Free DAMA CDMP Fundamentals Practice Exam: Data Management Fundamentals

This free full-length DAMA CDMP Data Management Fundamentals practice exam includes 100 original IT Mastery questions across the exam domains.

These are original IT Mastery practice questions. They are not official exam questions, copied live-exam content, or exam dumps. Use them for self-assessment, scope review, and deciding what to drill next.

Count note: this page uses the full-length practice count maintained in the Mastery exam catalog. Some certification vendors publish total questions, scored questions, duration, or unscored/pretest-item rules differently; always confirm exam-day rules with the sponsor.

Exam snapshot

• Practice target: DAMA CDMP Data Management Fundamentals
• Practice-set question count: 100
• Time limit: 90 minutes
• Practice style: mixed-domain diagnostic run with answer explanations

Full-length exam mix

Use this as one diagnostic run. IT Mastery gives you timed mocks, topic drills, analytics, code-reading practice where relevant, and interactive practice.

Practice questions

Questions 1-25

Question 1

Topic: Data Quality

A customer operations team reports that address-validity and completeness scores have improved in its source system for three months. Enterprise customer reports still show inconsistent regional counts because downstream teams apply different standardization and survivorship rules after extraction. What is the best next action?

Options:

• A. Implement end-to-end quality monitoring using shared business rules

• B. Increase sampling frequency in each source application

• C. Replace quality scorecards with report-formatting standards

• D. Assign database custodians to tune reporting tables

Best answer: A

Explanation: Local quality improvement does not guarantee enterprise-quality outcomes. In continuous monitoring, measures should follow critical data elements across the lifecycle, including transformations, integration points, and reporting outputs. When downstream teams apply different rules, the priority is to align quality rules and monitor them end to end so that defects caused after extraction are visible and can be escalated through stewardship. More frequent source checks may confirm that the source is improving, but they will not reveal or resolve inconsistent downstream handling.

• More source sampling may improve confidence in local metrics, but it misses rule variation introduced after extraction.
• Report formatting affects presentation, not the quality rules that produce inconsistent counts.
• Table tuning may improve performance, but it does not address inconsistent standardization or survivorship logic.

Question 2

Topic: Big Data

A company is moving high-volume customer interaction data from a single relational warehouse into a distributed data lake and scalable processing platform. Business users expect faster analytics, but the data will still support regulatory reports and customer segmentation. Which statement best distinguishes the platform capability from the continuing data-management responsibilities?

Options:

• A. Schema-on-read removes the need for business definitions and lineage.

• B. Scalability increases processing capacity, while governance, quality, metadata, and security controls remain necessary.

• C. Raw data zones should be exempt from ownership and access controls.

• D. Replication across nodes makes formal data quality rules unnecessary.

Best answer: B

Explanation: Big data technologies address characteristics such as volume, velocity, and scalability, but they do not remove core data-management obligations. Data still needs accountable ownership, stewardship, quality expectations, metadata for meaning and lineage, and security controls based on sensitivity and allowed use. A distributed platform may change how controls are implemented, but the need for trusted, well-understood, and protected data remains. This is especially important when outputs support regulated reporting or customer analytics.

• Replication misconception fails because multiple copies can improve availability, not correctness or fitness for use.
• Schema-on-read misconception fails because delayed physical structure does not eliminate business meaning, definitions, or lineage.
• Raw-zone exemption fails because sensitive data and accountability requirements apply before data is curated.

Question 3

Topic: Data Storage and Operations

A retailer has 8 years of completed order transactions in its operational database. Customer service needs fast access to the most recent 18 months, finance must retain completed orders for 7 years for audit, and analytics only needs a stable monthly history refreshed after close. Which lifecycle action best fits these needs?

Options:

• A. Retain all 8 years in the operational database

• B. Archive older orders and supply analytics from the archive

• C. Purge all orders older than 18 months

• D. Replicate every order change to analytics immediately

Best answer: B

Explanation: Data lifecycle support aligns availability, retention, performance, and downstream use. Here, recent orders belong in the operational database because customer service needs fast access. Completed orders older than 18 months still have audit value for 7 years, so they should not be purged. Moving them to an archive preserves retention obligations and reduces operational database load. Because analytics needs stable monthly history rather than immediate operational changes, the archived or curated historical store can feed downstream reporting after the close process.

