Organizations are detecting and containing attacks faster as the global median dwell time, defined as the duration between the start of a cyber intrusion and it being identified, was 56 days. This is 28% lower than the 78-day median observed in the previous year, according to FireEye.
Consultants attribute this trend to organizations improving their detection programs, as well as changes in attacker behaviors such as the continued rise in disruptive attacks (e.g. ransomware and cryptocurrency miners) which often have shorter dwell times than other attack types.
Global internal and external detection times have also reduced.
- Median dwell time for organizations that learned of their incident by an external party: Stands at 141 days, a 23% decrease since the previous M-Trends report (184 days).
- Median dwell time for organizations that self-detected their incident: Stands at 30 days, a 40% decrease year over year (50.5 days). While internal dwell time saw the greatest level of improvement, still 12% of investigations continue to have dwell times of greater than 700 days.
Internal detection reaches a four-year low
Although the dwell time for intrusions identified internally by organizations has gone down, the overall percentage of self-detected security incidents versus external sources has also reduced.
There has been a 12-percentage point decrease in the proportion of compromises detected internally, year-over-year. This comes after a steady increase of internal detections since 2011.
2019 is the first time in four years in which external notifications, when an outside entity informs an organization that it has been compromised, exceeded internal detections.
This shift is potentially due to a variety of factors, such as increases in law enforcement and cybersecurity vendor notifications, changes in public disclosure norms, and compliance changes.
It is unlikely that organizations’ ability to detect intrusions deteriorated, as other metrics show continued improvements in organizational detections and response.
Hundreds of new malware families identified
The report details how of all the malware families observed in 2019, 41% had never been seen before. Furthermore, 70% of the samples identified belonged to one of the five most frequently seen families, which are based on open source tools with active development.
These points demonstrate that not only are malware authors innovating, cybercriminals are also outsourcing tasks to monetize operations faster.
Also of note, the majority of new malware families impacted either Windows or multiple platforms. While new malware families solely impacted Linux or Mac, this activity remains in the minority.
More ransomware attacks
Of the attacks that professionals responded to, the greatest majority (29%) were likely motivated by direct financial gain. This includes extortion, ransom, card theft, and illicit transfers. The second most common (22%) was data theft likely in support of intellectual property or espionage end goals.
The successful monetization of ransomware attacks and the availability of ransomware as a service have contributed to an increase in overall ransomware cases. Established cybercrime groups that historically targeted personal and credit card information have also been increasingly turning to ransomware as a secondary means of generating revenue.
Given the ease with which ransomware attacks can be carried out and their continued financial success for attackers, it is expected that ransomware will continue to be used as a secondary means for monetizing access to victim environments.