ServiceNow Certified System Administrator (CSA) Practice Exam

What the ServiceNow Certified System Administrator (CSA) exam covers

• User Interface and Navigation119 questions
• Collaboration95 questions
• Database Administration127 questions
• Self-Service and Automation142 questions
• Data and Reporting103 questions
• Administration and Security112 questions
• Data Management and Migration95 questions

Free CSA practice test questions

A sample of 10 questions with answers and explanations. Sign up free to practice all 793.

1. Question 1Administration and Security

   Which table stores individual user records in ServiceNow, such as name, email, and login information?

   • Asys_userCorrect
   • Bsys_user_group
   • Csys_user_has_role
   • Dsys_user_grmember
   ✓ Correct answer: A

   sys_user is the platform's core User table, holding each person's identity directly on the record: first and last name, email, User ID and password, title, department, and manager. Everything else in the sys_user_ family describes a relationship rather than a person, which is the naming pattern worth memorising - sys_user for the individual, sys_user_something for a link or a related object. Spotting that pattern makes the user administration data model easy to navigate without looking anything up.

   Why the other options are wrong

   • Bsys_user_group stores the group records themselves, such as name and manager, not the user's own identity data.
   • Csys_user_has_role is a join table linking a user to a role; it does not hold the user's personal details.
   • Dsys_user_grmember records which groups a user belongs to; it does not store the user's core profile fields.
2. Question 2Administration and Security

   Following the principle of least privilege when assigning access, which approach is most appropriate for a new fulfiller who only works incidents and problems?

   • AGrant the security_admin role so they can adjust ACLs as needed
   • BGrant no role at all, since itil is optional for working tasks
   • CGrant the itil role rather than the broader admin roleCorrect
   • DGrant the admin role so no future access issues arise
   ✓ Correct answer: C

   Least privilege means giving an account exactly the access its job requires and nothing beyond it. A fulfiller who works incidents and problems needs the itil role, which carries the permissions to read, create and update task records, and stops there. The reason it matters is blast radius. A compromised or careless itil account can mishandle tickets; an over-privileged one can rewrite security rules, change configuration or delete data across the whole instance. Access can always be widened later if that person's responsibilities genuinely grow.

   Why the other options are wrong

   • Asecurity_admin is scoped to security configuration, not to the routine work of a task fulfiller; granting it here far exceeds what is needed.
   • BGranting no role at all would leave the fulfiller without the access typically needed to work incidents and problems; some scoped role is still required.
   • DGranting admin far exceeds what a fulfiller needs for incident and problem work and directly contradicts least privilege.
3. Question 3Self-Service and Automation

   What is a knowledge article?

   • AA piece of content in a knowledge base documenting information for self-help readingCorrect
   • BA record producer that is specifically configured to write directly to the Incident table
   • CA variable set that is shared for reuse across several different catalog items
   • DA Catalog Task record that has been assigned to a specific fulfillment group
   ✓ Correct answer: A

   A knowledge article is the unit of documentation content inside a knowledge base. Users and agents reach it by searching or browsing categories, read it, and resolve the question themselves without opening a ticket at all. Its purpose is informational rather than transactional, and that is what separates it from everything in the catalog and fulfillment world. An article never creates work or tracks it; it carries the information that makes the work unnecessary, which is why deflection through knowledge is measured as a benefit in its own right.

   Why the other options are wrong

   • BA record producer configured to write to the Incident table is a catalog object used to create records, not a piece of documentation content.
   • CA variable set shared across catalog items is a form-building object used on catalog items, unrelated to knowledge documentation.
   • DA Catalog Task assigned to a fulfillment group is a fulfillment record, not a piece of self-help reading content.
4. Question 4Self-Service and Automation

   What is a Data Lookup Rule commonly used for?

   • AAutomatically populating default field values from matching conditionsCorrect
   • BDefining the schedule that an SLA definition uses for business hours
   • CPublishing a new version of a classic workflow after editing
   • DSending a notification once a task record has been approved
   ✓ Correct answer: A

   A data lookup rule matches values already present on an incoming record, such as category, subcategory or location, against a lookup table and then writes the matched result into another field, most often assignment group or priority. The effect is fewer fields for a user to fill in and the same answer every time the same inputs appear. The defining characteristic is that a lookup rule only sets field values. It does not time anything, move configuration between instances, or send messages, and because the mapping is maintained as data a process owner can extend it without writing a script.

