Skip to content
System access
ASD ISMISM-1504

ASD ISM 1504 Multi-factor authentication for human users of sensitive data online services

What it says

Multi-factor authentication is used to authenticate human users to their organisation's online services that process, store or communicate their organisation's sensitive data.

Quoted as published. Everything else on this page is written by Control Stack.

What it means

In plain English

Every person logging in to an organisation's own online services that handle its sensitive data must prove who they are with more than a password.

This control says that whenever a human user (a staff member, contractor or other person, as opposed to a system account) signs in to one of the organisation's online services, and that service processes, stores or communicates the organisation's sensitive data, the sign-in must use multi-factor authentication (MFA). MFA means the user has to present something beyond a password, such as an authenticator app code, a hardware security key or a passkey. It matters because a password on its own is the weakest link in an online service. Passwords are phished, reused across sites, guessed, or leaked in third-party breaches, and an online service is by definition reachable from anywhere. If a password is all that stands between an attacker and a webmail, file-sharing, HR, finance or customer portal that holds sensitive data, one successful phishing email is enough. Requiring a second factor means a stolen password alone does not get the attacker in. The scope is deliberate: it is about the organisation's own online services, it is about the ones that touch sensitive data, and it is about the human users who log in to them. Any service that meets that description is in scope, whether it is hosted in-house, in the cloud, or provided as software-as-a-service.

Why it matters

If MFA is not enforced on these services, a single compromised password gives an attacker the same access to sensitive data as the legitimate user. Credential phishing, password spraying and reuse of passwords leaked elsewhere are the most common ways organisations are breached, and an online service without MFA turns each of those into a direct path to the data. The likely consequences are theft or exposure of sensitive information, tampering with records, follow-on attacks launched from a trusted account (such as business email compromise), regulatory and notification obligations, and loss of trust from customers, partners and staff.

How to implement

  1. The information security manager, working with system owners, inventories every online service the organisation operates or subscribes to that processes, stores or communicates its sensitive data, and records each one in a register marked as in scope for MFA.

  2. The identity and access administrator configures the organisation's identity provider (or the service's own authentication settings) to require MFA for all human user sign-ins to each in-scope service, using an enforcement policy rather than a per-user option.

  3. The IT service desk enrols every human user in an MFA method (for example an authenticator app, hardware security key or passkey) as part of account provisioning, and verifies enrolment is complete before the user is given access to an in-scope service.

  4. The identity and access administrator disables or blocks any legacy or alternative sign-in path to in-scope services that would let a user authenticate with a password alone, such as basic authentication or older protocol endpoints.

  5. The system owner for each in-scope service runs a regular report of human user accounts without MFA enrolled or sign-ins that did not complete MFA, and works with the service desk to remediate each account or close each gap.

Keep it running

Day to day, this control lives in the identity and access configuration of each in-scope service. Someone needs to own an up-to-date list of the organisation's online services that process, store or communicate sensitive data, so that new services (including ones adopted by business units on their own) are identified and brought under the MFA requirement before people start using them.

For each of those services, MFA should be enforced by policy in the identity provider or the service itself rather than left as an opt-in setting users can decline. Onboarding of new human users should include enrolling their second factor, and offboarding should remove it along with the account. Expect a steady trickle of support tasks: lost or replaced phones, new hardware keys, re-enrolment after a device change, and temporary exceptions. Those exceptions are where the control tends to erode, so they should be time-limited, recorded and reviewed.

Periodic checks are also part of routine operation: confirm that no in-scope service still permits password-only sign-in, look for accounts that have never enrolled a second factor, and review sign-in logs for logins that bypassed MFA.

Evidence and audit

  1. Check 1 of 5What an auditor asks for

    Ask for

    Ask for the register or list of the organisation's online services that process, store or communicate its sensitive data.

    Look at

    Check that the list is current, names an owner for each service, and is not missing obvious services such as email, file sharing, HR, finance or customer portals.

    Good evidence

    A maintained register exists, covers all sensitive-data online services, and each entry is marked as requiring MFA for human users.

  2. Check 2 of 5What an auditor asks for

    Ask for

    Ask for the MFA enforcement policy configuration from the identity provider or from each in-scope service.

    Look at

    Confirm the policy applies to all human user accounts for those services and that MFA is mandatory rather than optional or self-selected.

    Good evidence

    Exported policy settings or screenshots show MFA required for every human user sign-in to each in-scope service, with no broad exclusions.

  3. Check 3 of 5What an auditor asks for

    Ask for

    Ask for a report of human user accounts on in-scope services and their MFA enrolment status.

    Look at

    Look for accounts with no second factor enrolled, and for how long they have been in that state.

    Good evidence

    Every active human user account on an in-scope service has an MFA method enrolled, or is disabled until enrolment is completed.

  4. Check 4 of 5What an auditor asks for

    Ask for

    Ask for recent sign-in logs for a sample of in-scope services.

    Look at

    Check whether any successful sign-ins by human users were completed with a password only, and whether any legacy authentication paths are still in use.

    Good evidence

    Sampled sign-ins all show a completed second factor, and no password-only or legacy-protocol logins to in-scope services appear.

  5. Check 5 of 5What an auditor asks for

    Ask for

    Ask for the record of MFA exceptions or bypasses granted for in-scope services.

    Look at

    Review why each exception was granted, who approved it, when it expires, and whether it was actually removed.

    Good evidence

    Exceptions are rare, documented, approved, time-limited, and closed on schedule, with none left open indefinitely.

Mappings

Closest match in ASD Essential Eight: E8-MF-ML1.1, Require multi-factor authentication for sensitive online services, which asks for the same thing.

ISO 27001

ISO/IEC 27001: 2 mappings

Essential Eight

ASD Essential Eight: 10 mappings

Mappings show how controls relate across frameworks. They do not imply full equivalence or certification.