
An information security management system: what it is, and what must be in place after three months
An information security management system is not a folder of documents and not a software package. It is the way your organisation decides which risks it tackles, who does it and how it checks that it works. This article covers where to start, what should be in place after three months and which part of it sits with IT.
An ISMS is a management system: a fixed cycle of identifying risks, taking measures, measuring and adjusting, with an owner for each part.
Start with scope and the risk assessment, not the documents. After three months the system should be running, even if it is not yet ready for an audit.
The board owns the ISMS, not IT. We deliver the technical controls and the evidence; we do not certify anyone.
What an information security management system is
An information security management system, or ISMS, is a management system for information security. ISO/IEC 27001 sets out what such a system must do in clauses 4 to 10: define context and scope, organise leadership and roles, assess and treat risks, provide people and resources, operate the controls, measure and improve. The controls themselves are in the annex.
Three things an ISMS is not:
- A policy document. The policy is one part of it; how to write one people actually read is covered in our article on an information security policy.
- A certificate. You can have an ISMS without a certificate. The certificate is external confirmation that the system works, and that question comes later.
- A tool. Software can help with reminders and registers, but a completed tool without owners and decisions is an archive.
In the UK, Cyber Essentials is often the first thing customers ask for. It covers a set of technical basics and is a sensible baseline, but it is not a management system: it says nothing about risk ownership, measurement or review.
Start with scope and context
The first decision is what the ISMS covers. It sounds formal, and everything else depends on it.
- Which parts of the organisation, sites and processes. A scope covering only the IT department is rarely what a customer means when they ask for ISO 27001.
- Which information and systems. Customer data, HR records, finance, source code, designs.
- Who has an interest in it. Customers, regulators, insurers, your own staff, and what they require.
- Where suppliers come in. What an IT provider, a cloud provider or a payroll bureau does for you falls within your ISMS as an agreement, even if you do not carry it out yourself.
The trap on both sides: a scope so wide you finish nothing in the first year, or so narrow it falls outside what your customer is asking. Write the scope on half a page, including what is deliberately excluded and why.
Roles: who owns what
An ISMS without owners is a project that stalls as soon as its champion leaves. The standard explicitly requires the involvement of top management, and that is not a formality: the board decides which risks are acceptable and provides the time and the budget.
- The board: sets the objectives, accepts residual risks and reviews at least once a year whether the system works.
- An ISMS owner: keeps the cycle going, watches the plan and is the point of contact in an audit. In a small business this is often part of a role, but a named one.
- Risk owners: for each risk, someone who decides what happens to it. Usually a head of department or process owner, not IT.
- Information and system owners: who decides who gets access to the customer database or the finance system.
IT, in-house or outsourced, carries out a large share of the technical controls. But hand the ISMS to IT and you get a technical project, not a management system, and an auditor sees that in the first conversation with the board.
The risk assessment as the engine
Everything in an ISMS follows from the risk assessment: which controls you apply, which you leave out and why. So agree the method first and only then the risks.
- Criteria. How you rate likelihood and impact, and from what level a risk is no longer acceptable. Keep it simple: a scale of three or four steps that everyone fills in the same way works better than a refined one nobody understands.
- The register. For each risk the cause, the consequence, the owner, the current controls and the rating.
- The treatment plan. For each unacceptable risk: what you do, who does it and when. Accepting a risk is allowed, if the owner does so consciously and in writing.
Do not start with a hundred risks. Start with the information you cannot afford to lose and the processes that cannot stop, and work out from there. A register of twenty-five well-considered risks is worth more in the first year than a list of two hundred that nobody maintains.
What must be in place after three months
In three months you do not build a complete ISMS, but you do build a system that runs. This is the order we see working:
- Month 1: the board decision, the ISMS owner named, the scope on paper, interested parties and their requirements, and a first overview of information and systems with an owner per line.
- Month 2: the risk method, a first risk assessment of the core, the treatment plan, the top-level policy, and an agreement on document control: who changes, who approves, where the current version lives.
