They should do both, but not confuse them. Replacement addresses the root cause, while segmentation reduces exposure while replacement is delayed or impossible. If the component is unmaintained or cannot be updated safely, containment becomes the immediate control, but it should never become the permanent excuse for leaving the dependency in place.
Why replacement and segmentation solve different parts of the problem
Replacement and segmentation answer different questions. Replacement removes the vulnerable component and its defect surface. Segmentation limits what the component can reach while it is still present. For third-party components, that distinction matters because some dependencies can be patched or swapped quickly, while others are embedded in business flows, integrations, or vendor release cycles.
When a component is actively maintained and can be updated safely, replacement or upgrade should usually be the end state. When the component is unmaintained, broken by a fix, or too risky to swap immediately, containment becomes the practical control that reduces blast radius while teams work the long path to removal.
Teams get into trouble when they treat segmentation as a substitute for remediation. It is a control boundary, not a cure. The security objective is to reduce exposure now without normalising permanent dependency on a known weak link.
How to decide which one comes first
Use replacement as the default priority when there is a realistic path to a safe update, supported upgrade, or alternate component. Use segmentation first when immediate replacement would break production, create unacceptable instability, or take longer than the exposure can reasonably remain open. In other words, the more direct and low-risk the replacement path, the more it should be prioritised.
That decision often turns on operational evidence, not policy preference. If the component has no vendor support, no safe patch path, or no credible roadmap, then isolation, access restriction, and monitoring are the fastest meaningful reduction in risk. For a maintained component with a known fix, delaying replacement in favour of long-term containment usually just preserves avoidable exposure.
A useful way to frame this is: segmentation buys time, replacement buys closure. Good teams do both, but they distinguish between interim protection and permanent resolution. The top NHI issues pattern is similar whenever a risky dependency keeps broad access for too long, and the same logic appears in broader identity governance in IAM and IGA Basics.
What good containment looks like in practice
Containment should be narrow, explicit, and testable. The component should only be allowed to talk to the systems, APIs, data stores, and administrative paths it actually needs. If the dependency is internet-facing, shared across environments, or trusted by many services, segmentation should reduce that trust surface before the replacement project finishes.
Good containment also includes credential and key hygiene around the component. Third-party components often fail not only through code flaws but through exposed tokens, reused secrets, or overly broad service access. The attack path in many supplier incidents is not just “the component was vulnerable”, but “the component was allowed to reach too much once compromised.” That is why OWASP Non-Human Identity Top 10 and Third-Party, B2B and Contractor Access Guide are useful complements when the component is part of a vendor or integration chain.
If the dependency can authenticate to sensitive systems, segmentation should be paired with least privilege, short-lived access where feasible, and tight review of every route into production. The goal is not just to block network paths, but to prevent a compromised component from becoming a launch point for lateral movement.
Risk and Threat Considerations
Third-party components create risk in two directions at once: they can be the original weakness, and they can become the bridge to something more valuable. If a vulnerable component is left in place, attackers may target the easier compromise path and then pivot into data, admin functions, or downstream systems that were never meant to be exposed through that dependency.
Failure mechanism: The component remains reachable with more privilege, trust, or network access than its business function requires, so a flaw, stolen token, or unsupported version becomes an entry point for broader compromise.
Impact: Exposure can spread beyond the component itself, turning a local vulnerability into account abuse, data theft, service disruption, or an incident that is much harder to contain after the fact.
Standards & Framework Alignment
This section maps relevant standards and security frameworks to the operational risks and controls described in this guidance.
NIST SP 800-53 Rev 5, CIS Controls v8 and SLSA set the technical controls, while ISO/IEC 27001:2022 defines the regulatory obligations.
| Framework | Control / Reference | Relevance |
|---|---|---|
| NIST SP 800-53 Rev 5 | SI-2 — Flaw Remediation | Vulnerable components require patching or replacement once a fix exists. |
| SC-7 — Boundary Protection | Segmentation is the immediate control when a vulnerable dependency cannot be removed safely. | |
| Recommendation — Prioritise flaw remediation or component replacement once a safe update path is available. Restrict component communications to the minimum paths needed and contain exposed dependencies. | ||
| CIS Controls v8 | CIS-7 — Continuous Vulnerability Management | This topic is about reducing exposure from known third-party component weaknesses. |
| Recommendation — Track and remediate vulnerable components, and isolate them while remediation is pending. | ||
| ISO/IEC 27001:2022 | A.8.8 — Management of technical vulnerabilities | Third-party components with known weaknesses fall under technical vulnerability management. |
| Recommendation — Maintain vulnerability tracking and remediation for third-party components until they are retired. | ||
| SLSA | Supply-chain Levels for Software Artifacts | Replacement decisions are stronger when build and dependency provenance are controlled. |
| Recommendation — Use provenance and integrity controls to reduce reliance on untrusted third-party components. | ||
Practitioner Guidance
What to prioritise: If the dependency is exploitable and replacement is feasible, start the removal plan immediately; if not, prioritise blast-radius reduction, access narrowing, and monitoring before anything else.
What to verify: Confirm that the component cannot reach systems it does not need, that any credentials it uses are scoped tightly, and that there is a named owner for the replacement or retirement plan.
Common mistake: Treating segmentation as a closed ticket. If the component is still present six months later, the team has usually converted an urgent risk reduction measure into a permanent risk acceptance without saying so.
Practitioner takeaway: The right question is not whether to segment or replace first in the abstract, but whether the dependency still has enough trust and reach to make delay unsafe. If it does, segment immediately and keep replacement on the critical path.
Related resources from NHI Mgmt Group
- What do teams get wrong about third-party software components?
- How should security teams manage vulnerable open source libraries in third-party dependencies and sub-dependencies?
- How should security teams prioritise external exposures when web servers and third-party software are both in scope?
- How should security teams prioritise third-party vulnerabilities when CVE severity alone is not enough?