A practical framework for evaluating third-party vendor security posture, validating compliance claims, and managing supply chain cyber risk — with ready-to-use checklists and templates.
Third-party vendors are the most exploited attack vector in enterprise cybersecurity today. The SolarWinds, Kaseya, and MOVEit breaches all shared a common element: the attacker didn't need to breach the target organisation directly — they compromised a trusted supplier instead.
Despite this, most organisations still evaluate vendor security through self-assessment questionnaires and a brief review of certification logos. This approach confuses compliance documentation with actual security posture, and leaves organisations exposed to supply chain risks they believe they've addressed.
This whitepaper provides a structured, practical framework for vendor security assessment and compliance validation. It is designed for security leaders, procurement teams, and compliance officers who need to move beyond checkbox exercises and towards evidence-based vendor risk management.
Key principle: A vendor's security posture is not what they claim in a questionnaire. It's what they can demonstrate through evidence, what they've implemented in their infrastructure, and what they can sustain over time. This whitepaper provides the tools to verify all three.
The framework consists of six stages — from initial risk tiering through to ongoing monitoring — and includes ready-to-use checklists, scoring matrices, and contract clause templates that can be adapted to any organisation's vendor management programme.
Vendor security is no longer a best practice — it's a regulatory obligation. Three major European regulations have placed explicit requirements on supply chain security management:
| Regulation | Vendor Security Requirement | Consequence of Non-Compliance |
|---|---|---|
| NIS2 | Article 21(2)(d): supply chain security, including security-related aspects of relationships with direct suppliers and service providers | Fines up to €10M or 2% of global turnover; management liability |
| DORA | Chapter V: comprehensive ICT third-party risk management with mandatory contractual provisions | Regulatory sanctions; supervisory authority can restrict vendor relationships |
| ISO 27001:2022 | Annex A Control 5.19–5.23: supplier relationships, information security in supplier agreements, managing supply chain risk | Certification failure or withdrawal |
Supply chain attacks increased by 300% between 2021 and 2025, according to ENISA's annual threat landscape reports. Attackers have learned that compromising one vendor can provide access to hundreds of downstream organisations. The economics favour the attacker: why breach one target when you can breach one supplier and reach them all?
Real-world pattern: In most supply chain breaches, the compromised vendor had certifications, passed audits, and completed security questionnaires. The gap was between documented controls and implemented controls — and nobody was checking.
Most vendor security programmes fail in predictable ways. Understanding these patterns is the first step towards building something more effective:
Not every vendor requires the same level of scrutiny. The first step in any vendor security programme is to tier vendors by the risk they present to your organisation. This ensures assessment effort is proportional to actual exposure.
The risk tier should be determined by evaluating four factors:
| Factor | Questions to Ask | Weight |
|---|---|---|
| Data access | Does the vendor access, process, or store your organisation's data? What type — public, internal, confidential, personal data? | High |
| Network access | Does the vendor connect to your network? To which zones — corporate IT, DMZ, OT/ICS? | High |
| Business criticality | What happens if this vendor is unavailable for 24 hours? 7 days? 30 days? | Medium |
| Regulatory scope | Is this vendor in scope for NIS2, DORA, GDPR, or sector-specific regulation? | Medium |
Direct access to sensitive data or production networks. Business-critical dependency. Examples: cloud hosting provider, MSSP, ERP vendor, OT system integrator. Assessment: Full evaluation + technical validation + annual reassessment.
Access to internal data or limited network connectivity. Significant operational dependency. Examples: HR SaaS platform, managed network provider, external development team. Assessment: Full questionnaire + evidence review + biennial reassessment.
Limited data access, no network connectivity, moderate operational role. Examples: marketing automation platform, recruitment agency, facilities management. Assessment: Shortened questionnaire + certification review.
No data access, no network connectivity, easily replaceable. Examples: office supplies, catering, general consulting. Assessment: Basic due diligence only.
