Explore / Microsoft Security Operations Analyst (SC-200) Exam Prep / Practice questions
Microsoft Security Operations Analyst (SC-200) practice questions
10 questions from the bank behind this course's timed mock. Free to read and to answer, no account needed, and each has its answer and an explanation. They belong to Microsoft Security Operations Analyst (SC-200) Exam Prep.
Answer them one at a time →Read lesson 1 free
Question 1
You are a security operations analyst who passed SC-200 eleven months ago. Your manager wants you to keep the certification current with the least cost and disruption. What should you do?
- Score 700 or greater on the Free Practice Assessment
- Complete the exam sandbox walkthrough of the current question types
- Book and pass the full SC-200 exam again at a test centre or online
- Pass the free online renewal assessment on Microsoft Learn before the certification expires
Show the answer
Answer: D. Pass the free online renewal assessment on Microsoft Learn before the certification expires
Associate certifications expire every year, and you renew them by passing a free online assessment on Microsoft Learn. Retaking the full exam would also keep you certified, but it costs money and takes far more effort. The Practice Assessment and the sandbox are only for preparation and don't renew anything.
Question 2
You are a security operations analyst. You created a URL/domain indicator with the action Block execution for a phishing domain and scoped it to all devices. Custom network indicators is turned on in Advanced features. Users who browse to the domain in Microsoft Edge are blocked, but users on Google Chrome can still reach it. What should you do so that Chrome is blocked too?
- Create a web content filtering policy that blocks the Phishing category
- Change the indicator action from Block execution to Block and remediate
- Turn on the Allow or block file advanced feature
- Turn on network protection in block mode on the devices
Show the answer
Answer: D. Turn on network protection in block mode on the devices
Edge enforces network indicators through SmartScreen, but Chrome, Firefox and processes that aren't browsers rely on network protection in block mode. Allow or block file is the prerequisite for file hash indicators, not network indicators, so turning it on changes nothing here.
Question 3
You are a security operations analyst. Your SOC lead asks for a list of every action automatic attack disruption has taken across the whole organisation in the last 30 days, including the action type, the affected device or account, and any remote IP. You must use advanced hunting. Which table should you query?
- DeviceEvents, filtered on ActionType containing "ContainedDevice"
- DeviceNetworkEvents, filtered on blocked connections
- DisruptionAndResponseEvents
- AlertInfo, filtered on alerts from automated investigation and response
Show the answer
Answer: C. DisruptionAndResponseEvents
DisruptionAndResponseEvents holds one row for each disruption action across the organisation, with columns such as ActionType, AttackDisruptionCategory, DeviceName, AccountUpn and RemoteIP. DeviceEvents filtered on "ContainedDevice" only shows traffic that was blocked because a device was already contained, so it isn't a list of the disruption actions themselves.
Question 4
You are a security operations analyst at a retailer with a tight Microsoft Sentinel ingestion budget. Your detections only need breach indicators from the company's Windows servers, such as failed logons (4625) and process creation (4688). A full user audit trail is not required. Which event set should you choose in the Windows Security Events via AMA data collection rule?
- All events
- Minimal
- Common
- Custom, with an XPath query that collects every Security event except 4624
Show the answer
Answer: B. Minimal
Minimal collects breach indicators such as 4624, 4625 and 4688 and leaves out the full audit trail, so it is the cheapest of the predefined sets that still supports detection. Common adds the full audit trail, including logoff 4634, group changes and Kerberos events, so it costs more volume than this requirement needs.
Question 5
You are a SOC analyst at Contoso. You have a hunting query that counts AntivirusDetection events in DeviceEvents for each DeviceId. It uses summarize count() by DeviceId and keeps devices with more than five hits. You want to save it as a custom detection rule with the fewest changes. What should you do?
- Change the summarize to bring back Timestamp and ReportId with arg_max(Timestamp, ReportId), grouped by DeviceId
- Add | where Timestamp > ago(1d) before the summarize so the rule knows its time window
- Add | project Timestamp, ReportId, DeviceId after the summarize line
- Remove the summarize and set the rule to Continuous (NRT) so every event becomes its own alert
Show the answer
Answer: A. Change the summarize to bring back Timestamp and ReportId with arg_max(Timestamp, ReportId), grouped by DeviceId
summarize drops Timestamp and ReportId, and the rule needs both. arg_max keeps them from the most recent event for each device. A project after the summarize can't help, because those columns no longer exist at that point.