The key is to match the lifecycle state to the business need: active for current service, archived for retained history, and made available to analytics through a controlled downstream process.

• Early purging fails because finance has a 7-year retention requirement for completed orders.
• Immediate replication adds unnecessary operational change flow when analytics only needs monthly closed history.
• Keeping everything active may satisfy access, but it ignores the performance and lifecycle benefits of archiving inactive history.

Question 4

Topic: Data Quality

A sales operations team spends several hours each week correcting customer addresses in the reporting database before monthly revenue reports are published. The same address errors continue to appear after each source-system load. The data quality manager wants a sustainable improvement rather than another cleanup cycle. What is the best next action?

Options:

• A. Increase the reporting database cleanup frequency

• B. Analyze the source capture process and remediate the cause

• C. Create a one-time corrected address extract

• D. Add more report-level filters for invalid addresses

Best answer: B

Explanation: Root-cause analysis focuses on why data defects are created, not only where they are detected. In this case, the recurring address errors appear after each source-system load, so repeated correction in the reporting database treats a symptom. A sustainable remediation effort would examine how addresses are entered, validated, transformed, or loaded; identify the control failure; and change the upstream process or rule so future loads contain fewer defects. Downstream cleanup may still be needed temporarily, but it should not be the primary improvement mechanism when the same error pattern keeps returning.

• More cleanup reduces short-term reporting pain but leaves the defect-generating process unchanged.
• Report filters may hide bad records from reports, but they do not improve the underlying data.
• One-time extract can support an immediate need, but it does not prevent the same errors from reappearing.

Question 5

Topic: Metadata Management

A CRM source system begins sending a new Customer.status value, Dormant. The current transformation rule handles only Active and Inactive.

Exhibit: Lineage excerpt

Which impact assessment best follows from the lineage?

Options:

• A. Review only the CRM source because the defect starts there.

• B. Review the finance report because it also uses a status field.

• C. Review the BI metric and analytics feature for impact.

• D. Review every report containing customer data.

Best answer: C

Explanation: Lineage shows how data moves from a source through transformations to downstream data stores, reports, and analytical uses. The new Dormant value affects assets that depend on CRM.Customer.status through the standardize_status rule and EDW.DimCustomer.status_code. That makes the BI metric and analytics feature in scope for impact analysis. The finance report uses a different field, Billing.Invoice.status, so a similar name is not enough to infer dependency. Lineage-based impact analysis follows actual data flow and transformation paths, not broad subject-area similarity or where the issue first appeared.

• Similar field name fails because Billing.Invoice.status is not shown as downstream from the CRM status field.
• Source-only review misses downstream consumers that may receive invalid, unmapped, or defaulted values.
• All customer reports is too broad because lineage supports targeted impact analysis, not subject-area guessing.

Question 6

Topic: Data Storage and Operations

A finance application is online, response times are normal, and storage utilization is within expected limits. During month-end reconciliation, analysts find that several approved journal entries were loaded twice after an interrupted batch restart, causing account balances to be overstated.

Which type of operational data issue does this most directly indicate?

Options:

• A. Capacity

• B. Performance

• C. Integrity

• D. Availability

Best answer: C

Explanation: Operational data issues are distinguished by the effect observed. Integrity concerns whether data remains accurate, complete, consistent, and trustworthy during storage and processing. In this scenario, the system is available, queries are not slow, and storage is not exhausted. The decisive symptom is that a failed restart created duplicate financial records, so the stored balances no longer faithfully represent the approved transactions.

Availability would focus on whether users or processes can reach the system. Performance would focus on response time or throughput. Capacity would focus on whether storage or processing resources are nearing limits. The key takeaway is to classify the issue by its primary operational impact, not merely by the batch process where it occurred.

• Availability trap fails because users can access the application and the system is online.
• Performance trap fails because response times are normal and no throughput problem is described.
• Capacity trap fails because storage utilization is within expected limits.

Question 7

Topic: Data Modelling and Design

A data governance council is reviewing why sales reports, CRM screens, and finance extracts use different meanings for “customer.” Business stakeholders need to agree on the meaning before application teams change interfaces and analytics teams revise metrics. Which artifact best supports communication and validation across these groups?