The security questionnaire is a starting point, not an endpoint. It should be designed to surface areas that require deeper investigation, not to produce a compliance score. The most effective questionnaires are concise, specific, and require evidence — not just yes/no answers.
| Domain | Question | Evidence Required |
|---|---|---|
| Governance | Who is accountable for information security in your organisation? What is their reporting line? | Org chart showing security function |
| Governance | Provide your information security policy with the last review/approval date. | Policy document with date stamp |
| Certifications | List all current security certifications (ISO 27001, SOC 2, etc.) with scope and expiry dates. | Certificate copies with scope statements |
| Access Control | How do you manage access to client data and environments? Describe your provisioning and de-provisioning process. | Access control procedure; sample evidence of joiner/leaver process |
| Access Control | Is multi-factor authentication enforced for all access to client environments? | Configuration screenshot or policy extract |
| Data Protection | Describe your encryption approach for data at rest and in transit, including algorithms and key management. | Technical documentation; encryption policy |
| Data Protection | Where is client data stored geographically? Can data residency be restricted by region? | Infrastructure documentation; data flow diagram |
| Vulnerability Mgmt | What is your patch management SLA for critical, high, medium, and low vulnerabilities? | Patch management policy; metrics from last 12 months |
| Vulnerability Mgmt | Do you conduct regular penetration testing? Provide the date and scope of the most recent test. | Pen test executive summary (redacted) |
| Incident Response | Provide your incident response plan. What is your client notification timeline for security incidents? | IR plan document; notification clause |
| Incident Response | Have you experienced a security incident affecting client data in the last 24 months? | Incident summary or attestation of no incidents |
| Business Continuity | Provide your business continuity and disaster recovery plan. When was it last tested? | BCP/DR plan; test report |
| Subprocessors | List all subprocessors who have access to our data, their location, and the service they provide. | Subprocessor register |
| Personnel | Do you conduct background checks on employees with access to client data? Describe your security awareness training programme. | Background check policy; training completion records |
| Network Security | Describe your network segmentation architecture. How is client data isolated from other tenants? | Network architecture diagram (high-level) |
| Logging & Monitoring | What security events do you log? What is your log retention period? Do you operate or outsource a SOC? | Logging policy; SOC details |
Tip: Send the questionnaire with a clear deadline and a named contact for questions. Allow 3–4 weeks for Tier 1 vendors. Incomplete responses are a finding in themselves — a vendor that can't describe their security controls probably doesn't have effective ones.
This is where most vendor assessment programmes stop too early. The vendor has submitted a questionnaire and provided some documentation. Now comes the critical step: verifying that what they've documented is actually implemented, effective, and current.
Evidence has different levels of reliability. Prioritise your validation effort accordingly:
| Level | Evidence Type | Reliability | Example |
|---|---|---|---|
| 1 | Vendor self-attestation | Low | "Yes, we encrypt data at rest" |
| 2 | Policy documentation | Low–Medium | Written encryption policy provided |
| 3 | Third-party certification | Medium | ISO 27001 certificate with scope statement |
| 4 | Independent audit report | Medium–High | SOC 2 Type II report with test results |
| 5 | Technical evidence | High | Configuration screenshot showing AES-256 encryption enabled |
| 6 | Your own testing | Highest | Penetration test of vendor-provided environment |
Red flag: A vendor who is unwilling to provide evidence beyond self-attestation is telling you something. Their resistance to transparency is itself a risk indicator. Document it, escalate it, and consider whether the business relationship justifies the unvalidated risk.
For Tier 1 (critical) vendors, document review is not sufficient. A technical assessment — either conducted by your team or an independent third party — provides the highest level of assurance that controls are actually implemented and functioning.
The scope depends on the vendor's access to your environment and the type of service provided. Common technical assessment activities include:
| Assessment Type | When to Use | What It Validates |
|---|---|---|
| External vulnerability scan | All Tier 1 vendors with internet-facing services | Publicly exposed vulnerabilities, outdated services, misconfigured SSL/TLS |
| Configuration review | Vendors with direct access to your infrastructure | Firewall rules, access controls, encryption settings, logging configuration |
| Penetration test | Vendors providing application services or hosting environments | Exploitable vulnerabilities in the vendor's infrastructure or application |
| Architecture review | Vendors building or managing critical infrastructure | Design quality, separation of concerns, resilience, security-by-design |
| Source code review | Vendors providing custom-developed software | Code quality, injection vulnerabilities, authentication weaknesses, hardcoded credentials |
Practical approach: You don't need to penetration test every vendor. Use risk tiering to focus technical assessments on the vendors who present the most risk. For most organisations, 3–5 vendors warrant a technical assessment. The rest can be managed through questionnaire, evidence review, and ongoing monitoring.
