Free Collibra Data Steward Certification Sample Questions

Correct answer: B — To approve data sets for use per governance criteria

• A. Incorrect. Cataloging and discovering data assets is a function of data cataloging, not certification. Certification focuses on validating that data sets meet quality and governance standards, rather than simply listing assets.
• B. Correct. Data certification in Collibra is primarily used to formally approve data sets for use based on predefined governance criteria. It signals to consumers that the asset is trusted, validated, and suitable for business use.
• C. Incorrect. Archiving datasets for long-term storage is a records management or retention activity, unrelated to certification. Certification is about validating data for active use, not storage lifecycle management.
• D. Incorrect. Masking personally identifiable information is a data protection or privacy control, not the purpose of certification. While certification may consider compliance, it does not itself perform masking.

Correct answer: C — Investigate the claim and launch a certification review if needed

• A. Incorrect. Certification does not mean a dataset is immune to scrutiny; the data steward must review claims of outdated information to ensure ongoing data quality and certification standards.
• B. Incorrect. Immediately revoking certification without investigation is too drastic and disrupts business. The steward should first validate the claim and assess impact before taking action.
• C. Correct. The data steward should investigate the challenge to verify its validity. If the dataset is indeed outdated, a certification review should be launched to address the issue and maintain certification standards.
• D. Incorrect. Certification is the responsibility of the data steward or designated authority, not the business user. The user's challenge should be reviewed, but they should not be asked to certify the dataset.

Correct answer: A — The articulation score measures metadata completeness, while the data quality score measures data validity.

• A. Correct. The articulation score measures the completeness and richness of metadata for an asset (e.g., descriptions, relationships, classifications), while the data quality score assesses the validity, accuracy, and reliability of the actual data.
• B. Incorrect. Both scores can be calculated automatically or manually depending on configuration; the key difference is the object of measurement: metadata completeness versus data quality.
• C. Incorrect. The articulation score applies to various asset types, not just business terms; similarly, data quality scores are not limited to data assets alone.
• D. Incorrect. The articulation score is not a simple true/false indicator; it is typically a percentage or numerical value representing metadata completeness, and data quality scores can also be percentages.

Correct answer: B — A mandatory metadata field was cleared

• A. Incorrect. An automated approval process typically changes workflow state or governance status, but it does not cause a sudden drop in articulation score. Articulation score is driven by metadata completeness and required attributes, not approval actions.
• B. Correct. Articulation scores are directly tied to metadata completeness. If a mandatory metadata field was cleared, the asset immediately loses points for required completeness, causing a sharp drop. A sudden overnight decrease is most consistent with a required field becoming empty or invalid.
• C. Incorrect. A platform version update may introduce new features or changes, but it does not directly alter the articulation score of an individual asset unless it retroactively affects metadata requirements. A score change of this size is more commonly explained by a metadata change on the asset itself.
• D. Incorrect. The number of views an asset receives does not impact its articulation score. Articulation scoring is based on metadata and governance completeness, not usage metrics like views.

Correct answer: B — Technical lineage is harvested from the connected data source

• A. Incorrect. While registering and authenticating a data source is necessary for connecting Collibra to the system, it is not the direct prerequisite for viewing technical lineage. Lineage will not appear unless it has been harvested from that source.
• B. Correct. Technical lineage must be harvested from a connected data source in order to be visible in the technical lineage viewer. Without harvested lineage metadata, Collibra cannot display the data flow between assets.
• C. Incorrect. Assigning business terms to assets is a business governance activity that enriches metadata but is not required for technical lineage analysis. The technical lineage viewer focuses on technical relationships and data flow, not on whether glossary terms are mapped.
• D. Incorrect. A System Administrator role is not required to view or analyze technical lineage. Users with appropriate permissions, such as Data Steward or Lineage Viewer roles, can access the lineage viewer provided lineage data has been harvested.

Correct answer: B — Mostly unique values suggest an ID field.