Options:

• A. A physical database index design

• B. A conceptual data model with agreed business definitions

• C. A server capacity forecast

• D. A dashboard color and layout standard

Best answer: B

Explanation: Data models support communication by making business concepts, definitions, relationships, and rules explicit. In this situation, the main problem is not database performance or report formatting; it is shared understanding of what “customer” means across business processes, applications, and analytics. A conceptual data model is especially useful because it avoids unnecessary implementation detail while giving business stakeholders, data professionals, application teams, and analytics teams a common structure to review and validate. Once the concept is agreed, logical and physical design work can proceed with less ambiguity. The key value is alignment before downstream systems and metrics are changed.

• Physical design too late focuses on implementation details such as indexes, not shared business meaning.
• Dashboard standard mismatch may improve report usability, but it does not define the underlying business concept.
• Capacity planning mismatch addresses infrastructure sizing, not semantic agreement across stakeholders.

Question 8

Topic: Document and Content Management

A legal department stores signed supplier contracts in an enterprise content management repository. The same PDFs are also attached to procurement records and copied to a shared workspace for project teams. When the contract terms are later disputed, which characteristic best distinguishes the authoritative record from a convenience copy?

Options:

• A. It is easiest for project staff to retrieve

• B. It is stored in the most frequently used repository

• C. It is declared and managed under records controls

• D. It has the newest file modification timestamp

Best answer: C

Explanation: In document and content management, the authoritative record is the managed record that an organization relies on as evidence of a business transaction or obligation. It is distinguished by records controls such as formal declaration, required metadata, retention rules, disposition authority, version control, and legal hold capability. Copies in procurement systems, shared workspaces, email, or search indexes may support convenience, collaboration, or retrieval, but they do not become authoritative merely because users access them often or because they are easier to find. The key distinction is controlled record status, not popularity, location, or file freshness.

• Frequent use can indicate operational convenience, but it does not prove the copy is the official evidence of the transaction.
• Newest timestamp may reflect a copied, renamed, or edited file rather than the controlled record.
• Easy retrieval improves usability, but records authority depends on governance and records controls.

Question 9

Topic: Data Quality

A retail company has profiled its customer master data and created validation rules for required email, postal code, and consent fields. Business leaders now want a monthly view showing pass rates, agreed target levels, trends, and areas needing stewardship follow-up. Which approach best fits this need?

Options:

• A. Create an exception report listing failed records

• B. Run one-time profiling to discover field patterns

• C. Create a monitored data quality scorecard with thresholds

• D. Add validation rules at data entry only

Best answer: C

Explanation: Data quality measurement uses several related practices. Profiling helps discover current patterns, anomalies, and candidate rules. Validation rules test data against agreed expectations. Thresholds define acceptable levels, such as a required pass rate. Exception reports identify specific records that failed and need remediation. A scorecard summarizes quality results over time, often by dimension, rule, data domain, owner, or target level. In this scenario, leaders need monthly pass rates, targets, trends, and stewardship attention areas, so a monitored scorecard is the strongest fit. Detailed exception reports still matter, but they support operational correction rather than executive or governance-level monitoring.

• One-time profiling helps discover issues and rule candidates, but it does not provide recurring target and trend reporting.
• Exception reporting identifies failed records for correction, but it is too detailed for monthly oversight.
• Data-entry validation can prevent some defects, but it does not summarize quality performance across the data set.

Question 10

Topic: Data Modelling and Design

A CRM replacement project is defining the Customer subject area. Sales defines Customer as a company account, while Support defines Customer as an individual who opens cases. They also disagree about whether email address belongs to Customer or Contact. What should the data modeler do next?

Options:

• A. Choose the source system with the most complete records

• B. Ask database administrators to define keys and indexes

• C. Facilitate agreement on business definitions and rules

• D. Create separate physical tables for each department

Best answer: C

Explanation: Disagreement about entity meaning, relationships, or attribute granularity is a conceptual and logical modeling issue first. The data modeler should bring the relevant business stakeholders, data steward, and subject-matter experts together to agree on definitions, relationship rules, and where attributes belong. The decisions should be documented in the conceptual or logical model and aligned with glossary definitions or stewardship decisions. Physical structures, source selection, and database optimization depend on those business rules; they should not be used to settle unresolved semantics.

• Separate tables may preserve departmental views, but it avoids resolving shared business meaning and can institutionalize inconsistency.
• Source completeness is a data quality consideration, not a way to decide what Customer means.
• Keys and indexes are physical design concerns that come after entity definitions and relationships are understood.

