---
title: "Third-Party Risk: What Changed Since Onboarding?"
description: "A supplier approved last year may now hold more data, more access and a new subcontractor. How to keep supplier assurance in step with the relationship."
url: https://panop.io/blog/third-party-risk-after-onboarding/
---

Article November 16, 2026 [Panop Research](https://panop.io/company) · Security research team, Panop SA

# Third-Party Risk: What Changed Since Onboarding?

A supplier approved last year may now hold more data, more access and a new subcontractor. How to keep supplier assurance in step with the relationship.

On this page

1. [The approval was reasonable, and then the relationship moved on](#the-approval-was-reasonable-and-then-the-relationship-moved-on)
2. [A trusted connection can carry the risk](#a-trusted-connection-can-carry-the-risk)
3. [Start with what the supplier can affect](#start-with-what-the-supplier-can-affect)
4. [Apply comparable rigor, with evidence that fits](#apply-comparable-rigor-with-evidence-that-fits)
5. [Review when the risk changes](#review-when-the-risk-changes)
6. [Rehearse the uncomfortable conversation](#rehearse-the-uncomfortable-conversation)
7. [Does the approval still match the relationship?](#does-the-approval-still-match-the-relationship)
8. [Where Panop fits](#where-panop-fits)
9. [Sources](#sources)

A strong internal security program can still leave an organization exposed through the businesses it depends on. Supplier assurance has to keep pace with the data, access and responsibility those relationships pick up over time.

## The approval was reasonable, and then the relationship moved on

Supplier relationships change quietly. Take an illustrative example: a growing financial services business appoints a customer support provider. Procurement checks the supplier, security reviews its controls, and the agreement limits access to a small set of customer records. The service works well.

Over the next year the provider takes on more work. Its staff can now export customer documents. A new integration connects the support platform to internal systems. A subcontractor helps cover weekends. Each change has an operational reason, and nobody brings the whole relationship back through security review.

Meanwhile the organization keeps testing its own applications and tightening access. The supplier approval still looks current in the register. Yet the arrangement now exposes more personal information, and relies on more people and systems, than the original assessment considered.

Would you accept the same access and uncertainty inside your own organization? If not, the supplier relationship deserves attention. The same goes for engineering partners holding source code, advisers receiving acquisition plans and manufacturers handling sensitive designs.

## A trusted connection can carry the risk

Microsoft’s [October 21, 2026 analysis](https://www.microsoft.com/en-us/security/blog/2026/10/21/storm-0940-uses-artex-agentic-ai-connectwise-supply-chain-compromise/) of the ConnectWise supply-chain compromise, which it attributes to a state-aligned actor it tracks as Storm-0940, is a current example. Microsoft reported that compromised application identities were used to authenticate into customer environments through access the applications already held. Some environments showed further activity involving reconnaissance, credentials and privileges. Microsoft also cautioned that a successful authentication did not, by itself, establish that customer data had been accessed or stolen.

The lesson is about the permissions behind the relationship. A supplier can become a route into your systems without any employee opening a phishing email, so controls on human sign-ins cover only part of the picture when software applications also hold access.

The exposure may also sit entirely outside your environment. Customer identity documents, product designs or trade secrets held by a supplier can be compromised there, while your own systems keep running and your customers or your business take the harm. Data entrusted to others is a security concern in its own right.

## Start with what the supplier can affect

A supplier list is useful once it has business context. For each important service, identify who handles the information, who can change systems, and whose failure would interrupt delivery. Follow material dependencies into subcontractors and shared infrastructure. NIST’s [due diligence quick-start guide](https://csrc.nist.gov/pubs/sp/1326/final) includes resilience, foundational cyber practices and supply-chain tiers among its assessment components.

Prioritize by potential harm. A small specialist with privileged access to production may deserve more attention than a much larger supplier that receives public marketing material. A supplier with no system access at all may still hold sensitive customer records or irreplaceable intellectual property.

We start with a short record for each critical relationship: the business service it supports, the data it holds, the access it has, its material subcontractors, the consequence of a compromise or outage, and the person accountable internally. Record what is unknown as well. An empty field should prompt a question, never an assumption of low risk.

Check for concentration too. Two apparently independent providers may depend on the same cloud platform, identity service or subcontractor. Your fallback has to survive the disruption you are planning for.

## Apply comparable rigor, with evidence that fits

Suppliers don’t all need identical controls or the same testing schedule. They do need requirements proportionate to the information, access and service responsibility you hand them. NIST’s [supply chain risk management practices](https://csrc.nist.gov/pubs/sp/800/161/r1/upd1/final) treat this as part of wider enterprise risk management, given how little visibility customers often have into how the technology and services they buy are built and run.

For a provider holding sensitive data, ask how access is restricted, how long information is kept, what goes to subcontractors and how deletion is verified. For a privileged integration, focus on permissions, credential management, logging and your ability to withdraw access. For a critical service, recovery evidence matters as much as prevention.

Testing evidence needs context. Which service and environment were assessed, and when? Did the test cover the access model that matches your arrangement? What was excluded? Were significant findings fixed and retested? A certificate or summary report adds assurance, and you still need to read its scope and findings.

External scanning can pick up some visible exposures and changes. It can’t tell you whether a supplier’s internal access controls work, whether its backups restore, or whether its subcontractors handle data properly. Combine independent assurance, supplier evidence and agreed testing. Intrusive testing of supplier systems needs explicit authorization and a defined scope, including permission for any affected systems run by others.