During a technical assessment of a vendor's environment, these are the most common findings that indicate systemic security weaknesses:
Technical assessments validate the current state. Contracts protect the future state. Security requirements must be embedded in vendor contracts to ensure that the controls validated during assessment are maintained for the duration of the relationship.
| Clause | Purpose | Recommended Language (Summary) |
|---|---|---|
| Right to audit | Enables you to verify security controls at any time | Client reserves the right to audit vendor's security controls annually, or upon reasonable suspicion of a security incident, with 30 days' notice. Vendor shall provide reasonable access and cooperation. |
| Incident notification | Ensures timely awareness of breaches | Vendor shall notify Client within 24 hours of becoming aware of any security incident that may affect Client data or services, including suspected incidents. Notification shall include nature, scope, and remediation steps. |
| Data handling | Defines permitted use and location of data | Vendor shall process Client data only as instructed, store it within [specified geography], encrypt it at rest and in transit using industry-standard algorithms, and delete it within 30 days of contract termination. |
| Subprocessor control | Extends security requirements downstream | Vendor shall not engage subprocessors without prior written approval. All subprocessors shall be bound by security obligations no less stringent than this agreement. Client shall be notified of subprocessor changes with 30 days' advance notice. |
| Security standards | Establishes minimum security baseline | Vendor shall maintain security controls aligned with [ISO 27001 / NIST CSF / IEC 62443] throughout the contract term and provide evidence of continued compliance annually. |
| Personnel security | Controls who accesses your data | Vendor shall conduct background checks on all personnel with access to Client data, provide regular security awareness training, and ensure access is restricted to personnel with a documented business need. |
| Termination & data return | Protects data at end of relationship | Upon termination, Vendor shall return or securely destroy all Client data within 30 days and provide written certification of destruction, including destruction method and scope. |
| Liability & indemnification | Allocates financial risk for breaches | Vendor shall indemnify Client against losses arising from Vendor's failure to comply with security obligations. Liability cap for security breaches shall be [specified, typically higher than general liability cap]. |
DORA-specific note: For financial sector organisations, DORA Article 30 prescribes mandatory contractual elements for ICT service providers including: clear descriptions of services, data processing locations, service level descriptions, notice periods, reporting obligations, exit strategies, and unrestricted audit rights. Ensure your contract template addresses all DORA Article 30 requirements if you fall under the regulation.
Vendor security is not a point-in-time activity. A vendor who was secure at onboarding may not be secure twelve months later. Changes in ownership, staffing, technology, or threat landscape can all erode a vendor's security posture without your knowledge — unless you're monitoring.
Beyond scheduled reviews, certain events should trigger an immediate reassessment regardless of the normal cycle:
Vendors who provide equipment, software, or services to operational technology environments present unique risks that standard IT vendor assessments don't adequately address. An OT vendor with remote access to your process control network is fundamentally different from a SaaS vendor who holds some marketing data.
Key principle for OT vendors: Assume breach. Design your controls around the assumption that the vendor's credentials, laptop, or update mechanism will be compromised at some point. Segmentation, monitoring, and session control are compensating controls for a risk you cannot fully mitigate through vendor assessment alone.
A well-designed vendor security programme satisfies requirements across multiple regulatory frameworks simultaneously. The table below maps each stage of this framework to the relevant controls in NIS2, DORA, ISO 27001, and NIST CSF.
| Framework Stage | NIS2 | DORA | ISO 27001:2022 | NIST CSF 2.0 |
|---|---|---|---|---|
| Vendor risk tiering | Art. 21(2)(d) | Art. 28(1) | A.5.19, A.5.21 | GV.SC-04 |
| Security questionnaire | Art. 21(2)(d) | Art. 28(2) | A.5.20 | GV.SC-06 |
| Evidence validation | Art. 21(2)(d) | Art. 28(3) | A.5.22 | GV.SC-07 |
| Technical assessment | Art. 21(2)(e) | Art. 28(4), Art. 26 | A.5.22, A.8.8 | ID.RA-09 |
| Contract requirements | Art. 21(2)(d) | Art. 30 | A.5.20 | GV.SC-05 |
| Ongoing monitoring | Art. 21(2)(d) | Art. 28(5), Art. 29 | A.5.22, A.5.23 | GV.SC-09 |
Efficiency gain: By implementing this single vendor management framework, you address the supply chain requirements of NIS2, DORA, ISO 27001, and NIST CSF simultaneously. One process, one evidence repository, four compliance outcomes.
