Join our Newsletter — 33% off our NHI Course
Home FAQ Cyber Security How should teams secure on-prem software when the…
Cyber Security

How should teams secure on-prem software when the vendor still manages part of the deployment?

← Back to all FAQ
By NHI Mgmt Group Editorial Team Updated September 23, 2026 Domain: Cyber Security

Treat the deployment as a shared responsibility model. The purchaser controls infrastructure and local security boundaries, while the vendor must still provide operational support, secure configuration guidance, and application-layer protections. Success depends on clear integration with the customer’s existing controls, especially identity, logging, and network segmentation. Simply shipping code is not enough when the vendor continues to operate the service.

What shared responsibility looks like in hybrid software operations

When a vendor still manages part of an on-prem deployment, the security model is no longer “vendor code, customer controls” in the simple sense. The deployment has two operating boundaries: the customer’s infrastructure, network, and local access controls, and the vendor’s supported configuration, patching, and service-layer behavior. Teams should define which side owns each control plane function, then verify that the vendor’s operational tasks fit inside the customer’s security architecture rather than bypass it.

This is why integration details matter as much as the software itself. If the vendor touches production systems, the deployment must still inherit the customer’s expectations for authentication, logging, segmentation, change control, and exception handling. A supported product that cannot fit those controls creates a governance gap even if the software is technically functional.

Vendor-managed components are easiest to assess when you map them to the actual boundary they cross: infrastructure, application, or identity. The State of Non-Human Identity Security is useful here because vendor-operated access often depends on service credentials, tokens, or other non-human access paths that still need lifecycle and privilege control. For a broader operational view, NHI Lifecycle Management Guide helps teams think through provisioning, rotation, and offboarding as part of deployment governance.

Controls that must remain on the customer side

The customer does not give up responsibility just because the vendor performs some tasks remotely or maintains a hosted management plane. Local network segmentation, administrative access, logging destinations, backup boundaries, and exception approvals should stay under customer control unless there is a deliberate, documented reason to delegate them. The key question is whether the vendor can operate without receiving broader access than the business actually wants to grant.

Identity is often the most important control point in these arrangements because it determines who can administer, troubleshoot, or alter the deployment. For that reason, vendor access should be constrained, time-bounded, and auditable, with clear separation between normal support and elevated intervention. Logging needs to be trustworthy enough to reconstruct vendor actions, and segmentation needs to prevent vendor maintenance paths from becoming a general-purpose bridge into production.

That operational discipline aligns with Top 10 NHI Issues, which is especially relevant where support accounts, API keys, or automation credentials persist beyond the work they were created to do. It also connects to The Critical Gaps in Machine Identity Management report, since certificates and workload credentials are often the practical mechanism that enables vendor-maintained components to communicate securely.

Why this model fails when boundaries are vague

The main failure mode is overtrust: teams assume the vendor will protect what it can see, while the vendor assumes the customer will harden what it cannot control. That gap can leave support channels, service accounts, or remote management interfaces with more privilege than either side intended. It also creates blind spots when logs are not integrated, when network rules are temporary but never reviewed, or when the customer cannot prove what the vendor changed during support activity.

Teams should treat those gaps as design flaws rather than administrative inconveniences. If the vendor cannot explain how its access is authenticated, recorded, rotated, and revoked, then the deployment is already weaker than its contract suggests. If the customer cannot correlate vendor actions with its own monitoring, then incident response will be slow even when the software is nominally “supported.”

Standards & Framework Alignment

This section maps relevant standards and security frameworks to the operational risks and controls described in this guidance.

OWASP Non-Human Identity Top 10 address the attack and risk surface, while NIST CSF 2.0, CIS Controls v8 and NIST Zero Trust (SP 800-207) set the governance and control requirements practitioners need to meet.

FrameworkControl / ReferenceRelevance
NIST CSF 2.0GV.OV-01 — Organizational Context and Shared ResponsibilityShared responsibility must be defined across customer and vendor control boundaries.
PR.AA-01 — Identity and Access ManagementVendor support paths depend on authenticated, bounded access into the deployment.
DE.CM-01 — Continuous MonitoringVendor actions and support activity need logging and monitoring for accountability.
Recommendation — Document ownership boundaries for vendor-managed and customer-managed controls. Restrict vendor access with strong authentication and least privilege. Correlate vendor-administered changes with monitoring and audit logs.
CIS Controls v86.8 — Unprivileged Access for Service AccountsVendor-operated support and automation often rely on service credentials that should not be broadly privileged.
8.2 — Audit Log ManagementVendor intervention must be observable for troubleshooting and incident review.
Recommendation — Limit vendor and service accounts to narrowly scoped access. Centralize logs for vendor-managed actions and retain them for review.
NIST Zero Trust (SP 800-207)5.1 — Least Privilege AccessVendor-managed access should be explicitly minimized and time-bounded.
Recommendation — Apply least privilege to all vendor support and maintenance paths.
OWASP Non-Human Identity Top 10NHI-02 — Secrets and Credential LifecycleVendor-managed deployments often hinge on credentials that need rotation and revocation.
NHI-06 — Overprivileged Non-Human IdentitiesSupport accounts and automation paths can accumulate excess privilege in shared deployments.
NHI-09 — Third-Party and Supply-Chain TrustThe vendor remains part of the trust boundary when it manages deployment functions.
Recommendation — Rotate and revoke vendor credentials on a defined lifecycle. Review vendor-linked non-human identities for excess privilege. Assess vendor-operated controls as third-party trust dependencies.

Practitioner Guidance

What to verify: Require a written control map that separates customer-owned infrastructure controls from vendor-owned support actions, then verify that every vendor access path is explicitly tied to logging and revocation procedures. If the vendor needs persistent access, treat that as a higher-risk condition that demands stronger review and tighter segmentation.

Common mistake: Do not accept “the vendor manages part of it” as a substitute for control evidence. The arrangement is only secure when the vendor’s operational role still fits inside the customer’s identity, logging, and network boundaries, with no standing access that outlives the support need.

Practitioner takeaway: The security question is not whether the vendor is involved, but whether the vendor’s involvement can be bounded, observed, and removed without weakening the customer’s own control plane.

Deepen Your Knowledge

Sign up to our weekly newsletter — get 33% off our NHI Foundation Level Course

    NHIMG Editorial Note
    Reviewed and updated by the NHIMG editorial team on September 23, 2026.
    NHI Mgmt Group — the #1 independent authority on Non-Human Identity, IAM, and Agentic AI security. nhimg.org