   Why the other options are wrong

   • BAn SLA definition's business-hours schedule is configured directly on the SLA Definition record, unrelated to data lookup rules.
   • CPublishing a new classic workflow version is a separate administrative action with no connection to data lookup rules.
   • DSending an approval notification is handled by the notification system, not by a data lookup rule, which only sets field values.
5. Question 5Collaboration

   An administrator wants normal changes above a certain risk level to automatically require CAB approval, while low-risk changes skip it. How is this typically achieved?

   • AThrough approval rule conditions that generate CAB approval only when the risk threshold is metCorrect
   • BBy manually emailing the CAB chair individually for every single change regardless of risk
   • CBy changing the change request record's number prefix format
   • DBy deleting the risk field from the change request form entirely
   ✓ Correct answer: A

   The approval rule's condition does the work. Write it to test the change's Risk field and the CAB approval is generated only for changes at or above the threshold, while lower-risk changes carry on without that step. This is what condition-based approvals are for. The rule evaluates the record's actual data at the moment an approval is needed, so routing adapts to each change instead of being hardcoded or left to a person reading every request. The pattern generalises well: any field on the record - risk, category, cost, affected service - can gate whether a particular approval is generated, which is how an approval process stays proportionate to what is actually being changed.

   Why the other options are wrong

   • BManually emailing the CAB chair for every single change regardless of risk defeats the purpose of automated, condition-based approval routing.
   • CChanging the change request's number prefix format has no bearing on which changes require CAB approval.
   • DRemoving the risk field would eliminate the exact data the approval rule's condition needs to evaluate, breaking the intended automation entirely.
6. Question 6Data Management and Migration

   Which of the following changes would normally be captured in an update set moved between instances?

   • AA change to a UI policy's conditionCorrect
   • BA new record in the Incident table
   • CA user adding a value to a Company record
   • DA new attachment on a Change request
   ✓ Correct answer: A

   Update sets capture configuration, and a UI policy is a configuration record, so editing one - including its condition - is recorded in the current update set and travels to the target instance on commit. The line the platform draws is between configuration and data. Business rules, UI policies, client scripts, form layouts, ACLs and dictionary changes are configuration and are captured automatically. The records users create while doing their jobs, the values they type into reference data, and the files they attach are all data, and data moves between instances by import or clone rather than by update set.

   Why the other options are wrong

   • BA new record in the Incident table is transactional data, which update sets are designed to exclude from capture.
   • CA user adding a value to a Company record is a data change on a reference record, not application configuration.
   • DA new attachment on a Change request is data tied to a transactional record, not a configuration change.
7. Question 7Database Administration

   Which field type is designed to hold multiple record references at once, such as multiple watchers on a record?

   • AChoice
   • BString
   • CGlide ListCorrect
   • DReference
   ✓ Correct answer: C

   Glide List stores a set of sys_ids rather than a single one, so one field can point at several records at once. That is what makes it the type behind fields like Watch list, where any number of people may need to be tracked in the same place. The contrast with a standard Reference field is the whole lesson: a Reference field is capped at exactly one related record, so expressing a one-to-many relationship from inside a single field calls for Glide List. Field types that hold fixed option values or plain text do not point at records at all.

   Why the other options are wrong

   • AChoice stores a value from a fixed predefined list; it does not point to records on another table at all.
   • BString stores plain text; it has no capability to reference or point to other records.
   • DA standard Reference field is limited to exactly one related record, not a set of multiple references.
8. Question 8Data and Reporting

   Which report type displays individual records in rows with selected columns, essentially mirroring what a filtered list view of a table shows, without grouping or aggregating the data?

   • AList reportCorrect
   • BColumn chart
   • CHeatmap
   • DFunnel chart
   ✓ Correct answer: A

   A List report deliberately skips aggregation, presenting matching records exactly as a filtered list view would, with the chosen fields as columns. This is the one report type built for showing actual records rather than a summarized count, useful whenever the audience needs to see the records themselves. Column charts and heatmaps both require grouping to produce their visuals, and a funnel chart tracks staged counts rather than listing records. List report is essentially a report-native version of a filtered list view.