• A. Incorrect. A large number of duplicate entries indicates low uniqueness, which is not typical for an ID field. IDs are expected to be unique or nearly unique to identify rows.
• B. Correct. Profiling results showing mostly unique values strongly suggest the column is an identifier field, as IDs are designed to uniquely identify each row.
• C. Incorrect. Many missing values relate to completeness, not uniqueness. While missing data is a profiling result, it does not imply the column is an ID field.
• D. Incorrect. The data type being string is not indicative of an ID field. Many non-ID fields are also stored as strings, so this alone is insufficient to conclude it is an ID.

Correct answer: A — True

• A. The statement is true because when multiple filters are selected in the basic search panel, Collibra uses the OR operator, meaning assets matching any filter are returned.
• B. The statement is false because Collibra actually applies the OR operator between multiple filters, not the AND operator; thus the original statement is correct.

Correct answer: B — The issues remain open with a warning about stale metadata.

• A. Incorrect. Re-synchronizing an asset does not automatically close all related open issues. Issues are preserved so users can still track and resolve the underlying concern; they remain active unless manually resolved.
• B. Correct. By default, open issues remain open after re-synchronization, but Collibra may flag them with a warning indicating that the underlying metadata has changed and the issue may be based on stale metadata, prompting review.
• C. Incorrect. Collibra does not automatically place open issues into a special review state for a system administrator when an asset is re-synchronized. Issues remain part of the normal workflow unless a user acts on them.
• D. Incorrect. Re-synchronization does not reassign issues to the steward who performed the synchronization. Issue ownership and workflow are managed separately from the synchronization process; reassignment is a manual action.

Correct answer: B — Re-synchronize the table asset, then review the issues.

• A. Rejecting the issues is premature because the schema change must first be reflected in Collibra. Updating the asset metadata allows you to accurately assess which issues are still relevant. Discarding issues without synchronization may lead to loss of important context.
• B. Re-synchronizing the table asset ensures Collibra reflects the latest schema from the data source. After synchronization, you can review the open issues and determine whether they should be closed, reassigned, or otherwise handled based on the updated asset state.
• C. Closing the issues as 'Resolved' without first updating the schema in Collibra is premature. The issues should be reviewed after the asset is re-synchronized to verify their validity based on the current schema.
• D. Creating new issues on the remaining columns does not address the existing open issues tied to the dropped columns. The immediate need is to align Collibra with the source system change by re-synchronizing the asset, then assessing the current issue set.

Correct answer: B — Write a comprehensive business description explaining the report’s purpose and key metrics.

• A. Incorrect. While data quality rules are important for improving trust in data, they do not directly add business context. The first step in making an ingested asset meaningful is to describe and contextualize it, not to monitor its quality.
• B. Correct. Writing a clear business description is the most direct way to add business context to a newly ingested report. It explains the report’s purpose, what it measures, and how business users should interpret it, thereby making the asset meaningful.
• C. Incorrect. Assigning the asset to a domain and designating a steward helps with governance and ownership, but it does not by itself explain the report's business meaning. This is more of an administrative governance step than a context-enrichment step.
• D. Incorrect. Access controls are important for security and compliance, but they do not add business context to the report. They determine who can see the asset, not how the asset is understood by business users.

Correct answer: A — Navigate to the business term, open the details panel, click "Generate AI description", review and accept

• A. Correct. This is the standard process in Collibra for generating an AI description. Navigate to the business term, open the details panel, click 'Generate AI description', then review and accept the suggested text before saving.
• B. Incorrect. This describes the manual edit path, not the AI description generator. While it updates the description, it does not use Collibra’s AI feature.
• C. Incorrect. Creating a custom workflow is not the typical method for generating an AI description for a single business term. The AI description generator is a built-in feature accessed directly from the asset’s details panel, not via the workflow editor.
• D. Incorrect. Bulk operations are not the standard mechanism for generating AI descriptions for a single term. The expected action is to open the asset and use the dedicated AI description generation option in its details panel.