## Review when the risk changes

[NIST CSF 2.0](https://nvlpubs.nist.gov/nistpubs/CSWP/NIST.CSWP.29.pdf) separates due diligence before a supplier is appointed from understanding and monitoring supplier risk throughout the relationship. It also calls for security requirements in contracts and for planning what happens after a relationship ends. That is a workable foundation for an operating process.

Set a review cadence in proportion to the risk, then add triggers: new data categories, broader privileges, a new material subcontractor, a major product change, an incident, a change of ownership or signs that controls are slipping. Agree which changes the supplier must report, how quickly, and who receives the notification.

A reassessment should end in a decision. Keep the arrangement as it is, ask for more evidence, reduce access, require remediation, put a workaround in place or prepare an exit. Give each unresolved issue an owner and a date. Where the business accepts residual risk, make that acceptance explicit and reviewable.

None of this means a longer questionnaire every quarter. A focused conversation about one material change is often more useful. The goal is to find out whether the original decision still holds.

## Rehearse the uncomfortable conversation

The UK Financial Conduct Authority’s [operational resilience observations](https://www.fca.org.uk/publications/good-and-poor-practice/operational-resilience-insights-observations-one-year) recognized stronger testing of third parties and supply chains, while noting more work was needed on third-party vulnerabilities. Its rules apply to specified financial firms, but the practice of mapping and testing dependencies is useful anywhere.

For a critical provider, rehearse the moment it reports an incident. Who is available? What evidence can it share? Can your team suspend the integration without stopping an essential service? Who decides when it is safe to reconnect? What do you tell customers while the facts are still incomplete?

Include recovery and exit. NIST’s [CSF 2.0 implementation examples](https://www.nist.gov/document/csf-20-implementations-pdf) cover incident exercises and crisis communications with critical suppliers, and, when a relationship ends, removing supplier access and returning or disposing of your data. Turning those topics into a rehearsal is a practical way to find gaps before a live incident does.

## Does the approval still match the relationship?

Choose one critical supplier. Note its business owner and compare the arrangement approved at onboarding with what happens today. Work through the table with security and the team that uses the service, and bring in the supplier where you need evidence or an agreed test.

| Change to check | Evidence to review | Action to decide |
| --- | --- | --- |
| Customer data, IP and trade secrets | Current data flows, retention, deletion and who receives copies | Reduce unnecessary holdings or tighten handling requirements |
| Access and integrations | Application permissions, privileged accounts, credential ownership and revocation procedure | Remove excess access; test safe suspension within agreed scope |
| Subcontractors and shared dependencies | Current service chain, material changes and notification commitments | Assess new exposure, and whether the fallback shares the dependency |
| Security and testing | Assessment scope, exclusions, dates, open findings and retest evidence | Agree proportionate remediation or request missing assurance |
| Response, recovery and exit | Named contacts, joint exercise results, service restoration and access removal | Rehearse the most consequential gap and assign an owner |

Record the decision, any accepted residual risk, the action owner and the due date. Set the next review and the changes that would bring it forward. Missing evidence is a question to resolve. It is never proof that the relationship is safe.

## Where Panop fits

[Panop’s third-party risk monitoring](https://panop.io/solutions/third-party-risk) tracks the external exposure of your suppliers as their infrastructure changes. That evidence can focus a supplier conversation or an authorized validation exercise. It belongs inside a broader program that covers access, data handling, assurance, contracts and recovery, because no external exposure platform can establish the full security of a supplier relationship.

## Sources

1. Microsoft Security: [AI-driven, large-scale nation-state campaign using ARTEX enables supply chain compromise](https://www.microsoft.com/en-us/security/blog/2026/10/21/storm-0940-uses-artex-agentic-ai-connectwise-supply-chain-compromise/), October 21, 2026
2. NIST: [SP 1326, NIST Cybersecurity Supply Chain Risk Management: Due Diligence Assessment Quick-Start Guide](https://csrc.nist.gov/pubs/sp/1326/final), July 2026
3. NIST: [SP 800-161 Rev. 1, Cybersecurity Supply Chain Risk Management Practices for Systems and Organizations](https://csrc.nist.gov/pubs/sp/800/161/r1/upd1/final), May 2022, updated November 2024
4. NIST: [The NIST Cybersecurity Framework (CSF) 2.0](https://nvlpubs.nist.gov/nistpubs/CSWP/NIST.CSWP.29.pdf), CSWP 29, February 26, 2024
5. Financial Conduct Authority: [Operational resilience: insights and observations one year on](https://www.fca.org.uk/publications/good-and-poor-practice/operational-resilience-insights-observations-one-year), March 27, 2026
6. NIST: [CSF 2.0 Implementation Examples](https://www.nist.gov/document/csf-20-implementations-pdf), February 26, 2024

About the author

[Panop Research](https://panop.io/company)

Security research team, Panop SA

Panop Research is the security research function of Panop SA, a Swiss exposure management company founded in July 2024 by three practitioners with nearly twenty years each inside large international organisations. It publishes analysis built on primary public datasets, with every figure attributed and linked to the publisher that produced it.

- [LinkedIn](https://www.linkedin.com/company/panop-io)

[← Back to blog](https://panop.io/blog/)