Question 6
You are a SOC analyst. A Defender for Cloud Apps alert shows that a user's account was probably compromised. Your organisation already has Microsoft Entra ID risk-based Conditional Access policies that block high-risk users or make them reset their password. You want those existing policies to deal with the user, and you don't want to lock the account out directly from Defender for Cloud Apps. Which governance action should you apply?
- Require user to sign in again
- Suspend user
- Revoke app
- Confirm user compromised
Show the answer
Answer: D. Confirm user compromised
Confirm user compromised sets the user's risk level to high, so your Microsoft Entra ID risk policies take over. Suspend user blocks access and sign-in directly, which skips the risk-policy route the scenario asks for.
Question 7
You are a SOC analyst using Microsoft Defender XDR. A user reported a phishing email, and the resulting incident is open with Copilot guided responses in view. Your manager asks you to find every other mailbox that got messages like this one. You must not write KQL by hand. What should you use?
- The View similar emails option in the guided responses
- File analysis on the email attachment
- The View similar incidents option in the guided responses
- Script analysis from the attack story, started with Analyze
Show the answer
Answer: A. The View similar emails option in the guided responses
For phishing incidents, View similar emails opens advanced hunting with a KQL query that is already written, so you can find related messages without writing any. View similar incidents is a real phishing option, but it lists related incidents matched by machine learning, not individual emails across mailboxes.
Question 8
You are a SOC analyst at Contoso. Your custom role includes Advanced live response. Live response sessions on Windows 11 workstations connect without problems. When you select Initiate live response session on a Windows Server 2022 domain controller that is onboarded and healthy, you can't start the session. You need to fix this with as little change to your security posture as possible. What should an administrator do?
- Turn on the Live Response advanced feature
- Add the File collection permission to your custom role
- Turn on the Live Response for Servers advanced feature
- Turn on the Live Response unsigned script execution advanced feature
Show the answer
Answer: C. Turn on the Live Response for Servers advanced feature
Sessions on Windows Server need their own advanced feature, Live Response for Servers. The base Live Response feature is already on, because workstation sessions work. Unsigned script execution affects only running unsigned PowerShell scripts, not connecting, and it would increase your exposure to threats.
Question 9
You are a security operations analyst. Defender for Office 365 delivered a phishing message to 40 mailboxes, and zero-hour auto purge (ZAP) later moved it out of some of them. Your manager wants to know exactly which mailboxes ZAP acted on after delivery. Which advanced hunting table should you query?
- EmailPostDeliveryEvents
- EmailUrlInfo
- EmailEvents
- UrlClickEvents
Show the answer
Answer: A. EmailPostDeliveryEvents
EmailPostDeliveryEvents records actions taken after a message is delivered, such as ZAP. EmailEvents is the closest wrong choice: it shows the original delivery verdict (Delivered or Blocked), not what happened to the message later.
Question 10
An analyst at Contoso is checking whether user Jo Okafor can reach key vault kv-payments-prod. She writes a KQL query that joins users to groups, groups to role assignments, and role assignments to resources. It returns no results. Later, a hunt shows that Jo could reach the vault through an OAuth app and a cached credential. Why would Sentinel graph have been the better tool for this question?
- The graph gets its data only from Defender XDR device telemetry, which records OAuth consent events that Entra tables leave out.
- The graph queries the analytics tier's hot, indexed tables, which keep more relationship data than the Sentinel data lake.
- The graph replaces KQL with GQL, and GQL can write the missing relationship edges back into the source tables.
- The graph stores the estate as nodes and edges, so it can follow every relationship type over several hops without the analyst having to predict the number of joins or the route.
Show the answer
Answer: D. The graph stores the estate as nodes and edges, so it can follow every relationship type over several hops without the analyst having to predict the number of joins or the route.
With flat tables you need one join per hop, and you have to guess the routes in advance. Miss one route, such as an OAuth app, and you get a false 'no'. Graph traversal follows whatever edges exist. GQL can't help here because it is read-only, and the graph is built from lake data from Entra, Microsoft 365 and Azure Resource Graph, not the analytics tier or Defender XDR telemetry alone.
The full timed mock
50 questions in 100 minutes, with a pass mark of 70%. Each paper is drawn from a bank of questions written for the mock alone, and these 10 are left out of it. Full access opens it, with every level of the course.
LevelUpWith is independent of the organisation that sets this exam. These questions were written for this course; they are not taken from the exam.