Correct answer: C — Review the term definition with stakeholders; update if wrong, else relink to a suitable term

• A. Incorrect. Merely changing the attribute name does not resolve the semantic misalignment between the attribute and the business term. The attribute name reflects its technical meaning, while the business term should capture business meaning. Renaming does not ensure accurate governance.
• B. Incorrect. Unlinking the term without further investigation leaves the attribute without business context, reducing its understandability and governance value. The data steward should first verify whether the term definition is correct or the linkage is inappropriate.
• C. Correct. The data steward should review the business term definition with stakeholders to confirm accuracy. If the definition is incorrect, update it; if the definition is correct but the link is mismatched, unlink and link to a more appropriate term. This ensures metadata remains semantically correct and useful.
• D. Incorrect. Creating a new business term without evaluating the existing one can lead to redundancy and confusion. While a new term may be needed if the concept is truly missing, linking both terms to the same attribute is not best practice and does not resolve the misalignment.

Correct answer: B — To indicate the asset's current lifecycle phase

• A. Incorrect. The 'Status' field does not control workflow assignment; workflows are configured separately and manage process steps, not the asset's status value.
• B. Correct. The 'Status' field indicates the asset's current lifecycle phase, such as 'Draft', 'Approved', or 'Deprecated'. It helps users understand where the asset is in its lifecycle.
• C. Incorrect. User access and permissions are managed through roles and domain restrictions, not by the 'Status' field. The status does not determine who can view the asset.
• D. Incorrect. Search visibility and priority are controlled by other metadata or configurations, not the 'Status' field. Asset status is about lifecycle state, not search behavior.

Correct answers: A, B, C — The 'Name' column was not present in the Excel file.; The 'Community' column contained invalid community names.; The 'Asset Type' column had types not defined in Collibra.

• A. Correct. The 'Name' column is mandatory for creating assets in Collibra. If it is missing, the import will fail for those records because a name is required for each asset.
• B. Correct. The 'Community' column must reference valid community names in Collibra. Invalid or non-existent community names cause the importer to fail on those rows.
• C. Correct. The 'Asset Type' must match an existing asset type configured in Collibra. Undefined or unsupported types will result in import errors for those records.
• D. Incorrect. Collibra supports both .xls and .xlsx file extensions for Excel imports. The file extension alone is not a reason for record-level failures; this would be a global issue.
• E. Incorrect. If the steward lacks the 'Asset: Create' permission for the target domain, the entire import would fail, not just some records. Thus, it cannot explain partial failure.
• F. Incorrect. Scheduled maintenance windows are not a typical cause of record-level import failures. Import failures are normally due to data validation, permissions, or configuration issues.

Correct answers: A, B — Adding a new custom attribute value; Creating a relation to another asset

• A. Correct. Adding or changing a custom attribute value modifies the asset's metadata, which is an enrichment action. Collibra records such changes in the Activity Log to maintain an audit trail, tracking who changed what and when.
• B. Correct. Creating a relation to another asset enriches the asset by establishing a connection, altering its metadata and context. Collibra logs this action in the Activity Log for traceability.
• C. Incorrect. Accessing the asset page for viewing is a read-only action and does not modify or enrich the asset. The Activity Log focuses on changes that alter the asset or its metadata.
• D. Incorrect. Exporting asset data as a CSV is a data retrieval action, not an enrichment change to the asset itself. Therefore, it is not recorded in the Activity Log.
• E. Incorrect. Retrieving an attached resource file is a viewing or download action, not a modification. The Activity Log records modifications, not access to attachments.
• F. Incorrect. Sending the asset link via email is a sharing action that does not alter the asset's metadata or content. Hence, it is not captured as an enrichment action in the Activity Log.

Correct answers: A, B, E — Browsing the Data Catalog and filtering by assigned data class; Using the Classification Dashboard to review overall coverage; Reviewing classification status of columns in a specific data set

• A. Browsing the Data Catalog and filtering by assigned data class allows data stewards to view and inspect data assets based on their classification, enabling verification of class assignments against expectations. This is a standard and direct method for reviewing ingested data and data classes.
• B. The Classification Dashboard provides an overview of classification coverage and metrics, which is a valid tool for reviewing the status and overall coverage of ingested data and data classes, helping stewards monitor governance progress.
• C. Running a data quality rule is used to assess data quality conditions, not to verify classification accuracy. Classification review relies on catalog metadata and classification views, not data quality rules.
• D. Exporting data asset records to a CSV file may support offline analysis but is not a primary in-application method for reviewing ingested data and data classes within Collibra. It adds manual effort and does not directly support active stewardship.
• E. Reviewing classification status of columns in a specific data set allows data stewards to inspect granular classification details after ingestion, validating whether sensitive or governed data elements have been correctly identified.
• F. Assigning a new data class directly from the workflow inbox is an action to classify data, not a method for reviewing existing classifications. The question asks about review, so this option is outside the scope of inspection and validation.

