At best, a new cybersecurity alert should trigger immediate action. But we all know in practice that work is not always clear cut. A new alert can find itself as just the latest un-addressed number in the inbox. In an inbox-zero case, the latest new alert is the most urgent task. But in a backed-up, bottleneck situation, it may be treated as the least urgent — because the previous alerts are older. There are more alerts than can be dealt with. And, so people can start to ignore new alerts. Security alert fatigue is real — and a real problem we can assuage with the right tools.
The Impact of Alert Fatigue
A survey from IDC and FireEye found that more than one-third of IT security managers and analysts ignore security alerts when their queue is full. And this impact ripples out across the operation. More than a quarter (28%) of all alerts are never addressed, according to the 2020 State of Security Operations study from Forrester Consulting.
Alert fatigue is crushing the productivity of security experts. It’s also driving up stress levels, according to the survey. A pervasive sense of fear of missing incidents sets in and adds to stress and burnout. That drives down job satisfaction and makes the chronic industry skills shortage worse, which in turn contributes to alert fatigue. The human toll is the most important, but may be the least known, factor in the alert fatigue crisis. The people who work in a security operations center (SOC) are by far the main factor that determines the organization’s success or failure. Alert fatigue affects more than the events that trigger SOC alerts.
A single security analyst may look into around two dozen incidents every day. Each one requires gathering information, checking indicators of compromise against threat intelligence feeds and other time-consuming work.
SOC analysts are swapped out like air traffic controllers to reduce error through fatigue. But it’s not enough. What’s needed are better (smarter and more automated) tools for threat management: artificial intelligence (AI), SOAR tools and SIEM software. Yet less than half of the respondents to the FireEye survey even use such tools.
A better SIEM system can improve the picture and function as part of a larger alert fatigue solutions package. SIEM systems detect anomalies — possibly malicious behavior — and generate an alert each time they find one.
Take a proactive approach to threats
The False Positive Problem
One reason for the volume of alerts is a constant flood of false positives — alerts that don’t really point to a real problem. Some 45% of all alerts are false positives, according to the survey. An analyst takes the time to figure out if each alert is an attack in progress or a false positive. When they confirm an attack, the incident response team addresses it. The problem is that each false positive takes up time, which could be used to check out alerts that lead to stopping an attack. The SIEM software is the first place where the machines can help humans in efficient threat hunting.
Adding machine learning in SOAR tools to this process can reduce false positives because these trained algorithms analyze past alert data to give a more accurate assessment. They also shorten the time it takes to chase down incidents. AI-enhanced tools can apply multiple analytics techniques, including supervised learning and automated data mining, which speeds up the process — in some cases, from hours to minutes — and reduces errors. Machine learning tools can also sort alerts. So, it moves the alerts most likely to be false positives to the bottom of the queue.
How SOAR Cuts Down on Alert Fatigue
A SOAR platform uses scripts and tools to automate and coordinate otherwise manual processes. It represents a unified threat management system designed to make the process of responding to anomalies consistent and repeatable. To do this, it looks at guidelines for responding that are based on past responses. This gives all incident responders practical knowledge and best practices for addressing potential threats in a systematic way.
Besides solving real problems and reducing alert fatigue and SOC burnout, SOAR platforms can equate to a major return on investment (ROI) for security spending. The way better SOAR systems boost ROI and make the SOC more efficient is hard to overstate. They can shatter the negative feedback loop of alert fatigue that contributes to lower productivity, which leads to alert fatigue, and so on.
The Inbox-Zero Approach
People often talk about the ‘inbox zero’ concept when dealing with email. But an inbox-zero approach to security alerts is even more important. The solution lies in using advanced systems for detecting, sorting, investigating and addressing potential threats. It’s time for humans in the SOC to partner with AI to address alert fatigue and achieve an inbox-zero approach to managing incoming alerts.