Question 11

Topic: Data Modelling and Design

A logical data model for customer orders stores CustomerName, CustomerEmail, ProductName, ProductCategory, and CategoryManager on every OrderLine. Business stakeholders confirm that an order line identifies one ordered product, each product belongs to one category, and each category has one manager. What design action best improves consistency and reduces redundancy?

Options:

• A. Separate Product and Category entities and relate them to OrderLine

• B. Store category details in a JSON column on OrderLine

• C. Create a monthly summary table by product category

• D. Add indexes on ProductName and ProductCategory

Best answer: A

Explanation: Normalization uses business rules and dependencies to place facts where they belong. In this case, ProductName and ProductCategory repeat across many order lines, and CategoryManager depends on category rather than on an individual order line. A better logical design separates stable business concepts into entities such as Product and Category, then links OrderLine to Product and Product to Category. This reduces update anomalies, such as changing a category manager in many order lines, and supports more consistent definitions across the model.

Indexes may improve access performance, but they do not fix redundancy or misplaced facts. Summary tables serve reporting needs, not normalized logical design.

• Indexing repeated columns may speed searches, but it leaves the same product and category facts duplicated on many order lines.
• Embedding category details can increase denormalization and make category manager changes harder to control consistently.
• Monthly summaries are analytical structures and do not address the underlying logical dependencies in operational order data.

Question 12

Topic: Data Modelling and Design

An insurance company is replacing a claims reporting data mart. Business leaders disagree on what counts as a “closed claim,” application teams only have source-system table layouts, and analytics teams need consistent KPI definitions before building dashboards. Data governance is new, but claims data stewards are available. Which action is the best professional decision?

Options:

• A. Select a modeling tool and enforce one notation

• B. Facilitate a validated conceptual and logical model review

• C. Build dashboard prototypes to expose definition gaps

• D. Let application teams derive definitions from source tables

Best answer: B

Explanation: Data models are communication tools as much as design artifacts. In this situation, the main risk is semantic disagreement: “closed claim” means different things to different stakeholders. A conceptual model can align business entities and relationships in business language, while a logical model can add attributes, definitions, and rules that data professionals, application teams, and analytics teams can validate. Involving data stewards supports governance without making the effort overly technical. Physical table layouts and dashboards are useful later, but they should be grounded in agreed business meaning first.

• Source-table definitions fail because physical structures often reflect system design, not agreed business meaning.
• Dashboard prototypes may reveal conflicts, but they do not establish governed definitions before analytics work begins.
• Tool enforcement addresses documentation consistency, not stakeholder validation of claims concepts and KPI definitions.

Question 13

Topic: Data Quality

A bank’s customer onboarding data has recurring defects: missing tax identifiers and duplicate customer records. The data quality team already publishes a monthly scorecard, and the integration team has added validation rules that reject incomplete records. Business users report that the same defects continue to appear from several branches. Which action best fits a quality improvement approach?

Options:

• A. Add more fields to the monthly quality scorecard

• B. Implement stricter technical validation at the integration layer

• C. Analyze root causes with stewards and change the onboarding process

• D. Increase the frequency of data quality audits

Best answer: C

Explanation: Data quality improvement uses measurement as input, but it goes further by identifying root causes and changing the processes that create or maintain data. In this scenario, the scorecard and validation rules reveal and contain defects, but they have not stopped branches from creating missing identifiers and duplicate customers. A suitable improvement action involves business data stewards, process owners, rules, training, and monitoring to prevent recurrence at the source. Audit-only measurement can show that a problem exists, and technical validation can block bad records from moving downstream, but neither necessarily fixes the upstream behavior or accountability gap. The key distinction is prevention and sustained correction, not just detection.

• More scorecard fields may improve visibility, but it remains measurement unless it leads to corrective process action.
• More frequent audits can confirm the defect trend, but auditing alone does not remove the cause.
• Stricter integration validation can reject bad data, but it treats symptoms after capture rather than improving the originating process.

Question 14

Topic: Data Governance

A data governance council is reviewing its first six months of stewardship activity. Executives want a KPI that shows whether governance is reducing business disruption from recurring customer data issues. Issues are logged with severity, assigned to stewards, and have agreed resolution targets. The council needs a trend suitable for monthly executive reporting. Which KPI is the best professional choice?

