain riskGV.SCSR-6, SA-9Chapter V ICT third-party risk, register of informationService provider and back-office scopeReq. 12.8 and 12.9
Testing and assuranceID.IMCA-2, CA-8Chapter IV testing, TLPTIndependent assessmentReq. 11 (scans, penetration testing)
  • Write controls at the lowest common detail: where SWIFT and PCI DSS ask for MFA on administrative access to their environments, write one MFA control with scope covering both, not two.
  • Keep regime-specific controls tagged: some requirements have no equivalent elsewhere (for example SWIFT architecture-type decisions or PCI DSS cardholder data rendering). Tag them so they do not get lost in the merge.
  • One owner per control: assign owners in the first line and testers in the second line. A control owned by "compliance" has no owner.
  • Version the mapping: each annual CSCF update or PCI revision triggers a mapping review, not a rebuild.

Governance and board accountability

DORA Article 5(2) sets the clearest expectation: the management body defines, approves, oversees and is responsible for the ICT risk management framework, sets the digital operational resilience strategy and ICT risk tolerance, approves and reviews business continuity and response and recovery plans, approves ICT audit plans, allocates budget including ICT skills training, and approves the policy on ICT third-party services. US examiners look for similar board engagement through the FFIEC handbook.

  • Risk appetite in numbers: express ICT risk tolerance as measurable limits, for example maximum tolerable downtime for payments, card authorisation and SWIFT messaging, and maximum open critical vulnerabilities on internet-facing systems.
  • Three lines that actually work: technology and security run controls; technology risk challenges and owns the control library; internal audit tests independently. SWIFT accepts internal second or third line assessment for its independent assessment, which only works if that line is genuinely independent of the operators.
  • Board reporting pack: one quarterly pack covering risk against appetite, major incidents and near misses, third-party concentration, testing results and open findings from every regime. Separate packs for each regime fragment the board's view.
  • Training: DORA expects ICT skills training for the management body itself, not only staff awareness. Record it.

One incident process, several clocks

The incident process is where duplication hurts most, because each regime runs its own clock from a slightly different trigger. Build a single classification step early in the response that answers every regime's question at once.

ObligationTriggerClockRecipient
US banking organisation noticeNotification incident: a computer-security incident with actual harm that is reasonably likely to materially disrupt or degrade operations, core business lines or critical operations, or threaten financial stabilityNo later than 36 hours after determinationPrimary federal regulator (OCC, Federal Reserve or FDIC)
US bank service provider noticeIncident that has disrupted or is reasonably likely to disrupt covered services for four or more hoursAs soon as possibleBank-designated contact, or CEO and CIO
DORA major ICT-related incidentIncident classified as major under the DORA criteria and technical standardsInitial, intermediate and final reports on the timelines set in the technical standardsCompetent authority
SWIFT, card schemes, privacy lawPer scheme rules and law (for example GDPR personal data breach)Per ruleSWIFT, card brands and acquirer, data protection authority
  1. Define the determination point: the US clock runs from when the bank determines a notification incident occurred, so record that decision, who made it and when.
  2. Pre-draft notices: the US rule requires no specific form and allows email or phone; keep contact lists and templates for each recipient current.
  3. Bind providers by contract: require service providers to notify a named contact, aligned to the four-hour threshold in the US rule and the information your DORA reports need.
  4. Exercise it: run tabletop exercises with legal, communications and the business, using the FDIC Cyber Challenge or FS-ISAC CAPS scenarios the FFIEC lists.

See the US Bank Computer-Security Incident Notification Rule Guide and the DORA Major ICT Incident Reporting Guide.

Third parties, concentration and the register of information

Banks depend on a small number of core banking, card processing, cloud and messaging providers. DORA requires a register of information on all contractual arrangements with ICT third-party service providers, and its oversight framework designates critical ICT third-party providers. The US incident rule puts direct notification duties on bank service providers. PCI DSS Requirement 12.8 asks for a list of third-party service providers and their responsibilities.

  • One inventory: a single third-party register feeds DORA's register of information, PCI DSS 12.8 and FFIEC vendor management. Record the services supported, data handled, critical function supported and exit plan.
  • Concentration view: show which critical functions depend on the same provider or cloud region. This is what supervisors ask about after a large outage.
  • Contract clauses: security requirements, incident notification, audit and access rights, subcontracting and termination assistance.
  • Connections, not just companies: inventory every integration credential a provider holds into your estate: VPN accounts, API keys, OAuth grants and file transfer keys.

See the DORA ICT Third-Party Risk and Register of Information Guide.

Where identity fits

Every regime in this stack treats access control as central, and in a bank the hardest accounts to govern are not people. Payment hubs, SWIFT middleware, card switches, batch schedulers and cloud workloads run under service accounts, certificates and API keys that often have broad, standing access. PCI DSS v4.0.1 Requirement 8.6 added specific controls for system and application accounts. The Capital One breach in 2019 shows the pattern: a firewall's cloud role held far more data access than it needed, its temporary credentials were obtained from outside, and the OCC later imposed an $80 million civil money penalty citing cloud risk management weaknesses.

  • Inventory machine identities per critical function: for payments, card authorisation and SWIFT messaging, list every service account, key and certificate in the path.
  • Remove standing privilege: vault credentials, rotate them, and use just-in-time elevation for administrators in the SWIFT secure zone and cardholder data environment.
  • Phishing-resistant MFA for operators: payment operators and administrators are prime targets; the FFIEC's 2021 authentication guidance asks for risk-based authentication and access controls for all users including employees and third parties.

See the PCI DSS Requirements for Service Accounts and NHIs Guide (8.6) and the Online Banking Security Guide.

