IT Partner/ Blog/ How to detect attacks before they cause damage

    How to detect attacks before they cause damage

    #advanced threat protection, #cloud security, #microsoft 365, #office 365 tenant

    We are excited to start a series of scenarios on how to improve your Microsoft 365 security and productivity, prepared by Microsoft engineers. The first one is about how to detect attacks before they cause damage.

    Traditionally, security investments were focused on protection. With Microsoft Solutions such as EM+S, it is imperative to also have good detection and response. IT organizations should focus on an approach that looks at how to protect, detect, and respond to threats.

    To address the requirements of this scenario, EM+S uses "Advanced Threats Analytics", Cloud App Security and Azure Active Directory Premium . By implementing these technologies, organizations will be able to:

    1. Detect or identify abnormal behavior using innovative behavioral analytics and anomaly detection technologies leveraging machine learning.
    2. Detect known malicious attacks (i.e., Pass the Hash, Pass the Ticket) and known security vulnerabilities.
    3. Focus on what is important, fast, clear, and relevant attack information.
    4. Identify anomalies and policy violations that may be indicative of a security breach.

    When you open the attack timeline in ATA, you see a comprehensive report with suspicious activities showing the entities that were involved in this activity and what the recommendations are.

    Let's see the high-priority issues caused by suspicion of identity theft. According to the report here, the user, Michael Dubinsky's, credentials may have been stolen and used to gain access to resources that Michael doesn't usually access. Let's trace the sequence of events that led to Michael's identity theft.

    As with many attacks, this one begins with a reconnaissance phase where we see the attacker attempting to guess usernames. Ultimately, the attacker(s) succeeded and guessed three different accounts, including Michael's user account.

    In the next phase of the attack, we will clearly see the attacker attempting a brute force attack, including them guessing Michael's password.

    Once Michael's account was compromised, we can see the user behaving abnormally. With the list of alerts prior to this, we have sufficient evidence to conclude that this user's credentials are now compromised.

    As you might see, in this instance, the attack was detected by ATA with the help of data provided by a third party SIEM solution that was configured to forward Windows security events to ATA. The third-party software was already collecting these events, so no additional configuration was required beyond the event forwarding itself.

    All of ATA's detection algorithms are self-learning, allowing it to detect suspicious activities from the first minute it's deployed, without the need to configure or tweak rules, baselines, or thresholds; you simply plug it in and off it goes. You can also configure ATA to send an event to your SIEM.

    In summary, ATA uses machine learning in its deterministic and detection engine to establish an understanding of the normal patterns of behavior for both users and entities, and it's that unique capability that provides timely and accurate alerts across a huge variety of attack vectors.

    Also, we provide Free Office 365 Security Assessment. We can help you evaluate how secure your business really is and find a solution that will help to simplify security and reduce costs while still getting the protection you need.


    Request a call back

    Do you have questions? Want to discuss your project? Please schedule a call back.
    To top