One of the ways that the most important countermeasures can be identified is by prioritizing the risk elements. Risks occur when a threat exploits a vulnerability. The importance of a risk can be determined by estimating its likelihood and impact. The likelihood of a risk is a reflection of how likely it is that a threat will exploit a vulnerability, and the impact identifies the damage to the organization. Risks that are highly likely to occur and will have a high impact are the most important.
Threats can negatively affect confidentiality, integrity, or availability. The severity of a threat is evaluated by identifying the likelihood that a threat will affect one of these elements, and the impact is evaluated by determining the extent to which it will be affected.
Phishing is a form of social engineering cybercrime whereby the phisher attacks its targets by posing as a trusted source, often luring the targets to provide sensitive data, for example, their bank details. Phishing can also be used to deliver malware. Individuals who are not educated in recognizing threats of this nature often lose out, especially when the attackers are able to access their personally identifiable information.
TABLE 11-2 shows a sample threat likelihood/impact matrix. This matrix can be used to determine the priority of various threats. Threats with a 0 to 10 percent likelihood of occurring would be assigned a value of 10 percent, threats with a likelihood between 11 and 50 percent would have a value of 50 percent, and threats with a likelihood between 51 and 100 percent would have a value of 100 percent. Similarly, impact values of 10, 50, or 100 would be assigned, depending on the impact to the organization.
A threat likelihood/impact matrix can be used to prioritize risks and countermeasures. Risks with higher scores would result in a higher loss and should be addressed before risks with lower scores.
Risks are evaluated based on current in-place countermeasures. For example, if an organization was not using antivirus software, the likelihood would be high that systems could become infected. If several systems became infected, the impact would also be high. A high likelihood of 100 percent times a high impact of 100 equals a score of 100.
In another example, the company has antivirus software installed on all its systems, and, in the past year, only one malware incident had caused problems after a single user had disabled the antivirus software. The malware tried to spread but was quickly detected by antivirus software on other systems. In this example, both the likelihood and impact are low, giving the occurrence of the threat a score of 1.
The numerical values assigned to the word values can be different if desired. For example, a low impact could be assigned a value of 0 instead of 10. A high likelihood could be assigned a value of 90 percent instead of 100 percent. Additionally, there can be more than three data points, and different names can be used, such as low, moderately low, moderate, moderately severe, and severe. TABLE 11-3 shows an example of how the threat likelihood/impact matrix can be used to prioritize threats. Each of the threats is assigned a likelihood and impact based on current countermeasures.
The information from Table 11-3 shows that the greatest current threats are the two with a score of 50:
These two threats would probably be given high priority for addressing their recommended countermeasures.
The attacks on DMZ servers are a threat because these servers are updated only once every six months. These updates are intended to fix bugs and vulnerabilities that have been discovered since the software was released. If the bugs aren’t fixed, the servers are vulnerable. Many attackers look for servers that do not have recent patches installed, giving this risk a high likelihood.
In this case, the solution is simple. A countermeasure would be implemented to ensure that the servers are up to date. Several ways are available to do this, but, if a risk assessment recommended a specific countermeasure and it was approved, it should be used.
Similarly, Table 11-3 indicates holes in the backup procedures. First, backups aren’t reliable. Their unreliability could be because there is no backup plan or no backup procedures, or test restores are never done to test the backups. A common countermeasure to establish the reliability of backups is to develop a backup plan and backup procedures. The plan could include a requirement to perform test restores on a weekly basis.
The threat scores aren’t necessarily perfect. They do take a little human interaction to ensure that the organization’s needs are met. For example, the threat of “loss of data due to a fire” has a score of 10. Just because this score is less than the two scores of 50 doesn’t mean it can’t be addressed earlier.
Management may decide that, even though the score is low, the impact is sufficiently high that it needs to be addressed as soon as possible. The countermeasure for this threat is simple. Store a copy of backup tapes off-site.