Can cybersecurity offer value for money?

(Markus Spiske, Unsplash)

This article is brought to you thanks to the collaboration of The European Sting with the World Economic Forum.

Author: Paul Stokes, Co-founder, Managed Security Forum, Prevalent AI

Cybersecurity looks like a waste of money. Companies set new global records with $37 billion invested in security in 2018, with spending expected to surpass $42 billion by 2020. Gartner research found that security spending per employee doubled between 2012 and 2018. And yet the bad guys are still gaining the upper hand. Cybercrime cost the world economy $600 billion in 2017, up from $500 billion in 2014. People are starting to question whether increasing budgets actually reduce cyber incidents.

Nor is the problem is a lack of awareness. UK government research found that the perceived importance of cyber-risk among FTSE350 companies has almost tripled in five years, proving that the most powerful decision-makers in UK business are concerned about security. The UK is no outlier, either, as the 2019 WEF Global Risk Report has cyberattacks occupying the fourth and fifth spots for international risks most likely to increase.

When we rule out investment and intent as the reasons behind our collective failure at security, we are left with implementation. Is there something wrong with the way that well-meaning professionals invest growing security budgets that prevents us from improving our cyber-resilience?

The first thing to consider is how cyberdefences are formed. A company’s security apparatus typically grows iteratively, with employees, tools and procedures added in response to changing budgets, threats and regulations. It is easier in the short-term to deal with emergent threats reactively rather than revisit the entire security strategy. Over time, this has led to an excessive number of tools, many of them point solutions, and this progresses into security teams that are overwhelmed by alerts, lacking a cohesive strategy and in a constant state of firefighting.

Addressing this vicious circle requires an adaptive framework that is agreed by a company’s senior management and implemented across the organisation. The most effective approach is a structured cyber-risk management framework, where security threats are expressed in terms of financial risk to a business and the cost of mitigation. When used correctly, it leads to security spending being informed by risk specialists in the C-suite such as the CFO, CRO and even CEO. The next progression to make spending more efficient is to track the return on investment in cyberdefences, to allow companies to better prioritize investment and provide more accurate reporting on their security posture.

Cyber-risk management stems from a philosophy of oversight that demands metrics to inform decisions as well as to assess them in retrospect. Security has historically struggled with this concept, as chief information security officers (CISOs) who have not suffered major incidents are unable to make the case for greater investment, and those who weathered breaches and attacks are forced to pay “security penance” and overspend in the aftermath. Measuring return on investment resolves both situations.

The first step to measuring ROI is to quantify security risks. One approach to this is to use a Value at Risk (VaR) approach, such as the FAIR framework. The results of the quantification can then be represented in a loss exceedance curve. These plot the likelihood of sustaining a loss in a given year against the financial cost that would be sustained. In other words, the loss exceeded by an event compared with the chances of that event taking place.

The second step is to compare the forecast losses against a company’s agreed risk appetite, enabling the organisation to decide where to invest in mitigations to reduce the risk to an acceptable level. Finally, once these mitigations have been implemented, the same risks are then quantified again using the VaR approach, comparing before and after, in order to assess the return on investing in the mitigations. This is best illustrated using a loss exceedance curve:

The above example can be used to determine ROI on an investment in security. This is done by plotting the inherent risk before the investment took place, using a VaR model to quantify it, and then comparing it with the agreed risk tolerance. If the risk is above the tolerance, then after any mitigations are in place, the risk is calculated again. The before and after curves are then compared to determine ROI.

Only a quantitative approach such as this can ensure budgets are spent in the most effective way, with the highest impact investments. It prevents reactive spending by building oversight into the investment process, and taps into senior expertise by presenting security in the language of business and risk. This approach can then be used to continuously monitor security projects as they progress, both through implementation and then over time once they are live.

Most organizations rely heavily on static tests such as vulnerability scans and penetration tests to determine cyber-risk. These are useful exercises, but are not enough in themselves.

Qualitative assessments based on subjective factors, another common approach, often do more harm than good when measuring ROI. Some CISOs protest that quantitative data is not always available, and soft scoring can provide a “feel” for security posture, but these methods rarely alleviate a lack of data in a highly complex field, while they can often obscure it.

Dynamic, quantitative data is required to provide boards and executives with the information needed to make informed decisions about security investment. This quantitative data needs to be up to date. As ESG’s Jon Oltsik explains in Oltsik’s Law, you cannot measure a dynamic environment with static data. Risk scoring, and resulting Return on Investment analysis, need to be based on continuously updated data.

UK government research shows that just one in three FTSE350 companies has even agreed security risk appetite and communicated it to staff. And that is based on self-reporting; the real number is likely to be lower. There is still a lot more waste than measurement in security. Yet it is clear that change is under way, with Gartner naming risk appetite statements as one of its top seven security and risk management trends for 2019.

Cyber-resilience is a moving target, and sometimes it can be hard to tell if a company is even moving in the right direction. Determining ROI allows for tangible, clearly defined measurement of progress that is understood throughout the organisation. For some companies this means demanding continuous risk reporting on security initiatives with full board oversight, while others might start by progressing from qualitative heat maps to quantifiable risk metrics.

The ultimate goal of cyber ROI is to maximize investment and reduce risk. This includes tapping into the expertise of the board and building towards a shared language of risk and ROI, which can only be done when risk and ROI are accurately and continuously measured. When done correctly, the value of security investment is plain to see.