I have heard of military operations that were clumsy but swift, but I have never seen one that was skillful and lasted a long time. Master Sun (Chapter 2 – Doing Battle, the Art of War).
The GRC (governance, risk and compliance) market is driven by three factors: government regulation such as Sarbanes-Oxley, industry compliance such as PCI DSS 1.2 and growing numbers of data security breaches and Internet acceptable usage violations in the workplace. $14BN a year is spent in the US alone on corporate-governance-related IT spending .
It’s a space that’s hard to ignore.
Are large internally-focused GRC systems the solution for improving risk and compliance? Or should we go outside the organization to look for risks we’ve never thought about and discover new links and interdependencies .
This article introduces a practical approach that will help the CISOs/CSOs in any sized business unit successfully improve compliance and reduce information value at risk. We call this approach “GRC 2.0” and base it on 3 principles.
1. Adopt a standard language of GRC
2. Learn to speak the language fluently
3. Go green – recycle your risk and compliance
GRC (Governance, Risk and Compliance) was first coined by Michael Rasmussen. GRC products like Oracle GRC Suite and Sword Achiever, cost in the high six figures and enable large enterprises to automate the workflow and documentation management associated with costly and complex GRC activities.GRC - an opportunity to improve business process
GRC regulation comes in 3 flavors: government legislation, industry regulation and vendor-neutral security standards. Government legislation such as SOX, GLBA, HIPAA and EU Privacy laws were enacted to protect the consumer by requiring better governance and a top-down risk analysis process. PCI DSS 1.2; a prominent example of Industry regulation, was written to protect the card associations by requiring merchants and processors to use a set of security controls for the credit card number with no risk analysis. The vendor-neutral standard, ISO27001 helps protect information assets using a comprehensive set of people, process and technical controls with an audit focus.
The COSO view is that GRC is an opportunity to improve the operation:
"If the internal control system is implemented only to prevent fraud and comply with laws and regulations, then an important opportunity is missed...the same internal controls can also be used to systematically improve businesses, particularly in regard to effectiveness and efficiency."GRC 2.0
The COSO position makes sense, but in practice it's difficult to attain process improvement through enterprise GRC management.
Unlike ERP, GRC lacks generally accepted principles and metrics. Where finance managers routinely use VaR (value at risk) calculations, information security managers are uncomfortable with assessing risk in financial measures. The finance department has quarterly close but information security staffers fight a battle that ebbs and flows and never ends. This creates silos - IT governance for the IT staff and consultants and a fraud committee for the finance staff and auditors.
GRC 1.0 assumes a fixed structure of systems and controls. The problem is that, in reducing the organization to passive executives of defense rules in their procedures and firewalls, we ignore the extreme ways in which attack patterns change over time. Any control policy that is presumed optimal today is likely to be obsolete tomorrow. Learning about changes must be at the heart of day-to-day GRC management.
A fixed control model of GRC is flawed because it disregards a key feature of security and fraud attacks - namely that both attackers and defenders have imperfect knowledge in making their decisions. Recognizing that our knowledge is imperfect is the key to solving this problem. The goal of the CSO/CISO should be to develop a more insightful approach to GRC management.
The first step to making that happen is to get everyone speaking the same language.Adopt a standard language of GRC - the threat analysis base class
We formalize this language using a threat analysis base class which (like any other class), has attributes and methods. Attributes have two sub-types - threat entities and people entities.Threat entities
Assets have value, fixed or variable in Dollar, Euro, and Rupee etc. Examples of assets are employees and intellectual property contained in an office.
Vulnerabilities are weaknesses or a lacking in the business. For example - a wood office building with a weak foundation built in an earthquake zone.
Threats exploit vulnerabilities to cause damage to assets. For example - an earthquake is a threat to the employees and intellectual property stored on servers in the building.
Countermeasures have a cost, fixed are variable and mitigate the vulnerability. For example - relocating the building and using a private cloud service to store the IP.People entities
Business decision makers encounter vulnerabilities and threats that damage company assets in their business unit. In a process of continuous interaction and discovery, risk is part of the cost of doing business.
Attackers create threats and exploit vulnerabilities to damage the business unit. Some do it for the notoriety, some for the money and some do it for the sales channel.
Consultants assess risk and recommend countermeasures. It's all about the billable hours.
Vendors provide security countermeasures. The effectiveness of vendor technologies is poorly understood and often masked with marketing rhetoric and pseudo-science.Methods
The threat analysis base class prescribes 4 methods:
- SetThreatProbability -estimated annual rate of occurrence of the threat
- SetThreatDamageToAsset - estimated damage to asset value in a percentage
- SetCountermeasureEffectiveness - estimated effectiveness of the countermeasure in a percentage.
A language with 8 words is not hard to learn, it's easily accepted by CFO, CIO and CISO since these are familiar business terms.
The application of our 8 word language is also straightforward.
Instances of the threat analysis base class are "threat models" - and can be used in the entire gamut of GRC activities: Sarbanes-Oxley, which requires a top down risk analysis of controls, ISO27001 - controls are countermeasures that map nicely to vulnerabilities and threats (you bring the assets) and PCI DSS 1.2 - the PAN is an asset, the threats are criminals who collude with employees to steal cards and the countermeasures are specified by the standard.
You can document the threat models in your GRC system (if you have one and it supports the 8 attributes). If you don't have a GRC system, there is an excellent free piece of software to do threat modeling - available at http://www.ptatechnologies.comGo green - recycle your threat models
Leading up to the Al Qaida attack on the US in 9/11, the FBI investigated, the CIA analyzed but no one bothered to discuss the impact of Saudis learning to fly but not land airplanes.
This sort of GRC disconnect in organizations is easily resolved between silos, by the common, politically neutral language of the threat analysis base class.Summary
Effective GRC management requires neither better mathematical models nor complex enterprise software. It does require us to explore new threat models and go outside the organization to look for risks we've never thought about and discover new links and interdependencies that may threaten our business. If you follow the Tao of GRC 2.0 - it will be more than a fulfillment exercise.