Procurement wants to sign a new payroll provider by Friday and asks the security team one question: is the risk acceptable? A strong questionnaire came back, certifications are current, and the vendor will still hold the salary and bank details of every employee. So a plain “yes” or “no” misses the point. What the security team owes procurement is a residual risk figure, compared against a level the business has already agreed it can live with.
Controls never remove all of a vendor’s risk, and that leftover share is what this guide measures. It explains what it is, how it differs from inherent risk, how to calculate it with a worked example across three vendors, what an acceptable level looks like, and when the number should send a vendor back for reassessment.
What is residual risk?
Residual risk is the risk that remains after controls have been applied. In third-party risk management, it is the exposure a vendor still brings once you account for its security controls, the contract terms you agreed to, and any compensating controls on your side. No set of controls takes exposure to zero, so the question is whether what remains sits within your organization’s risk appetite.
Leadership needs this figure more than any other. Inherent risk tells you how dangerous a relationship could be. What remains after controls tells you how dangerous it is now, after everyone has done their part, and so it sits behind every decision to sign, renew, restrict, or exit.
What is the difference between inherent and residual risk?
Inherent risk is the exposure a vendor relationship carries before any controls, based on what the vendor can reach and how much damage a failure could do. Residual risk is what is left after controls reduce that exposure. The first decides how closely you look at a vendor. The second decides what you do about it.
| Inherent risk | Residual | |
|---|---|---|
| What it measures | Exposure before controls | Exposure after controls |
| Driven by | Data access, system access, business criticality, regulatory scope | Inherent risk, reduced by the strength of verified controls |
| When you assess it | At intake, before due diligence | After assessment, and again whenever something changes |
| What it decides | The vendor’s tier and how deep the assessment goes | Whether to approve, add conditions, remediate, or exit |
| Who owns it | Usually the TPRM team at intake | A named risk owner, who accepts it or not |
A payroll provider and a catering supplier start far apart on inherent risk. Good controls can bring the payroll provider down a long way, but rarely to the catering supplier’s level, because the data it holds does not change. Our guide to inherent risk covers the intake side in more detail.
How do you calculate residual risk?
The common formula is residual risk = inherent risk × (1 − control effectiveness). Score inherent risk as likelihood × impact on a fixed scale, rate how well the vendor’s controls reduce it as a percentage, and apply that reduction. Some teams subtract a control score instead (inherent risk − control impact). Either works if every vendor is scored the same way.
The five steps:
- Score impact. Rate from 1 to 5 how badly a failure at this vendor would hurt you, based on the data it holds, the systems it touches, and how much of the business stops without it.
- Score likelihood. Rate from 1 to 5 how likely a security failure is before controls, based on the service type, the vendor’s exposure, and its threat landscape.
- Multiply for inherent risk. Impact × likelihood gives a score from 1 to 25.
- Rate control effectiveness. Estimate how much the vendor’s verified controls reduce that score, as a percentage. Count only what you have evidence for: an external scan, a current audit report, a tested process. A questionnaire answer on its own deserves a lower weight.
- Apply the reduction. Inherent score × (1 − control effectiveness) gives the residual score.
Here is a worked calculation for three vendors.
| Vendor | Impact | Likelihood | Inherent score | Control effectiveness | Residual score |
|---|---|---|---|---|---|
| Payroll provider (all employee financial data) | 5 | 4 | 20 | 70% (SOC 2 Type 2 report, clean external scan, MFA confirmed) | 6.0 |
| Managed IT provider (remote access to your network) | 5 | 4 | 20 | 40% (questionnaire only, remote access tool on an outdated version) | 12.0 |
| Event catering supplier (names on a guest list) | 1 | 2 | 2 | 20% (no evidence) | 1.6 |
The payroll provider and the managed IT provider start at the same inherent score. Verified controls bring the payroll provider down to 6, while the IT provider sits at 12 because most of its controls are self-reported and the one observable fact points the wrong way. Weak controls still leave the catering supplier with the lowest result, because there is little to lose. Spending assessment time on its controls would be effort in the wrong place.
None of these numbers is precise, and that is fine. The point of the scoring is to rank vendors consistently and explain decisions, so consistency across vendors matters more than precision or the exact scale you choose.
What is an acceptable level of residual risk?
An acceptable level is whatever your organization has decided, in advance, it is willing to carry for a given type of relationship. It comes from the risk appetite and tolerance set by leadership. Write it down as thresholds tied to actions, so a score of 12 triggers the same response no matter who is reviewing the vendor.
Frameworks expect this to be explicit. NIST CSF 2.0, released in February 2024, asks organizations to establish and communicate risk appetite and risk tolerance statements as part of its Govern function, which also covers supply chain risk. ISO/IEC 27001:2022 requires risk owners to approve the risk treatment plan and accept the residual information security risks. In both cases, someone with authority makes the acceptance on the record.
