The auditor opens the inventory, picks a row and asks: “Which risk came out of this asset?” An inventory copied from accounting’s fixed asset list or IT’s hardware dump can’t answer this question. Yet in an ISMS, risk analysis, classification and access decisions rest on this list. Below we explain, as it works in practice, at what level to group the inventory, which fields it should carry, what the asset owner does and how to keep the list current with source systems.
The difference between information assets and IT assets
ISO/IEC 27001:2022 Annex A 5.9 requires an inventory of “information and other associated assets” to be developed and maintained, with owners assigned to the assets. The emphasis is on information. A file server is an IT asset; the “customer technical drawings” on it are an information asset. Risk mostly arises from the information’s need for confidentiality, integrity and availability. The hardware is the medium carrying this information, nothing more.
Thinking in three layers makes the job easier in practice:
- Information assets: Personnel files, payroll data, customer contracts, product design data, financial reports, source code.
- Supporting assets: Applications (ERP, HR software, PDM), infrastructure (servers, storage, network devices), cloud services, user devices, printed archives.
- Supporting elements: People who hold critical knowledge, suppliers, physical areas (archive room, server room).
These layers can sit in separate lists; the real work is building the links between them. The “payroll data” row should show which application it’s processed in, where it’s hosted and which supplier accesses it. The hardware list can be kept separately. The ISMS inventory doesn’t replace it; it refers to it.
The right grouping level
A design mistake you may come across is the wrong level of detail. A list that gives every laptop and every Excel file its own row rots within a few months; nobody updates it, and the date column gives it away at the first audit. The opposite is just as useless: a single-row group called “company data” allows no distinction at all in risk analysis.
The grouping rule is simple: assets with the same owner, the same classification and the same protection needs can be gathered into a single row. 180 standard user laptops go in one row as “User laptops (standard image)”; serial numbers stay in the endpoint management tool. If the design team’s workstations carrying customer prototype data need different protection, they get a separate row. If protection differs, so does the row.
| Level | Example | Assessment |
|---|---|---|
| Too coarse | “All servers” | Risk differences invisible |
| Suitable | “ERP production database” | Owner and protection needs are clear |
| Suitable | “Standard user laptops” | Managed by a common policy |
| Too fine | “An employee’s 2024 budget file” | Unsustainable |
Inventory fields and sample rows
Ask one question for each column: which decision will this field be used in? A column not used in any report, risk calculation or access decision only creates maintenance overhead. In the field we see 25-column inventories; half the columns are empty and the other half are as they were two years ago. The core fields below are enough for most ISMSs:
| Field | Explanation | Where it’s used |
|---|---|---|
| Asset ID | A unique code that never changes (IA-012) | Risk and SoA references |
| Asset name and type | Information, application, infrastructure, service, person | Reporting |
| Asset owner | Role/title, not a person | Approval and review |
| Classification | E.g. Public, Internal, Confidential | Labeling and access |
| Location / hosting | On-premises, cloud, supplier | Supplier and backup scope |
| Supporting assets | IDs of linked applications and infrastructure | Dependency analysis |
| Contains personal data? | Yes/No, special category or not | Mapping to the KVKK inventory |
| Last review | Date and reviewer | Evidence of currency |
Two sample rows show how to fill in the fields. IA-012 · Payroll and personnel data · Information · Owner: HR Manager · Confidential · Hosting: HR software (SaaS) · Supporting: AP-07 HR application, SP-03 payroll service provider · Personal data: Yes, contains special category data. Second example: AP-02 · ERP system · Application · Owner: Finance Manager (business owner), IT Manager (technical custodian) · Internal · Hosting: on-premises virtualization cluster · Supporting: IN-04 SQL Server cluster, IN-01 virtualization platform.
The ERP row has two roles. The business owner decides who the information is opened to; the technical custodian runs the system. This is the shortest way to document that access approval belongs to the business unit, not IT; in an audit, the question “who approved this permission?” is also answered from here.
The asset owner’s responsibilities
In ISO 27001, “asset owner” means responsibility, not ownership. The owner is the person accountable for protecting the asset throughout its lifecycle; that’s why they’re usually chosen from the business unit. Tie ownership to a job title. If a person’s name is written, the inventory becomes wrong with every resignation and role change, and nobody notices.
What should go into the asset owner’s job description:
- Setting the asset’s classification and updating it when things change (Annex A 5.12).
- Approving access requests and signing off the periodic access review.
- Taking part in the impact assessment for the asset in risk analysis and, as risk owner, accepting treatment decisions.
- Starting the secure disposal or return process when the asset goes out of use (Annex A 5.11, 7.14).
- Reviewing and approving the inventory row at the set interval.
The owner can delegate operational tasks to IT. They can’t delegate accountability. How to write this distinction into the matrix is explained in detail in the ISMS RACI matrix guide.