Correct answer: C — Reject both suggestions and manually assign a custom classification or data attribute.

• A. Incorrect. While you can accept or reject individual suggestions, this does not address the conflict resolution process. The steward must explicitly resolve the conflict by choosing one or rejecting both. Without context, simply accepting one over the other is not necessarily valid.
• B. Incorrect. A column cannot have two mutually exclusive classifications (e.g., 'Email Address' and 'Phone Number') simultaneously. Accepting both would create ambiguity and not resolve the conflict.
• C. Correct. When Edge suggestions are conflicting or not trustworthy, the data steward can reject the suggestions and manually apply the appropriate classification or data attribute. This ensures the column has a single, accurate classification based on steward judgment.
• D. Incorrect. Leaving a suggestion in a pending state indefinitely does not resolve the conflict and may lead to incomplete or inconsistent data classification. The steward is expected to accept or reject each suggestion.
• E. Incorrect. Bulk accepting all suggestions does not resolve conflicts on individual columns. Conflicting classifications would still exist, and the steward must address these conflicts individually to ensure accuracy.
• F. Incorrect. Requesting a re-run of Edge profiling does not guarantee the conflict will be resolved, as similar suggestions may reappear. The steward should first review and manually resolve the conflict rather than relying on reprofiling.

Correct answer: B — The filters are combined with an OR condition instead of an AND condition.

• A. Incorrect. The 'Assigned to' filter in Collibra is not case-sensitive by default, so a case mismatch would not cause the view to show assets assigned to other users. Instead, the issue points to how the filters are being combined.
• B. Correct. If the filters are combined with an OR condition, the view displays assets that meet either criterion: assigned to the user OR due within 30 days. This explains why assets assigned to other users appear if their certification due date is within the next 30 days.
• C. Incorrect. Filters in Collibra do not override each other; they are evaluated together based on the logical operator (AND or OR) set between them. The most likely cause is the operator, not override.
• D. Incorrect. A lack of permissions would typically prevent the user from applying or seeing the filter at all, not cause it to include extra assets. The described behavior is about filter logic, not access control.

Correct answer: B — Add an OR-based filter for Asset Type equals "Data Model" alongside the existing filter.

• A. Deleting and recreating the view is unnecessary and inefficient, as Collibra allows modifying existing filters directly without discarding any other configurations.
• B. Adding an OR-based filter for Asset Type equals "Data Model" alongside the existing filter for "Data Dictionary" will include both asset types in the view. This is the standard way to combine multiple conditions in Collibra, broadening the result set without replacing the existing view.
• C. Collibra does not support multi-value input in a single filter value for asset types; comma-separated entries are not interpreted as multiple discrete values. Each asset type must be specified as a separate filter condition combined with OR logic.
• D. Manually combining views is not a supported or efficient method in Collibra. Views cannot be merged; the existing view should be edited directly to include filter criteria for both asset types.

Correct answer: B — To include assets from a selected community and all its sub-communities.

• A. Incorrect. Custom logical operators (AND/OR) are used in filter conditions, not hierarchy paths. Hierarchy paths traverse the hierarchy to include assets based on their location, not to apply logical operations.
• B. Correct. Hierarchy paths allow you to include assets from a selected community and all its sub-communities by traversing the hierarchy structure. This is useful when you want the view to reflect the full hierarchical scope under a parent node.
• C. Incorrect. Restricting a view to a specific asset type is done with type-based filters, not hierarchy paths. Hierarchy paths focus on structural relationships and inclusion based on ancestry, not asset types.
• D. Incorrect. Sorting assets by their position in the hierarchy is a display or ordering concern, not the purpose of hierarchy paths. Hierarchy paths determine inclusion based on ancestry, not sorting order.