Below are practical templates that can be adapted to your organisation's vendor management programme. Each template is designed to be used directly — fill in the bracketed fields and customise the scoring to match your risk appetite.
Score each domain 1–5. Weight by importance. A weighted average below 3.0 requires remediation before onboarding or contract renewal.
| Domain | Weight | Score (1–5) | Weighted Score | Evidence Quality | Notes |
|---|---|---|---|---|---|
| Security governance & organisation | 10% | [ ] | [ ] | [ ] | |
| Access control & identity management | 15% | [ ] | [ ] | [ ] | |
| Data protection & encryption | 15% | [ ] | [ ] | [ ] | |
| Vulnerability & patch management | 15% | [ ] | [ ] | [ ] | |
| Network security & segmentation | 10% | [ ] | [ ] | [ ] | |
| Incident response | 10% | [ ] | [ ] | [ ] | |
| Business continuity & DR | 10% | [ ] | [ ] | [ ] | |
| Logging & monitoring | 5% | [ ] | [ ] | [ ] | |
| Personnel security & training | 5% | [ ] | [ ] | [ ] | |
| Third-party / subprocessor management | 5% | [ ] | [ ] | [ ] | |
| TOTAL WEIGHTED SCORE | 100% | [ ] / 5.0 |
Scoring key: 1 = Not implemented, 2 = Partially implemented, 3 = Implemented but not evidenced, 4 = Implemented with evidence, 5 = Implemented, evidenced, and independently validated.
Evidence quality: A = Technical evidence / independent audit, B = Documentation provided, C = Self-attestation only, D = Not provided.
| Weighted Score | Decision | Required Actions |
|---|---|---|
| 4.0 – 5.0 | Approve | Proceed with onboarding. Schedule reassessment per tier schedule. File evidence. |
| 3.0 – 3.9 | Conditionally Approve | Approve with remediation plan. Vendor must address gaps within 90 days. Implement compensating controls in the interim. Reassess in 6 months. |
| 2.0 – 2.9 | Defer | Do not onboard until remediation is completed. Provide vendor with specific requirements. Set a 6-month review deadline. Explore alternative vendors. |
| Below 2.0 | Reject | Do not proceed. Security posture presents unacceptable risk. Document rationale and communicate to business stakeholder. Identify alternatives. |
| Finding ID | Domain | Finding Description | Severity | Remediation Required | Deadline | Status | Evidence |
|---|---|---|---|---|---|---|---|
| [F-001] | [e.g., Access Control] | [Specific finding] | [Critical/High/Medium/Low] | [Specific remediation action] | [Date] | [Open/In Progress/Closed] | [Link to evidence] |
| [F-002] | |||||||
| [F-003] |
| Field | Detail |
|---|---|
| Vendor name | [ ] |
| Vendor tier | [1 / 2 / 3 / 4] |
| Services provided | [ ] |
| Data classification | [Public / Internal / Confidential / Personal] |
| Network access | [None / Corporate IT / DMZ / OT] |
| Contract expiry | [Date] |
| Last full assessment date | [Date] |
| Current security score | [ ] / 5.0 |
| Previous security score | [ ] / 5.0 |
| Score trend | [Improving / Stable / Declining] |
| Active certifications | [ISO 27001 (exp: ), SOC 2 (exp: ), etc.] |
| Open remediation items | [Number — Critical: , High: , Medium: ] |
| Incidents during review period | [Number and summary] |
| Subprocessor changes | [Yes/No — details] |
| Recommendation | [Continue / Continue with conditions / Escalate / Terminate] |
| Assessor | [Name, Date] |
| Approved by | [Name, Date] |
Building an effective vendor security programme is not a one-time project — it's an ongoing operational capability. The following recommendations summarise the key actions for organisations at different maturity levels:
Final thought: The goal of vendor security management is not to eliminate all third-party risk — that's impossible in a connected economy. The goal is to understand your third-party risk exposure, make informed decisions about which risks to accept, mitigate, or transfer, and maintain the capability to respond when a vendor incident occurs. This framework provides the structure to do exactly that.