Options:

• A. Number of customer data governance policies approved

• B. Number of customer data stewards formally appointed

• C. Percentage of customer data elements with approved definitions

• D. Percentage of high-severity customer data issues resolved within target

Best answer: D

Explanation: Governance KPIs should connect the measurement to the stakeholder concern and the intended business outcome. Here, executives are not mainly asking whether governance roles, definitions, or policies exist; they want evidence that stewardship is reducing disruption from recurring customer data issues. Because issues already have severity, ownership, and target resolution times, tracking the percentage of high-severity issues resolved within target creates an outcome-oriented, repeatable monthly KPI. It also focuses attention on the most business-relevant defects rather than counting governance activity. Counts of stewards, definitions, or policies can be useful maturity indicators, but they do not show whether disruptive issues are being resolved effectively.

• Role count shows governance capacity, but appointed stewards do not prove that customer data disruptions are being reduced.
• Definition coverage supports shared understanding, but glossary completeness does not measure issue resolution performance.
• Policy count indicates governance output, but approved policies alone do not show operational impact on recurring issues.

Question 15

Topic: Data Modelling and Design

A bank is designing a conceptual data model for a customer reporting initiative. Retail banking uses “account” to mean a deposit product, commercial banking uses “account” to mean a customer relationship, and marketing uses “client” for the same party that risk calls “customer.” The report must reconcile customer exposure across units, and the data governance function is still informal. What is the best professional decision?

Options:

• A. Create one generic Account entity with a broad type code

• B. Facilitate agreement on distinct business concepts, definitions, synonyms, relationships, and rules

• C. Adopt the terminology used by the largest business unit

• D. Keep each unit’s terms unchanged and resolve meanings in report code

Best answer: B

Explanation: When stakeholders use the same term for different concepts, or different terms for the same concept, the modelling response should separate terminology from meaning. A conceptual model should identify the real business entities, their definitions, relationships, and business rules. The business glossary can then record preferred names and synonyms, such as “client” and “customer,” while distinguishing homonyms such as the two meanings of “account.” This is especially important for cross-unit reporting, where unresolved semantic differences create reconciliation errors and inconsistent metrics. Informal governance makes facilitated agreement even more important, because the model becomes a shared baseline for later logical design, integration, and reporting.

• Largest unit terminology may be politically easy, but it does not resolve conflicting meanings across the enterprise.
• One generic entity hides important business distinctions and can produce ambiguous relationships and rules.
• Report-code resolution pushes semantic conflict into implementation, increasing inconsistency and maintenance risk.

Question 16

Topic: Data Warehousing and Business Intelligence

A data warehouse team finds that a five-year sales trend dashboard has a visible break starting in March, when the CRM source mapping and revenue transformation logic changed. Finance needs a comparable trend for board reporting, auditors require traceability for any restatement, and retained source extracts are available for the affected periods. What is the best professional decision?

Options:

• A. Restate affected history using governed, versioned transformation logic

• B. Create a new dashboard that starts after the mapping change

• C. Adjust only the current month to match prior reports

• D. Leave history unchanged and add a dashboard footnote

Best answer: A

Explanation: Historical trend breaks in a warehouse should be remediated through governed data quality and lineage controls, not by hiding or bypassing the break. Because source extracts are retained and Finance needs a comparable five-year trend, the team should identify the impacted data, apply approved and versioned transformation logic consistently, reload or restate the affected history, and document the lineage and change rationale for auditability. The key is to preserve both analytical comparability and traceability. A footnote may be useful after remediation, but it does not correct inconsistent measures across time.

• Footnote only explains the break but leaves Finance with non-comparable trend values.
• Current-month adjustment creates another unmanaged inconsistency and does not address the affected historical periods.
• New dashboard window avoids the defect but fails the stated need for a five-year comparable trend.

Question 17

Topic: Data Warehousing and Business Intelligence

A finance team is reconciling the monthly Sales Amount in a data warehouse to the sales value shown in the order-entry system. The warehouse total is consistently lower.

Exhibit: Lineage notes

Options:

• A. The order-entry system has stronger access controls.

• B. The warehouse reports net invoiced sales, not gross ordered sales.

• C. The warehouse has a physical partitioning problem.

• D. The report layout uses a different visualization format.

Best answer: B

Explanation: Lineage and reconciliation compare where a measure comes from, which transformations are applied, and whether the compared values represent the same business concept. Here, the operational screen captures gross order value at order placement, while the warehouse measure is based on invoiced value and subtracts credit memos. Those are not equivalent measures, so a lower warehouse total can be expected even when the warehouse load is working correctly. The decisive issue is the measure definition and transformation path, not storage design, security, or presentation.

