Implementation Project
/
Step 6

Implementation Project

var(--variable-QsTa4u5mL)

How to Define an ISO 27001 Risk Methodology

var(--variable-Cl6cRG5u3)

Written by

Aron Lange

Published

Nov 6, 2026

Implementation Project

var(--variable-QsTa4u5mL)

How to Define an ISO 27001 Risk Methodology

var(--variable-Cl6cRG5u3)

Written by

Aron Lange

Published

Nov 6, 2026

Before anyone in your organization rates a single risk, someone has to decide how risks are rated. What counts as high? Who decides whether a risk is acceptable? If you skip this, every risk assessment ends up as a discussion about opinions.

That’s Step 6: the rules come first. You know your scope, you have management support and a policy, and in Step 5 you found out which assets your ISMS covers. Now you decide how you’ll assess and treat the risks to them. The assessment itself is Step 7, and the treatment is Step 8.

The rules come first

Here’s something that surprises many people: ISO 27001 doesn’t tell you how to assess risks. There’s no required matrix, no required scale and no required tool. A three-by-three matrix can be just as compliant as a quantitative model.

The rules come first: ISO 27001 prescribes no method, but it requires criteria set in advance, with acceptance criteria approved by management, and results that are consistent, valid and comparable

What the standard does require is this. You set your criteria in advance, before the first risk is assessed, and management approves the acceptance criteria. And your method has to produce consistent, valid and comparable results: if two people assess the same risk, they should come to the same result, and next year’s assessment can be compared with this year’s. That’s what an auditor looks for.

Clause 6.1 has three parts, and Step 6 defines all of them. In Steps 7 and 8, you run them.

1. Risks and opportunities for the ISMS (6.1.1)

When people hear risk management, they think of hackers and data breaches. But before ISO 27001 gets to those, clause 6.1.1 asks a different question: what could stop your ISMS itself from working? You consider your issues and the requirements of your interested parties from Step 1, and you determine the risks and opportunities that need to be addressed.

Risks and opportunities for the ISMS: risks like a key person leaving, a budget cut or no buy-in from the departments; opportunities like an existing management system, customer demand and existing tools

On the risk side: the ISMS knowledge sits in one person’s head, and that person leaves. The budget gets cut halfway through. The departments see the ISMS as an IT project and don’t take part. On the opportunity side: an existing management system like ISO 9001 whose processes you can reuse, a key customer demanding the certificate, which gets you management attention, or tools you already have that collect evidence automatically. None of these is about a hacker. They’re all about whether your ISMS works.

For your methodology, you decide one thing: do you handle these risks in the same process as your information security risks, or in one you already have? An ISO 9001 organization, for example, already keeps a register of risks and opportunities and can simply add the ISMS to it. Either works, as long as it’s clear where these risks are handled. And a list alone isn’t enough: you plan actions, build them into your ISMS processes, and check in the management review whether they worked.

2. The risk assessment process (6.1.2)

Two people look at the same risk. One says it’s high, the other says it’s low. Who’s right? Without agreed criteria, nobody is. Clause 6.1.2 fixes that. And notice that clause 8.2 has exactly the same title. That’s not a mistake: in 6.1.2 you define the process, in 8.2 you apply it.

The risk assessment process: risk identification, risk analysis and risk evaluation

The process has to establish the risk criteria, make sure repeated assessments give consistent, valid and comparable results, and then identify, analyse and evaluate the risks. Identify means finding the risks of losing confidentiality, integrity and availability of information in your scope, and their owners. Analyse means determining their level. Evaluate means comparing that level with your criteria.

The criteria for performing risk assessments

The level of risk combines the consequences, so how bad it would be, and the likelihood, so how likely it is to happen. ISO 27001 asks for both, and for the realistic likelihood, not the worst case you can imagine.

You can do this qualitatively, with descriptive scales like low, medium and high, which is quick but depends on judgment. Or quantitatively, with money and frequencies, which is more precise but needs data most organizations don’t have. Most use a mix: descriptive scales with numbers assigned, multiplied in a risk matrix.

Whichever way you choose, the most important part is the description of each level.

Defining impact: four levels from insignificant to catastrophic, each with thresholds for financial loss, downtime of critical processes, and legal and reputational consequences

Each impact level gets thresholds in the dimensions that matter to you: financial loss, downtime of critical processes, legal and reputational consequences. The numbers come from the organization and its top management. A loss of €100,000 is a disaster for a small company and a rounding error for a corporation. And if you classified your information in Step 5, use the same thresholds here, so that the classification and the risk assessment speak the same language.