Testing calendar and assurance

Testing obligations overlap heavily, so plan them as one calendar. PCI DSS requires internal and external vulnerability scans quarterly and penetration testing at least annually. SWIFT requires annual attestation backed by independent assessment. DORA requires a digital operational resilience testing programme for all in-scope entities and threat-led penetration testing (TLPT) for entities identified by the competent authority, at least every three years under Article 26.

ActivityRegimes servedCoordination tip
Quarterly vulnerability scanningPCI DSS, DORA testing programme, FFIECOne scanning platform, scope tagged by regime
Annual penetration testPCI DSS, SWIFT evidence, FFIECInclude SWIFT secure zone and payment hubs in the same engagement
Threat-led penetration testDORA (selected entities), TIBER-EU style programmesTreat findings as programme input; retest key paths after remediation
Independent CSCF assessmentSWIFTSchedule before the 31 December attestation deadline; reuse PCI and audit evidence
Business continuity and recovery testsDORA, FFIEC, PCI DSS 12.10Test recovery of payments from backups, not only failover

Implementing the unified programme

  1. Months 0 to 2, baseline: list every regime that applies, its scope and its next deadline. Build the inventory of critical functions, supporting systems and third parties.
  2. Months 2 to 4, control library: adopt NIST CSF 2.0 as the spine, write controls with owners, test procedures and evidence sources, then map to each regime. Pilot with one function, such as SWIFT messaging, where scope is well defined.
  3. Months 4 to 6, evidence automation: pull evidence from systems of record (identity platform, vulnerability scanner, SIEM, CMDB) rather than screenshots. Run the next SWIFT attestation from the unified library in parallel with the old method to prove it.
  4. Months 6 to 9, incident and third parties: merge incident classification and notification, and consolidate the third-party register.
  5. Months 9 to 12, testing calendar and board pack: one calendar, one quarterly pack.
  6. Rollback: keep the previous regime-specific workpapers for one full cycle. If an assessor rejects unified evidence, fall back for that regime and fix the mapping rather than abandoning the library.

Small versus large banks: a community bank or small credit union may not face DORA or TLPT and can run the library in a spreadsheet with CSF 2.0 and PCI DSS mapping. A large or cross-border bank needs a GRC platform, a dedicated control testing team and a mapping owner, because the number of regimes and entities makes manual mapping fail.

Measuring the programme

  • Control reuse ratio: share of regime requirements satisfied by a control also used for another regime. Rising means the library is working.
  • Evidence freshness: percentage of control evidence collected automatically within its test period.
  • Notification timeliness: time from incident detection to classification decision, and from decision to regulator notice, against the 36-hour and DORA clocks.
  • Critical third-party coverage: share of critical providers with current assessment, contract clauses and a tested exit plan.
  • Machine identity hygiene: share of service accounts in payment paths that are vaulted, owned and rotated.
  • Repeat findings: number of audit or assessor findings that recur across cycles or regimes.

Common failure patterns

  • Parallel programmes: four teams answer the same MFA question four different ways, and assessors find the inconsistency.
  • Scope drift: the SWIFT secure zone or cardholder data environment grows through new integrations that never get added to scope.
  • Over-privileged cloud and service roles: workloads in payment and data paths hold far more access than they need, as in the Capital One case.
  • Missing the determination record: the bank cannot show when it decided an incident was notifiable, so it cannot show it met 36 hours.
  • Paper third-party risk: questionnaires are filed but the provider's actual connections and credentials into the bank are not inventoried.
  • Annual cliff edges: evidence is gathered in November for a December attestation, so controls drift for the rest of the year.

Practitioner checklist

  • List every applicable regime with scope, owner and next deadline.
  • Adopt one control spine (NIST CSF 2.0) and map each control to FFIEC, DORA, SWIFT CSCF and PCI DSS v4.0.1 as relevant.
  • Record board approval of ICT risk tolerance, strategy and third-party policy.
  • Build one incident classification step that records the determination time and drives every notification.
  • Maintain one third-party register that also lists each provider's credentials into your estate.
  • Inventory and vault service accounts, keys and certificates in payment, card and SWIFT paths.
  • Publish a single testing calendar covering scans, penetration tests, TLPT, recovery tests and CSCF assessment.
  • Automate evidence collection from systems of record.
  • Report one quarterly pack to the board covering all regimes.

Standards and references

Related NHI Mgmt Group resources: FFIEC Cybersecurity Guidance Guide · DORA Guide · SWIFT Customer Security Programme (CSP) Guide · PCI DSS 4.0 Guide. This guide summarises banking cyber security obligations and is not legal advice.

PDF for subscribers

Take this guide with you

Every section, table and checklist in one PDF, with links to the sources and related guides. Subscribe to our weekly NHI & AI Identity Journal and we’ll email you a download link.

Free weekly newsletter

Subscribe to the NHI & AI Identity Journal

The latest on NHI and Agentic AI security – articles, research, breaches, news and events every week.

Bonus 33% off our NHI Course when you subscribe.

NHIMG Editorial Note
Researched and written by the NHI Mgmt Group editorial team using AI tools, under the editorial direction of Lalit Choda. Facts are drawn from the sources listed on this page and checked by our automated quality process. Last updated 27 October 2026. Spotted an error? Use our contact page.
NHI Mgmt Group, the independent authority on Non-Human Identity, IAM and Agentic AI security. nhimg.org

PDF Downloads for Newsletter Subscribers

Get the guide

Enter your details and we’ll email you a download link.

The full guide is free to read online. The PDF is a thank you for subscribing.

Fill in all fields marked * and tick the PDF box to continue.