   Why the other options are wrong

   • BA column chart groups records and plots aggregated counts as vertical bars; it does not list individual records.
   • CA heatmap shows a color-coded matrix of two grouped dimensions, not individual records.
   • DA funnel chart shows counts narrowing across sequential stages, not a row-by-row record list.
9. Question 9Data and Reporting

   A team wants a dashboard tab that shows both a Performance Analytics scorecard trend and a real-time incident report side by side. Is this combination possible on one dashboard?

   • AYes, a dashboard tab can host multiple widget types togetherCorrect
   • BNo, a dashboard can only ever contain Performance Analytics content
   • CNo, a dashboard can only contain standard reports, never PA content
   • DNo, each dashboard tab can only ever hold exactly one widget
   ✓ Correct answer: A

   A dashboard tab can host multiple widget types at once, so a Performance Analytics scorecard trend and a real-time incident report can sit side by side on the same tab. The widget catalog is deliberately not restricted to one content family. It spans PA content, standard reports, and other widget types precisely so different perspectives on the same subject can be read together rather than split across separate tabs. That is exactly what the scenario asks for, and it needs no workaround: add each widget to the tab and arrange them on the canvas.

   Why the other options are wrong

   • BDashboards are not restricted to only Performance Analytics content; standard reports and other widget types can also be added freely.
   • CDashboards are not restricted to only standard reports either; Performance Analytics scorecards and indicators can be added as widgets too.
   • DA dashboard tab can hold multiple widgets at once; it is not limited to exactly one widget per tab.
10. Question 10User Interface and Navigation

    A UI Policy makes a field mandatory only when Category equals Hardware. What happens when that condition is false?

    • AThe field stays mandatory regardless of the condition
    • BThe mandatory requirement is not enforcedCorrect
    • CThe field is permanently deleted
    • DThe record cannot be saved at all
    ✓ Correct answer: B

    A UI Policy's condition gates everything it does. While Category equals Hardware the mandatory requirement applies; the instant Category is anything else the condition evaluates false and the field reverts to whatever its normal state would otherwise be, so the record saves without a value in it. That on-or-off behavior is the whole reason to build a UI Policy around a condition instead of ticking Mandatory on the dictionary entry, which would demand a value on every record on the table regardless of category. Note also that a UI Policy only adjusts behavior such as visibility, read-only and mandatory status; it never alters the table's structure.

    Why the other options are wrong

    • AA conditional UI Policy's requirement only applies while its condition is true; it does not stay mandatory regardless of the condition.
    • CUI Policies adjust field behavior like visibility or mandatory status; they never permanently delete a field from the table.
    • DOnce the condition is false, the mandatory requirement lifts and the record can be saved normally; saving is not blocked.

Who this ServiceNow Certified System Administrator (CSA) practice exam is for

This practice set is for anyone preparing for the ServiceNow Certified System Administrator (CSA) exam at the intermediate level - from first-time candidates building a foundation to experienced ServiceNow practitioners doing a final review before test day. If you learn best by working through realistic questions and reading why each answer is right or wrong, it is built for you.

How to use this ServiceNow Certified System Administrator (CSA) practice exam

1. Start with the free sample questions above to gauge your current baseline.
2. Read the full explanation on every question, including why each wrong option is wrong.
3. Track your weak domains and focus your study where you are losing the most marks.
4. Once you are scoring consistently well, take a timed, full-length mock exam.
5. Use your readiness score to decide when you are ready to book the real ServiceNow Certified System Administrator (CSA) exam.

Related ServiceNow resources

• ServiceNow Certified System Administrator (CSA) study guideKey concepts
• ServiceNow practice examsAll ServiceNow
• Certification pathWhere this fits
• Certification exam guides & tipsBlog
• Plans & pricingFree & paid
• How these questions are written and reviewedMethodology
• Report a problem with a questionCorrections
• ServiceNow Certified Application Developer (CAD) practice examRelated

ServiceNow Certified System Administrator (CSA) practice exam FAQ