Today, organizations are faced with the increasingly difficult task of trying to protect their expanding digital estate from sophisticated cybersecurity threats. Migration to the cloud and a mobile workforce has dissolved the network boundary and projected the digital estate beyond its traditional confines. Data, users, and systems are everywhere. Additionally, these systems are increasingly domiciled in the cloud and generating a considerable amount of security data. To add to this, on average, companies with over 1,000 employees maintain about 70 security products from 35 different vendors, according to a recent report by CCS Insight. The end result? A vast amount of alerts that security operations center (SOC) teams have to contend with. Unsurprisingly, according to an ESG¹ study, 44 percent of these alerts go uninvestigated due to a combination of talent scarcity and the multiplicity of security solutions generating a huge volume of alerts.
To help our customers address alert fatigue but still maintain detection efficacy, Microsoft is leveraging the power of Threat Intelligence, native solution integration, AI, and automation to deliver a unique SIEM and XDR approach—to help tackle the challenge of alert fatigue. But first things first—what exactly are alerts, events, and incidents in the context of security operations? Below is a graphic that will help answer this question before we delve deeper into how Microsoft technology is helping SOC teams sift through high volumes of alerts and narrow down to manageable high-fidelity incidents.
Let us now look at the six strategies that Microsoft employs to help our customers deal with the alert fatigue problem:
1. Threat intelligence
To combat cyberthreats, Microsoft amalgamates trillions of daily signals, across all clouds and all platforms, for a holistic view of the global security ecosystem. Using the latest in machine learning and artificial intelligence techniques—plus the power of smart humans—we put these signals to work on behalf of our customers taking automated actions when threats are detected, and providing actionable intelligence to security teams when further contextual analysis is required.
2. Native integration
Microsoft leverages the tight integration across its threat protection solution stack to help customers connect the dots between disparate threat signals and develop incidents by grouping quality alerts from different parts of their environment and stitching together the elements of a threat. First-party security solutions within the Microsoft 365 Defender offering enable our customers to benefit from real-time interactions amongst the tools, backed by insights from the Intelligent Security Graph. As a result, the quality of alerts is improved, false positives are significantly reduced at source, and in some cases, automatic remediation is completed at the threat protection level. Additionally, this can be combined with log data drawn from third-party solutions such as network firewalls and other Microsoft solutions to deliver an end-to-end investigation and remediation experience, as depicted in the image below.
3. Machine learning
The third strategy that we employ is the ingestion of billions of signals into our security information and event management (SIEM) solution (Azure Sentinel) then passing those signals through proven machine learning models. Machine Learning is at the heart of what makes Azure Sentinel a game-changer in the SOC, especially in terms of alert fatigue reduction. With Azure Sentinel we are focusing on three machine learning pillars: Fusion, Built-in Machine Learning, and “Bring your own machine learning.” Our Fusion technology uses state-of-the-art scalable learning algorithms to correlate millions of lower fidelity anomalous activities into tens of high fidelity incidents. With Fusion, Azure Sentinel can automatically detect multistage attacks by identifying combinations of anomalous behaviors and suspicious activities that are observed at various stages of the kill-chain.
On the basis of these discoveries, Azure Sentinel generates incidents that would otherwise be difficult to catch. Secondly, with built-in machine learning, we pair years of experience securing Microsoft and other large enterprises with advanced capabilities around techniques such as transferred learning to bring machine learning to the reach of our customers, allowing them to quickly identify threats that would be difficult to find using traditional methods. Thirdly, for organizations with in-house capabilities to build machine learning models, we allow them to bring those into Azure Sentinel to achieve the same end-goal of alert noise reduction in the SOC. Below is a real-life depiction captured within a certain month where machine learning in Azure Sentinel was used effectively to reduce signal noise.
Watchlists ensure that alerts with the listed entities are promoted, either by assigning them a higher severity or by alerting only on the entities defined in the watchlist. Among other use-cases, Azure Sentinel leverages Watchlists as a high-fidelity data source that can be used to reduce alert fatigue. For example, this is achieved by creating “allow” lists to suppress alerts from a group of users or devices that perform tasks that would normally trigger the alert, thereby preventing benign events from becoming alerts.
User and entity behavior analytics (UEBA) is natively built into Azure Sentinel targeting use-cases such as abuse of privileged identities, compromised entities, data exfiltration, and insider threat detection. Azure Sentinel collects logs and alerts from all of its connected data sources, then analyzes them and builds baseline behavioral profiles of your organization’s entities (users, hosts, IP addresses, applications, and more) across peer groups and time horizons. With the UEBA capability, SOC analysts are now empowered to reduce not just false positives but also false negatives. UEBA achieves this by automatically leveraging contextual and behavioral information from peers and the organization that typical alert rules tend to lack. The image below depicts how UEBA in Azure Sentinel narrows down to only the security-relevant data to improve detection efficiency:
The lower tiers of a SOC are typically tasked with triaging alerts, and this is where the critical decisions need to be made as to whether alerts are worth investigating further or not. It is also at this point that automation of well-known tasks that do not require human judgment can have the most significant impact in terms of alert noise reduction. Azure Sentinel leverages Logic Apps native to Azure to build playbooks that automate tasks of varying complexity. Using real-time automation, response teams can significantly reduce their workload by fully automating routine responses to recurring types of alerts, allowing SOC teams to concentrate more on unique alerts, analyzing patterns, or threat hunting. Below is an example of a security playbook that will open a ticket in ServiceNow and send a message to an approver. With a click of a button, if they confirm activity from a malicious IP as a true positive, then automatically that IP is blocked at the firewall level, and the user’s ID is disabled in Azure Active Directory.
We have looked at 6 effective strategies that organizations can use to minimize alert fatigue and false positives in the SOC. When combined together across a unified ecosystem including Threat Intelligence, the Microsoft Security suite, UEBA, automation, and orchestration capabilities tightly integrated with the Azure platform and Azure Sentinel alert noise can be significantly reduced. Additionally, Azure Sentinel offers capabilities such as alert grouping and the intuitive Investigation Graph which automatically surfaces prioritized alerts for investigation and also provides automated expert guidance when investigating incidents. To significantly increase your detection rates and reduce false positives while simplifying your security infrastructure, including our unique SIEM and XDR solution comprising Azure Sentinel and Microsoft Defender capabilities into your threat defense and response strategy.
- Microsoft Threat Protection stops attack sprawl and auto-heals enterprise assets with built-in intelligence and automation.
- Azure Sentinel uncovers the real threats hidden in billions of low fidelity signals.
- Microsoft uses threat intelligence to protect, detect, and respond to threats.
- Tutorial: Set up automated threat responses in Azure Sentinel.
- How a customer significantly reduced alert fatigue using machine learning in Azure Sentinel.
- Use Azure Sentinel Watchlists.
- What’s new: Azure Sentinel User and Entity Behavior Analytics in Preview—Microsoft Tech Community
To learn more about Microsoft Security solutions, visit our website. Bookmark the Security blog to keep up with our expert coverage on security matters. Also, follow us at @MSFTSecurity for the latest news and updates on cybersecurity.
Special thanks to Sarah Young, Chi Nguyen, Ofer Shezaf, and Rafik Gerges for their input.
¹ESG: Security Analytics and Operations: Industry Trends in the Era of Cloud Computing 2019.
The post 6 strategies to reduce cybersecurity alert fatigue in your SOC appeared first on Microsoft Security.