Vendor questionnaires are where most third-party security programs start, and too often where they stop. The answers describe a supplier’s controls on the day the form was filled in, not on the day an attacker arrives.
This piece covers what slips past the form, what the gap costs, and how to move from attestation to evidence you can keep current.
Image placeholder: Hero image, 1200 × 630. Devensa AI branded graphic for “Beyond the Vendor Questionnaire: Evidence Over Attestation”.
Third-party security risk is the exposure an organization inherits through its vendors’ systems, credentials, and integrations. Vendor questionnaires record what a supplier says about its controls on one day; attackers exploit what those controls look like on every other day. Verizon’s 2026 DBIR found a third party involved in 48% of breaches, up 60% in a year.1
The standard security questionnaire, whether a SIG, a CAIQ, or a custom spreadsheet, asks a vendor to attest to its controls. That is useful for screening, but it is self-reported, dated the moment it is returned, and it describes intent more reliably than configuration. SecurityScorecard’s 2026 research found that 67% of organizations still rely on point-in-time audits to assess third-party threats.2
Third-party security risk reaches beyond the vendor’s own network. It includes the credentials you issued the vendor, the integrations linking their platform to yours, the software they ship, and the suppliers they depend on. It is often treated as a procurement checkpoint, a gate passed once. It functions better as a relationship monitored for as long as the access exists.
Verizon’s analysts cite missing multi-factor authentication, improper credential rotation, and absent least-privilege enforcement as core problems in third-party breaches.3 A questionnaire can ask whether a vendor requires MFA. It rarely shows whether MFA is enforced on the specific support login that touches your data.
The 2026 DBIR describes attackers exploiting vendors, SaaS platforms, and OAuth integrations.4 Connected apps often hold broad, long-lived permissions granted by one administrator years ago. They appear in no questionnaire because the risk sits in your tenant, not the vendor’s.
In Ponemon Institute and Imprivata research, 47% of organizations reported a breach or attack involving third-party network access in the past year, yet only 50% keep a comprehensive inventory of vendors with network access.5 Access that is not listed is not reviewed.
Exploited vulnerabilities overtook stolen credentials as an initial access vector in the 2026 DBIR, at 31% of breaches.1 A written patching policy says little about how fast a vendor closes an exploited flaw on an appliance sitting in front of your data.
Your vendor’s hosting provider, identity platform, and software dependencies sit outside your contract and usually outside the questionnaire. When one of them is compromised, the exposure travels downstream.
Image placeholder: Supporting chart or diagram for this section. Devensa AI branded.
IBM’s 2026 Cost of a Data Breach Report puts the global average breach at $4.99 million.6 Supply chain compromise ranked first among 30 cost factors, adding $227,250 and averaging 258 days to identify and contain.7 Leaders see the risk: 86% express concern about supply chain risk, yet 90% are confident their business could keep operating through a vendor breach.8 Those two findings are hard to reconcile without continuous evidence.
For U.S. public companies, Regulation S-K Item 106(b) asks whether the registrant has processes to oversee and identify cybersecurity risks from its use of third-party service providers, and SEC staff have said stating that a process exists is not enough.9 Vendor oversight is now a disclosed control. In Devensa AI, this is the territory of the vCISO capability, an AI-powered vOfficer augmented by Devensa subject-matter experts where needed, designed to connect vendor security evidence to the governance record a board reviews.
NIST’s Cybersecurity Framework 2.0 placed Cybersecurity Supply Chain Risk Management (GV.SC) inside its new Govern function. The category expects an established supply chain risk program, suppliers prioritized by criticality, security requirements written into contracts, and assessment and monitoring across the full relationship, including its end. Zero Trust principles carry this to the technical layer: vendor access is verified continuously, scoped to the task, and never trusted because of where it originates.
Devensa AI scores this through TruMaturity™, the methodology and six governance domains shared across Devensa. On the platform, the TruMaturity™ Assessment is designed as an AI-mediated, adaptive assessment, triggered on first sign-in and roughly annually after that, with scores mapped to the five CMMI maturity levels. For third-party security, the levels read like this:
Devensa AI scores Cybersecurity on the TruMaturity™ scale and keeps the evidence behind it current, including the vendor access, integrations, and review dates tied to each critical supplier. A virtual CISO (vCISO) on the platform tracks owners, due dates, and evidence across the vendor program.
They miss live configuration, the permissions your own tenant has granted to vendor integrations, fourth-party dependencies, and anything that changes after the answers are returned. They record attestations, which support screening but are not evidence of enforced controls.
Yes, as a screening and contractual tool. A signed attestation records what the vendor represented. The problem arises when the questionnaire is treated as the whole program rather than the first step.
The terms overlap. Third-party risk usually refers to direct vendors and service providers. Supply chain risk extends to the software components, hardware, and fourth-party providers behind them. NIST CSF 2.0 addresses both under GV.SC.
It pairs external signals, such as exposed services, leaked credentials, and breach disclosures, with internal signals, such as active vendor accounts and integration permissions. It supplements periodic reviews rather than replacing them.
Security typically owns the assessment method, while the business owner of each vendor relationship owns the decision to accept residual risk. Gartner recommends tracking those contract decisions so risk acceptance is visible.10 If a critical vendor were breached tonight, who in your organization would know first?
Devensa AI scores all six governance domains on TruMaturity™ and keeps the evidence behind each score current. Early access is open now.
Join the early access list: devensa.ai/early-access