Defining likelihood as frequency: rare, once in ten years or less often; unlikely, once every two years; possible, once a year; likely, several times a year

The same goes for the likelihood, and the simplest way is a frequency. It’s much easier to agree on “once a year” than on “possible”.

Three different methods: a startup with a three-by-three matrix in one workshop a year, a manufacturer with a four-by-four matrix with euro thresholds, downtime and safety, and a bank with a quantitative expected loss per year

And remember, the standard prescribes none of this. A startup with thirty people uses a three-by-three matrix and assesses its risks in one workshop a year. A manufacturer uses a four-by-four matrix with euro thresholds, downtime and safety. A bank calculates the expected loss per year, because it has the data and its regulator expects it. All three can meet clause 6.1.2. What matters is that the method fits the organization and is applied the same way every time.

The risk acceptance criteria

This is where the organization answers a question everyone asks sooner or later: how much security is enough? The organization decides which levels of risk it can accept, its risk appetite, and everything above has to be treated until it’s reduced to an acceptable level.

Who accepts what, approved by management before the first risk is assessed: low risks (1 to 3) accepted, medium (4 to 5) accepted and reduced where reasonable, high (6 to 9) by the risk owner with documented justification, critical (10 to 16) only with senior management approval

The acceptance criteria can also say who may accept what: low and medium risks are accepted, a high risk only by its owner with a documented justification, a critical one only with the approval of senior management. And the most important point: management approves the acceptance criteria before the first risk is assessed. Once you see the results, the temptation to move the line is big.

Consistent results and risk owners

Consistent means two people assess the same risk the same way. Valid means the results reflect reality, which only works if your levels describe real consequences. Comparable means you can put this year’s results next to next year’s. In practice, clear descriptions for each level, a few worked examples, and assessing in a workshop rather than alone get you most of the way.

And every risk gets an owner: the person with the accountability and authority to manage it. That’s usually not the IT department. If the risk is the loss of sales data, the owner is more likely the head of sales, who would suffer the loss and has the budget to act. That matters, because the risk owners approve the treatment plan and accept the residual risks.

3. The risk treatment process (6.1.3)

Once you know which risks aren’t acceptable, the question is: what now? Clause 6.1.3 defines how you answer it, and again, clause 8.3 is its twin where you apply it. The process selects the treatment options (avoid, modify, share or retain), determines the necessary controls from any source, compares them with Annex A to check nothing necessary is missing, produces the Statement of Applicability and the risk treatment plan, and gets the risk owners’ approval.

What to decide now: which treatment options and who chooses them, who approves the plan and accepts the residual risks, and which documents the process produces

In Step 6, you decide three things: which treatment options you use and who chooses them, who approves the treatment plan and accepts the residual risks, in line with your acceptance criteria, and what the Statement of Applicability and the risk treatment plan look like. Everything else happens in Step 8.

At Barvello: “What’s that going to cost me?”

Barvello Coffee Systems is the fictional company of our ISO 27001 Lead Implementer course: a family-owned maker of professional coffee machines in Munich, with about 500 employees and its own cloud platform, Barvello Connect. When project lead Nadia Berger showed the CFO her first draft of the risk criteria, he asked one question: if everything above a six has to be treated, what’s that going to cost me? That’s the moment risk criteria stop being theory.

Step 6 ran in parallel with the asset inventory, in the first two weeks of April, because the project charter had the risk assessment completed by 30 April.

Nadia’s first decision was about the risks and opportunities for the ISMS. Barvello has been ISO 9001 certified since 2004, and the Head of Quality already keeps a register of risks and opportunities for the quality management system. So Nadia didn’t build a second one. The ISMS goes into that register, and the information security risks get their own process and register.

For the impact, she didn’t invent new numbers. She took the thresholds from the classification policy, €100,000 and €5 million, and the 48 hours from the objective that critical systems are back within two days. So a major impact means a loss between €100,000 and €5 million, or critical processes down for more than 48 hours. And catastrophic includes something you won’t find in most templates: a risk of injury, because Barvello’s machines work with steam and hot water. For the likelihood and the risk level, she kept the template: a four-by-four matrix with scores from 1 to 16. An even number of levels has one advantage: there’s no comfortable middle to escape to.