A simple band structure for the 1 to 25 scale used above:
| Residual score | Action |
|---|---|
| 1 to 4 | Accept. Review at the normal cycle for the vendor’s tier |
| 5 to 9 | Accept with conditions: named owner, monitoring, a contract clause on incident notification |
| 10 to 15 | Remediate before approval or renewal, or apply compensating controls and get a documented acceptance with an expiry date |
| 16 to 25 | Do not approve as scoped. Reduce the vendor’s access, change the service, or choose another supplier |
On this scale, the managed IT provider from the example lands in the remediation band. The right next step is a remediation plan for the remote access tool, plus interim mitigation such as restricting when and from where that access works. Our guide to remediation vs. mitigation covers how to run that with a vendor.
When should residual risk trigger a reassessment?
Reassess a vendor when something changes the inputs to its score, not only when the calendar says so. Controls weaken, access grows, and threats arrive between annual reviews. A figure calculated in January can be wrong by March if nobody is watching what it was built on.
Common triggers:
- A new or exploited vulnerability in technology the vendor exposes, especially internet-facing systems and remote access tools.
- A security incident at the vendor or at one of its own key suppliers.
- A change in scope, such as new data shared, a new integration, or a new business unit using the service.
- Expired evidence, such as an audit report past its period or a certification not renewed.
- Corporate change, including an acquisition, a new hosting provider, or a move of operations to a new region.
- A score drifting into a higher band through continuous monitoring.
Each trigger should map to a defined response. A new exploited flaw at a critical vendor might mean a targeted question within days; a change of hosting provider might mean a partial reassessment at the next renewal.
What are the options for treating what is left?
Four treatments cover almost every case: accept, reduce, transfer, or avoid. Accepting means a named owner agrees to carry the exposure until a set date. Reducing means remediation by the vendor or mitigation on either side. Transferring means shifting part of the financial impact through contract terms or cyber insurance. Avoiding means changing the service or ending the relationship.
Transfer needs care. Insurance and indemnity clauses move some of the financial cost, but they do nothing to the chance of an incident or the operational damage, and regulators still hold you responsible for overseeing the vendor. Use transfer alongside reduction, never instead of it.
Where do vendor risk scores go wrong after controls?
Most scores go wrong in three ways: they trust self-reported controls too much, they go stale between reviews, and they average away the one finding that matters. All three make the result look better than reality, which is the most dangerous direction for an error.
Self-reported controls get full credit. A questionnaire says MFA is enforced everywhere, and the score drops accordingly, while an external scan shows a remote access portal without it. Weight controls by the evidence behind them.
Scores freeze at onboarding. The vendor that scored 6 at signing has changed providers twice since. Without monitoring, the score describes a company that no longer exists.
Averages hide outliers. A vendor with nine strong controls and one exposed, exploited system does not deserve a “mostly good” score. Let a critical finding hold the score up until it is fixed.
How does context change the score?
Third-party cyber risk management treats what remains after controls as a property of the relationship, not of the vendor alone. The same supplier, with the same controls, carries a different exposure for a bank that sends it customer records than for a manufacturer that uses it for marketing analytics. Contextual risk scoring builds that difference in, weighting what is observed about a vendor by what the vendor can reach in your business.
Context is what makes the number useful for decisions. Combined with continuous monitoring and remediation tracking, a contextual score shows which vendors moved, why, and which ones need a conversation this week. Our guide to third-party risk management covers the program around the score, and our piece on contextual risk scoring explains the method.
Want vendor risk scored by what each vendor can reach in your business? See how Panorays approaches third-party cyber risk management.
Residual Risk FAQs
How do you calculate residual risk?
Calculate inherent risk first, as impact × likelihood on a fixed scale such as 1 to 5 each. Then estimate how much the vendor’s verified controls reduce that score, as a percentage, and apply it: residual risk = inherent risk × (1 − control effectiveness). A vendor with an inherent score of 20 and controls rated 70% effective ends at 6. Score every vendor the same way so results are comparable.
What is a residual risk rating?
A residual risk rating is the score or band a vendor receives after controls are taken into account, such as low, medium, high, or a number on a 1 to 25 scale. It is compared with the organization’s risk appetite to decide the next step: accept, accept with conditions, remediate, or do not approve. Ratings are most useful when each band is tied to a defined action.
Who is responsible for accepting residual risk?
A named risk owner, usually the business owner of the vendor relationship or a senior leader, accepts it on behalf of the organization. The TPRM or security team calculates the score and recommends a decision. ISO/IEC 27001:2022 requires risk owners to accept residual information security risks, and good practice adds an expiry date so the acceptance is reviewed rather than left in place indefinitely.
Can residual risk ever be zero?
No. Every vendor relationship carries some exposure after controls, because controls can fail, new vulnerabilities appear, and the vendor’s own suppliers add risk you cannot fully see. The goal is to bring what remains within your agreed appetite and keep it there through monitoring and reassessment, and to know which risks you have decided to carry.
What is an example of residual risk with a vendor?
A payroll provider with strong, audited security controls still holds every employee’s salary and bank details. Its controls reduce the chance of a breach, but the impact of one would stay high. That remaining exposure, after the vendor’s controls and your contract terms are counted, is what the risk owner weighs when approving the relationship.