Keeping the inventory current: source systems
An inventory kept by hand starts going stale after the first audit. The lasting solution is to feed supporting assets from automatic sources and limit manual maintenance to information asset rows. Typical sources:
- Active Directory / Entra ID: Computer and server objects, service accounts, groups.
- Endpoint management (Intune, SCCM, etc.): User devices, operating system versions, encryption status.
- Virtualization platform and CMDB: Virtual machines and dependency relationships.
- Purchasing and contract records: SaaS subscriptions and suppliers. The most reliable way to catch cloud services bought without IT knowing is the invoice list in accounting.
A periodic computer list from AD and objects that haven’t signed in for a long time can be pulled with these PowerShell commands (requires the ActiveDirectory module):
Import-Module ActiveDirectory
Get-ADComputer -Filter * -Properties OperatingSystem,LastLogonDate |
Select-Object Name,OperatingSystem,LastLogonDate |
Export-Csv -Path .\ad-computers.csv -NoTypeInformation -Encoding UTF8
# Computer accounts that haven't signed in for 90 days
Search-ADAccount -AccountInactive -ComputersOnly -TimeSpan 90.00:00:00 |
Select-Object Name,LastLogonDate
This output is pulled for reconciliation, not to fill the inventory. A server that’s in AD but not in the inventory raises the question of an “unregistered system”; a server that’s in the inventory but not in AD raises the question of an “asset missing a disposal record”. Keep the reconciliation result as a dated record; in an audit, that’s the evidence the inventory is kept alive. A warning: the LastLogonDate value is derived from an attribute replicated with a delay between domain controllers. It isn’t an exact date. Read it as an approximate indicator and interpret the 90-day threshold accordingly.
Linking the inventory to risk analysis
The inventory becomes useful on the day it becomes an input to risk analysis. In the asset-based method, threat and vulnerability pairings are assessed for each significant information asset; in the scenario-based method, scenarios refer to the assets affected. The choice of method is discussed in the comparison of asset-based and scenario-based risk assessment. Whichever you choose, inventory IDs should appear as references in the risk register.
Concrete steps for building the link:
- Add an “affected asset IDs” column to the risk register.
- Check that every information asset classed “Confidential” in the inventory appears in at least one risk entry; an asset that doesn’t shows the analysis is incomplete.
- Keep the asset’s impact value (confidentiality, integrity, availability) in the inventory and use the same value in the risk calculation; there must not be two different impact scores in two places.
- Define a rule that triggers a risk review when a new asset is added or its classification goes up.
The second step is done with a simple filter and almost always turns up gaps in the field. How to structure risk scoring and records is the subject of the ISMS risk analysis guide. To build the inventory, risk analysis and Statement of Applicability as a single traceable chain, this can be handled together with our ISO 27001 risk analysis and SoA work.
Common mistakes and checklist
Most audit findings about the inventory come from the same few mistakes:
- Writing “IT” as the owner: The owner of business information is the business unit; IT is usually only the technical custodian.
- Forgetting cloud services: SaaS applications, shared cloud storage areas and supplier portals don’t appear on the list.
- Skipping printed information: Contracts and personnel files in the archive are information assets too.
- Undated list: An inventory with no last review date or approver can’t prove it’s current.
- Classification contradicting access: The “Domain Users” group having access to a folder marked “Confidential” shows the inventory isn’t reflected in practice. Looking at NTFS permissions on the file server reveals this in five minutes.
Check before going live:
- Does every row have a title-based owner and a classification filled in?
- Are the supporting assets and suppliers of critical information assets linked?
- Are the date and result of the last reconciliation with source systems recorded?
- Is the inventory versioned and access-restricted under document control?
- Are rows containing personal data matched with the KVKK inventory?
To set up the inventory and run it together with the data classification scheme, see the IT asset inventory and data classification service page. For the official text of the control, the ISO/IEC 27001 page is the starting point.
Frequently asked questions
Can the information asset inventory and the fixed asset list be the same document?
They shouldn’t be. The fixed asset list tracks financial value and physical assets; the information asset inventory shows the information’s owner, classification and protection needs. The inventory can refer to the fixed asset list or the CMDB; it doesn’t replace it.
Does every user computer need its own row in the inventory?
No. Devices with the same owner, classification and protection needs are grouped in a single row. Device-level detail is kept in the endpoint management tool and compared with the inventory periodically.
Who should the asset owner be?
The business unit manager who knows the business value of the information and who should access it. IT is usually the technical custodian. Tie ownership to a job title, not a person’s name, so the inventory doesn’t break when staff change.
How often should the inventory be reviewed?
The standard doesn’t give a frequency; the organization sets it. Common practice is at least one full review a year, plus interim updates on events such as a new system going live, a change of supplier or an organizational change.