• Partitioning concern could affect performance or completeness if a load failed, but no evidence shows missing partitions or unloaded data.
• Access controls govern who can see data, not why two authorized totals use different measure definitions.
• Visualization format changes presentation, but the discrepancy is explained by source fields and adjustment rules.

Question 18

Topic: Master and Reference Data Management

A bank uses a shared reference data set for customer lifecycle status across onboarding, servicing, risk reporting, and analytics. The servicing application team wants to add a new status value, Temporarily Suspended, for a local workflow. Which action best fits reference and master data lifecycle governance?

Options:

• A. Submit the value through the enterprise reference data change process

• B. Ask the database team to update indexes for the status table

• C. Create a local report filter that maps the new status

• D. Add the value directly to the servicing application lookup table

Best answer: A

Explanation: Shared reference data is managed as an enterprise data asset, not as a private application setting. A new lifecycle status used by multiple business processes can affect definitions, integrations, reporting, controls, and downstream interpretation. Reference data lifecycle governance should include change request review, business approval, impact assessment, effective dating or versioning when needed, and communication to consuming systems. Local application maintenance is appropriate for screen behavior, workflow configuration, or technical performance work that does not change shared meaning. Here, the proposed value changes a common business code set, so it belongs in the governed reference data change process.

• Local lookup update fails because it could create an unauthorized code value with inconsistent meaning across other consumers.
• Report mapping fails because it hides the governance issue instead of controlling the shared code set at its source.
• Index maintenance fails because performance tuning does not address definition, approval, or lifecycle control for reference data.

Question 19

Topic: Data Architecture

A data architecture team is supporting a cross-domain reporting initiative. Stakeholders need to see which applications are systems of record, how customer and product data moves between platforms, and where each domain should use authoritative data. Which artifact best fits this need?

Options:

• A. Business glossary term list

• B. Physical database schema

• C. Conceptual enterprise data model

• D. Enterprise data flow and lineage view

Best answer: D

Explanation: A data architecture view that combines data flows and lineage is the best fit when stakeholders need to understand where data originates, how it moves, and which uses are authoritative. Systems of record identify trusted originating applications or repositories. Data movement shows transfer, replication, transformation, and consumption paths. Authoritative use clarifies which domain or process should rely on which source for a given data subject. A glossary or model may support shared meaning, but it does not by itself show operational movement across platforms. The key distinction is that this need is about architecture traceability across systems and domains, not only definitions or database structure.

• Glossary focus fails because standard definitions do not show system-to-system movement or authoritative application paths.
• Conceptual model focus helps define subject areas and relationships, but not platform flows or systems of record.
• Schema focus is too implementation-specific and usually limited to one database or application.

Question 20

Topic: Data Security

A retail company wants to let an external analytics partner build a churn model using customer transaction history. The data includes customer identifiers, contact details, and purchase behavior. The partner needs behavioral patterns but does not need names or full contact details. The company has a data classification policy and requires business owner approval for sensitive data sharing. Which action should the data management lead recommend?

Options:

• A. Classify as restricted and provide approved, minimized access

• B. Classify as internal and share the complete dataset

• C. Deny all external access to customer data

• D. Grant full read access because a contract exists

Best answer: A

Explanation: Data classification and access control should align sensitivity, business purpose, and stakeholder role. Customer identifiers and contact details increase confidentiality and privacy risk, so the dataset should be treated as restricted or otherwise highly controlled under the organization’s policy. Because the partner only needs behavioral patterns, access should be minimized, approved by the business owner, and limited to the necessary fields or a de-identified view where practical. This supports least privilege while still enabling the legitimate analytics use. A contract may be required, but it does not replace classification, approval, and data minimization.

• Contract alone fails because legal agreement does not justify unrestricted access to sensitive customer fields.
• Internal classification understates the sensitivity of identifiable customer and behavioral data.
• Complete denial is too restrictive when a controlled, minimized dataset can meet the business need.

Question 21

Topic: Data Quality

A retailer’s customer address data is mostly complete, but failed deliveries and tax errors occur when international records use postal codes that do not match the recorded country. The data quality team needs a measurement approach that quantifies the defect pattern by country and produces actionable exceptions for daily remediation. Which approach best fits this need?