Then the acceptance criteria, and the CFO’s question. Nadia showed him two risks from the gap analysis that would end up as high: the SAP restore that was never tested, and the shared office and production network. He suggested accepting high risks in general. Nadia argued against it, because then nothing below a ten would ever have to be treated. The CEO decided in the management board: low and medium risks are accepted, a high risk can be accepted by its owner with a documented justification that Nadia reviews, and critical risks only by the management board. On 13 April, the steering committee approved the criteria. Before the first risk was assessed.

Your turn: Step 6 in the project plan

In the project plan of our ISO 27001 Toolkit, Step 6 is five tasks:

ID

Task

Clause

53

Define information security risk assessment process
Deliverable: Risk assessment process

6.1.2, 4.4

54

Establish information security risk assessment criteria
Deliverable: Risk assessment criteria

6.1.2 a) 2)

55

Establish information security risk acceptance criteria
Deliverable: Risk acceptance criteria

6.1.2 a) 1)

56

Obtain approval for information security risk acceptance criteria

6.1.2 a) 1)

57

Define information security risk treatment process
Deliverable: Risk treatment process

6.1.3, 4.4

At the end of Step 6, you have three deliverables, and all three are required by ISO 27001: the risk assessment and acceptance criteria (in the toolkit’s risk management policy, POL-009), the risk assessment process (PRO-006, with the procedure SOP-002) and the risk treatment process (PRO-007, with SOP-003).

The three Step 6 deliverables, all required: risk assessment and acceptance criteria (POL-009, 6.1.2 a)), risk assessment process (PRO-006 and SOP-002, 6.1.2) and risk treatment process (PRO-007 and SOP-003, 6.1.3)

Start with the impact levels, and reuse the thresholds from your classification policy. You’re done with Step 6 when management has approved your acceptance criteria.

Four of the 19, in one step.

Here are all 143.

Both risk processes and both sets of criteria are among the 19 mandatory documents and records of ISO 27001. The free checklist lists all of them, plus the 124 auditors expect, each mapped to its clause or Annex A control.

From the auditor’s chair

Two questions decide whether a risk methodology holds up in an audit. Were the acceptance criteria approved before the first risk was assessed? And would two people using this method rate the same risk the same way? If the levels are just words without descriptions, the answer to the second question is usually no.

Frequently asked questions

Does ISO 27001 prescribe a risk assessment method?

No. There’s no required matrix, scale or tool. It requires criteria set in advance, acceptance criteria approved by management, and consistent, valid and comparable results.

What’s the difference between clause 6.1.2 and 8.2?

They have the same title. In 6.1.2, you define the risk assessment process as part of planning. In 8.2, you apply it as part of operation. The same goes for 6.1.3 and 8.3 for the risk treatment.

Who approves the risk acceptance criteria?

Management, before the first risk is assessed. Afterwards, the risk owners accept the residual risks within those criteria.

Should we use a qualitative or quantitative method?

Most organizations use a mix: descriptive scales with numbers, combined in a risk matrix. A quantitative model only makes sense if you have reliable data on losses and frequencies.

Who should be the risk owner?

The person with the accountability and authority to manage the risk, typically a member of top management, a process or department manager, or an asset owner. It’s rarely the IT department.

Step 6 in the course.

With the templates to build it.

One lecture for each part of clause 6.1, from the risks to the ISMS to the criteria and the treatment process, and a case study. Your turn: 5 tasks in the project plan and the templates for the risk management policy and both risk processes.

Information Security Risk Assessment

6.1.2

Information Security Risk Treatment

6.1.3

Template: Risk Management Policy

POL-009

Template: Risk Assessment Process

PRO-006

With your methodology approved, you’re ready for Step 7: the risk assessment, where you apply everything you’ve defined in this step.

€690

/ year

incl. 19% VAT · billed yearly

12-hour Video Course

ISO 27001 Toolkit

Practice Exam

Weekly Live Q&A

Exam Voucher + Free Retake

Bundle

Study Guide

Bundle

€690

/ year

incl. 19% VAT · billed yearly

12-hour Video Course

ISO 27001 Toolkit

Practice Exam

Weekly Live Q&A

Exam Voucher + Free Retake

Bundle

Study Guide

Bundle