- Month 3: a first version of the statement of applicability, an incident procedure and register, the start of staff awareness, and a handful of metrics. Plus a date for the first internal audit and the first management review.
What is not there yet, and cannot be: evidence that it has worked for a while. An auditor wants to see access reviews, restore tests and incident logging running for a number of months. You cannot shorten that time, only start it as early as possible.
What IT contributes to an ISMS
A large share of the controls in the annex of ISO 27001 are technical, and they should leave a trail from day one. What IT delivers in the first months:
- A list of devices, systems and cloud services, with who manages them and who owns them.
- How access is organised: joiners and leavers, admin accounts, multi-factor authentication, and a first review of who has which rights.
- The state of patching and backups, including a restore test with a record.
- What is logged and for how long, and who looks at it.
- Which suppliers have access to data or systems, with the agreements already in place.
The rule of thumb for IT: set up every control so the evidence creates itself. A leavers process that disables the account automatically and raises a ticket produces its own evidence. A process that relies on a reminder email produces an argument.
Keeping the ISMS running
The difference between an ISMS and a project is that it is never finished. Three elements keep the cycle going:
- Measuring. A few metrics that mean something: how quickly leavers’ accounts are disabled, how many devices are behind on updates, how many incidents were reported and how many of those by staff.
- The internal audit. At planned intervals, someone independent of the work checks whether the system does what it promises. Findings go to the ISMS owner and the board.
- The management review. At least once a year the board looks at the metrics, audit results, incidents and changes in the organisation, and decides what needs to change.
On tooling: for a small organisation, a document management system, a shared risk register and the ticketing system you already have are enough. An ISMS tool helps once the number of owners grows and reminders start to matter, but it replaces no decision at all.
Where our work stops
What we do. We deliver the technical part of your ISMS: the overview of devices and systems, the set-up of access, patching, backup and logging, and the reports that show those controls work. For the part we manage, we write the procedures and keep the evidence up to date. We are ISO/IEC 27001 certified ourselves, so we also know the system from the inside.
What we do not do. We are not a certification body and not an auditor, and we certify no one. Scope, risk assessment, acceptance of residual risk and the statement of applicability are decisions for your board; we do not set or approve them. And the internal audit of controls we operate ourselves should be done by someone other than us.
Questions we get about this
The five that come up most often.
What is an information security management system?
An ISMS is a management system for information security: the fixed cycle in which your organisation identifies risks, takes measures, checks whether they work and adjusts, with an owner for each part. ISO/IEC 27001 sets out the requirements for such a system.
What is the difference between an ISMS and ISO 27001?
The ISMS is the system you set up yourself; ISO 27001 is the standard describing what it must meet. You can have an ISMS without a certificate. The certificate is confirmation by an accredited body that your system meets the standard.
How long does it take to implement an ISMS?
In three months you can have a system running: scope, roles, a first risk assessment, the policy and a first statement of applicability. What takes longer is evidence that it works, such as access reviews and restore tests that have run for a number of months. How long the whole thing takes depends on the size of the scope and on what is already in place.
Who is responsible for the ISMS?
The board. It sets the objectives, accepts residual risks and reviews at least once a year whether the system works. An ISMS owner keeps the cycle going and risk owners decide per risk. IT carries out a large share of the technical controls but does not own the system.
Can you set up our ISMS and certify it?
We deliver the technical part: implementing the controls and the evidence that they work. We do not certify; we are not a certification body and not an auditor. Scope, risk assessment and acceptance of residual risk are decisions for your own board.
Further reading
What the ISMS connects to.
Shall we get the technical part right from the start?
Tell us what the scope of your ISMS will be. We will tell you which technical controls are already there, which produce evidence and which do not yet; the decisions stay with your board.
Practical IT knowledge in your inbox
New guides on management, security and the workplace, written by the people doing the work. No sales talk, and you can unsubscribe in one click.