Correct answer: C — Add a hierarchy path for 'Risk Management', then add a filter group containing both Asset Type = 'Report' and Tag = 'Compliance' conditions with the group operator set to AND.

• A. Incorrect. While adding a hierarchy path and separate conditions with AND logic is valid in concept, the typical Collibra view setup places the hierarchy path separately from the filter conditions. The desired configuration uses a single filter group for Asset Type and Tag, not separate top-level conditions.
• B. Incorrect. Using OR logic between filter groups would return assets matching any of the conditions (e.g., under 'Risk Management' OR Asset Type = 'Report' OR Tag = 'Compliance'), which does not satisfy the AND requirement and broadens the result set instead of narrowing it.
• C. Correct. This approach uses the hierarchy path to scope the view to the Risk Management domain, then applies a single filter group containing both Asset Type = 'Report' and Tag = 'Compliance' conditions combined with AND. This ensures that only assets under Risk Management with Asset Type Report and Tag Compliance are returned.
• D. Incorrect. Hierarchy paths in Collibra cannot include tags as nested elements. Tags are separate metadata and must be filtered using dedicated tag conditions, not hierarchy paths.
• E. Incorrect. While the three conditions are all combined with AND, the Domain condition is not equivalent to setting the hierarchy path to Risk Management in the way Collibra views are designed. The hierarchy path should be used for domain scoping, with asset type and tag applied as filters.
• F. Incorrect. Splitting the requirements into two separate views does not create a single view that returns only the desired assets. The goal is a single view where the hierarchy path and filters work together, not two independent views.

Correct answer: A — Reopen the issue and add a comment describing the remaining problem.

• A. Reopening the issue maintains continuity and preserves the history and context of the original problem. Adding a comment ensures transparency about the remaining work and clarifies what still needs to be addressed.
• B. Creating a new issue duplicates the original record, fragments the discussion, and loses the historical context. The existing issue should be reopened instead.
• C. Leaving the issue as resolved misrepresents the status, as the fix is incomplete. The status should reflect the actual state of the issue to avoid oversight.
• D. Deleting the issue removes all historical context and tracking, disrupting audit trails and collaboration. The proper workflow is to reopen the existing issue and document the remaining problem.

Correct answer: C — The workflow stops, and the change is not implemented.

• A. Incorrect. Rejecting an approval task does not apply the change; it prevents the change from being implemented. A comment may be added, but the change itself is not applied.
• B. Incorrect. Rejecting a task does not permanently delete the change request. The request typically remains in the system for review or resubmission.
• C. Correct. When a data steward rejects an approval task, the workflow stops, and the proposed change is not implemented. This is the standard behavior in approval workflows.
• D. Incorrect. Automatic escalation to a senior approver is not the default behavior when a task is rejected. Escalation may occur in some workflows, but it is not guaranteed or standard.

Correct answer: B — To measure community sentiment quantitatively

• A. Incorrect. The rating feature in Collibra is not designed to record asset views; asset views are tracked through separate analytics or logging mechanisms. Ratings capture user feedback, not view counts.
• B. Correct. The rating feature allows users to express their sentiment about an asset quantitatively, and these ratings can be aggregated to measure overall community sentiment. This provides a quick signal about perceived quality or usefulness.
• C. Incorrect. Ratings do not automatically start certification reviews. Certification is a separate governance process with its own workflows and approval steps.
• D. Incorrect. Ratings are not used to allocate stewards. Steward assignment is based on roles, responsibilities, or domain expertise, not on user ratings.

