ISO/IEC 27004:2009 Information technology — Security techniques ― Information security management ― Measurement
ISO/IEC 27004 concerns measurements relating to information security management: these are commonly known as ‘security metrics’ in the profession (if not in SC 27!).
Scope and purpose
The standard is intended to help organizations measure, report on and hence systematically improve the effectiveness of their Information Security Management Systems.
It “provides guidance on the development and use of measures and measurement in order to assess the effectiveness of an implemented information security management system (ISMS) and controls or groups of controls, as specified in ISO/IEC 27001. This would include policy, information security risk management, control objectives, controls, processes and procedures, and support the process of its revision, helping to determine whether any of the ISMS processes or controls need to be changed or improved.”
The use of ‘monitoring, measurement, analysis and evaluation’ in ISO/IEC 27001 is confusing and will be explained/expanded-on in 27004.
The following sections are key:
Information security measurement overview;
Measures and measurement development;
Data analysis and measurement results reporting;
Information Security Measurement Program evaluation and improvement.
Annex A suggests a template on which to describe a metric, while Annex B offers some worked examples.
The standard is quite detailed in terms of the mechanics of measurement processes. It laboriously describes how to collect “base measures”, use aggregation and mathematical calculations to generate “derived measures”, and then apply analytical techniques and decision criteria to create “indicators” used for ISMS management purposes. Unfortunately, it does not offer much guidance on which base measures, derived measures or indicators might actually be worth all this effort in terms of enabling management to manage information security.
Status of the standard
Although ISO/IEC JTC1/SC 27 recognizes that metrics are an essential tool for managing an ISMS, the project that initially developed the standard had long and troubled history. Hundreds of pages of comments from the national bodies were received even at the late FCD stage, few of which were resolved prior to publication: since guidance on security metrics was sorely needed, the standard was deemed ‘good enough’ to publish rather than delaying it still further.
The standard was first published in 2009.
ISO/IEC 27004 is currently being revised. A new title has been proposed (“Information security management systems — Monitoring, measurement, analysis and evaluation”). The revised version will be limited solely to supporting clause 9.1 of ISO/IEC 27001, i.e. management metrics principally, rather than measures of the security controls etc.
The revised standard is at 2nd CD stage and is looking much better than the first edition. It may be published during 2016.
An ISMS is literally worse than useless without suitable metrics (thus it is appropriate for 27001 to list 27004 as a normative or essential standard) but information security metrics are of value in all organizations regardless of whether or not they have an ISO27k ISMS in place. I understand why the revised ISO27k standard will be aligned specifically to ISO/IEC 27001 clause 9.1: this narrow scope and tight focus increases the chances of the standard being completed and published in a reasonable timeframe (a problem that plagued the original version). However, I believe that leaves a gap for a broad-scope information security metrics standard ...
I’m delighted to report that the revised standard is much more sensibly worded than the first edition, making it useful for practitioners. Furthermore, approaches such as requiring the business managers and risk owners to define their [ISMS] measurement requirements, and concepts such as a ‘measurement programme’ (which might be termed a ‘measurement system’, except that the abbreviation of Information Security Measurement System is already spoken for!), are definitely steps in the right direction, in my opinion.
The example metrics are not very helpful though e.g. example B.10.2 “Percentage of personnel who received annual information security awareness training” is a very poor measure of security awareness (it simply measures attendance not proactive participation, engagement, learning, knowledge transfer, effectiveness, quality of awareness/training sessions or whatever) and further implies that ‘annual security awareness training’ is an acceptable approach (it’s not, and anyway it conflates awareness with training).