Correct answer: B — On the asset page in the Comments section and Activity stream

• A. Incorrect. The asset's overview summary dashboard provides a high-level view of the asset but does not include the Comments section or Activity stream where user interactions occur. Collaboration actions are performed on the asset page, not on a summary dashboard.
• B. Correct. In Collibra, the asset page includes a dedicated Comments section and an Activity stream where users can view, post, and interact with comments and questions about the asset. These areas are specifically designed for collaboration and discussion.
• C. Incorrect. The administrative console's collaboration audit log is used for tracking and auditing administrative activity, not for user collaboration on an asset. It does not allow users to post comments or ask questions.
• D. Incorrect. An asset relationship export report provides structural or relational data about assets but does not facilitate interaction. It cannot be used to rate, comment, or ask questions about an asset.

Correct answer: B — Select multiple tasks in the task list and apply the bulk Complete action.

• A. Incorrect. While this method works, it is inefficient for multiple tasks as it requires manual action for each individual task.
• B. Correct. Collibra allows bulk actions, including the ability to select multiple tasks and complete them simultaneously, which is the most efficient method.
• C. Incorrect. Delegating tasks does not complete them; it only transfers responsibility, which does not address the requirement to complete the tasks.
• D. Incorrect. Modifying the workflow is a long-term solution and does not directly address the immediate need to complete the existing tasks.

Correct answer: A — It remains in the same status as before reassignment.

• A. Correct. When a task is reassigned in Collibra, the task status remains unchanged; only the assignee is updated. The task does not transition to a new status or reset.
• B. Incorrect. Reassigning a task does not reset it to the initial status of the workflow. The task retains its current status and progress.
• C. Incorrect. Collibra does not have a dedicated 'Reassigned' status. Reassignment updates the assignee without changing the task's status.
• D. Incorrect. Reassignment does not complete the task or create a new one. The existing task is simply reassigned to a new assignee.

Correct answer: B — It lists error details including row number and failure reason.

• A. Incorrect. The import error log does not record successfully imported rows. Successful imports are tracked by the import job result or summary, not by the error log. It focuses on failures and issues encountered during the import process.
• B. Correct. The import error log is designed to capture error details, including the row number where the error occurred and the specific reason for the failure. This helps users identify and fix data issues.
• C. Incorrect. Performance metrics like import speed and memory usage are not the primary purpose of the error log. Those would belong to system monitoring or job execution statistics, not import error reporting.
• D. Incorrect. The import error log does not primarily show configuration settings for the import job. Its main purpose is to help diagnose failed records and understand why the import did not succeed. Configuration settings are typically found in the import job setup or configuration files.

Correct answers: A, C — Carefully download the error log to review failed rows and errors.; Fix the data issues in the source file based on the errors.

• A. Correct. Downloading the error log is the essential first step to identify exactly which rows failed and why, enabling targeted fixes for the validation errors.
• B. Incorrect. Deleting the import job discards progress and does not address the underlying data issues; the error log provides the information needed to fix them efficiently.
• C. Correct. Once the errors are understood, fixing the data issues in the source file ensures that the corrected data can be re-imported successfully.
• D. Incorrect. Re-importing the original file without corrections will simply repeat the same validation errors because the root cause remains unaddressed.
• E. Incorrect. Manually adding failed assets is time-consuming, error-prone, and bypasses the import process, which is designed to ensure consistency and traceability.
• F. Incorrect. While the 'Re-import failed rows only' feature can be used after data correction, it is not one of the two primary actions; the key initial steps are reviewing the error log and fixing the data.

Correct answer: B — .csv

• A. Incorrect. While .xlsx is a spreadsheet format, Collibra's spreadsheet import for bulk updates expects a CSV template, not Excel workbooks.
• B. Correct. .csv is the standard file format used for spreadsheet imports to update attributes and relations in bulk in Collibra. It is the expected tabular format for mapping rows and columns during import.
• C. Incorrect. .pdf is a document format and is not suitable for structured bulk data imports. It cannot be used to update attributes and relations through spreadsheet import.
• D. Incorrect. .txt is a plain text format and does not provide the structured tabular layout needed for spreadsheet import. Collibra does not use .txt for bulk attribute and relation updates.
• E. Incorrect. Although .xls is an older Excel format, it is not the accepted format for this import; CSV is required for bulk updates of attributes and relations.
• F. Incorrect. .json is a structured data format used in APIs or data exchange, not spreadsheet import. It is not the correct format for bulk updating attributes and relations via spreadsheet import.

Correct answers: B, C, E — Source systems feeding data into the dashboard; Intermediate business processes; The dashboard asset itself and its metadata

• A. Incorrect. Transformation logic is typically part of technical lineage, not business lineage. Business lineage focuses on high-level data flow and business context rather than detailed transformations.
• B. Correct. Source systems are a key component of business lineage, as they represent the origin of the data feeding into the dashboard. The steward would see upstream source systems in the lineage diagram.
• C. Correct. Intermediate business processes are often included in business lineage to show how data is transformed or enriched at a business level before reaching the dashboard. The diagram focuses on the business flow between sources and the dashboard.
• D. Incorrect. Column-level data type conversions are more detailed technical lineage information and are not part of business lineage, which operates at a higher, business-oriented level.
• E. Correct. The dashboard asset itself and its metadata are part of the lineage context because the steward is tracing the metric back to the dashboard where it is presented. The final business asset is typically shown as the endpoint of the lineage.

Correct answer: B — To provide a business-friendly view of data flow.

• A. Incorrect. Business lineage diagrams in Collibra are not designed for documenting technical ETL jobs in detail; that is the role of technical lineage or data flow diagrams. Business lineage focuses on making data movement understandable to business users.
• B. Correct. Business lineage diagrams provide a high-level, business-friendly view of how data flows through an organization, making it accessible to non-technical stakeholders. They help trace data from source to target while emphasizing business meaning and impact.
• C. Incorrect. Enforcing data quality rules is a function of data governance or data quality management tools, not the primary purpose of business lineage diagrams. Lineage may provide context for where rules apply but does not enforce them.
• D. Incorrect. Automating data cleansing and transformation processes is a technical function handled by ETL/ELT tools or data pipelines. Business lineage diagrams visualize relationships and flow, they do not execute data processing.

Correct answers: A, C, D, E — Selecting relevant assets from the asset catalog.; Arranging the assets to accurately represent data flow.; Using Auto Layout to automatically organize the diagram.; Publishing the diagram view for stakeholders to access.

• A. Correct. Selecting relevant assets from the asset catalog is a fundamental step in creating a diagram view, as it determines which assets will be included in the visualization. This is the core first step when building a diagram to represent lineage or relationships.
• B. Incorrect. Defining data quality rules is not part of the diagram creation process. Data quality rules are typically managed separately in Collibra, not on the diagram canvas. The diagram canvas is for visualizing relationships and lineage, not authoring quality rules.
• C. Correct. Arranging the assets to accurately represent data flow is a key part of diagram creation, ensuring the visualization clearly depicts relationships and lineage. The steward positions the assets so the relationships and lineage are clear and meaningful for compliance review.
• D. Correct. Using Auto Layout to automatically organize the diagram is a valid action during diagram creation, as it helps streamline the arrangement of assets and makes the lineage easier to understand.
• E. Correct. Publishing the diagram view for stakeholders to access is part of the process, as it makes the diagram available to others for review and collaboration. After the diagram is created and organized, it can be shared.
• F. Incorrect. Assigning classification levels to each diagram node is not part of the diagram creation process. Classifications are generally assigned to assets, not directly to diagram nodes. The diagram may display those classifications, but they are not assigned node-by-node as part of diagram building.

Correct answer: B — False

• A. The statement is false because users with roles such as Data Steward can also create diagram views if they have the appropriate permissions, not just Sysadmins.
• B. The statement is false because diagram view creation is not restricted to Sysadmins; it is based on the user's assigned permissions and roles, and other roles may also have the capability.

Correct answer: B — Any asset type available in Collibra

• A. Incorrect. Collibra diagrams can include a wider range of asset types beyond just Business Terms and Data Assets, depending on configuration and availability.
• B. Correct. Collibra diagrams support any asset type available in the platform, allowing flexible visualization of relationships across various asset types.
• C. Incorrect. Assets do not need a pre-defined icon to appear in diagrams; icons are optional visual aids, not a restriction.
• D. Incorrect. Diagrams can include assets from multiple domains, not limited to a single specific domain.