Options:

• A. Run duplicate detection on customer names

• B. Report the overall address completeness percentage

• C. Review a random sample of customer records manually

• D. Measure country-specific validity rules and exception rates

Best answer: D

Explanation: The defect is not missing data; it is invalid data relative to a domain rule: postal-code format must be appropriate for the recorded country. A suitable measurement approach defines data quality rules for each relevant country, applies them to the address population, reports exception rates by country, and identifies the records that fail. This supports both trend measurement and operational correction. Overall completeness may look high while country-specific validity remains poor. Sampling can estimate a problem, but it is weaker when daily exception handling is required.

• Completeness only misses values that are present but invalid for their country.
• Manual sampling may estimate defect levels, but it does not reliably produce daily record-level remediation lists.
• Duplicate detection addresses possible duplicate customers, not postal-code validity against country rules.

Question 22

Topic: Document and Content Management

A legal department stores contract templates and executed contracts in a shared content repository. During an internal review, many documents have no named business owner, confidentiality classification, retention period, or version history. Which control weakness is most directly indicated?

Options:

• A. Ineffective master data matching

• B. Incomplete relational data normalization

• C. Uncontrolled document lifecycle accountability

• D. Poor dashboard metric governance

Best answer: C

Explanation: Document and content management controls depend on metadata that supports ownership, protection, retention, and version control. Without a named owner, no accountable party can approve changes or resolve issues. Without classification, sensitive content may be handled incorrectly. Without retention metadata, records may be kept too long or disposed of too soon. Without version history, users may rely on obsolete or unauthorized documents. The weakness is not mainly about database design, master data consolidation, or BI reporting; it is a records and content lifecycle control gap.

• Normalization is about structuring relational data to reduce redundancy, not controlling document retention or versions.
• Master data matching addresses duplicate or inconsistent core business entities, not unmanaged content records.
• Metric governance concerns consistent BI definitions and reporting rules, not ownership and retention of documents.

Question 23

Topic: Metadata Management

A finance team finds a certified data set named Customer Revenue in the enterprise catalog. The catalog shows the owner, refresh schedule, source system, and field names. Analysts still avoid using it because they cannot tell whether “revenue” means invoiced amount, recognized revenue, or cash received. Which metadata gap most directly prevents correct use of this data asset?

Options:

• A. Missing physical storage location

• B. Missing refresh frequency

• C. Missing data owner assignment

• D. Missing business definition for the measure

Best answer: D

Explanation: Metadata supports trust and correct use by explaining what a data asset means, where it came from, how it is managed, and how it should be used. In this case, technical and operational details are already present: source system, field names, owner, and refresh schedule. The blocking issue is semantic. Analysts cannot determine which business concept the revenue measure represents, so they may produce inconsistent reports or compare incompatible figures. A business glossary definition or approved metric definition would clarify the intended meaning, calculation basis, and acceptable use.

• Storage details would help locate or manage the asset, but location does not resolve the meaning of the revenue measure.
• Refresh frequency is already available and addresses timeliness, not semantic interpretation.
• Owner assignment is already present and supports accountability, but ownership alone does not define the measure.

Question 24

Topic: Data Storage and Operations

A company is clarifying responsibilities for a critical customer database used by order processing and reporting teams. Which activity is primarily an operational database administration responsibility rather than a business stewardship or enterprise data architecture responsibility?

Options:

• A. Define the enterprise customer subject-area roadmap

• B. Monitor backups, recovery, performance, and availability

• C. Resolve conflicting business definitions of customer status

• D. Approve enterprise principles for shared customer data

Best answer: B

Explanation: Database operations and support are concerned with keeping database services reliable, recoverable, secure, and performant in day-to-day operation. These responsibilities commonly include backup and recovery, monitoring, capacity support, tuning, and operational maintenance. Business stewardship is different: it focuses on meaning, usage, quality expectations, issue resolution, and accountability for data from a business perspective. Enterprise data architecture is also different: it defines broader structures, principles, subject areas, and target-state direction for data across the organization. The key distinction is operational control of the database environment versus business accountability for data meaning or enterprise-level design direction.

• Business definitions belong to stewardship because they concern shared meaning and business accountability, not database operations.
• Enterprise principles belong to data architecture or governance because they guide organization-wide design decisions.
• Subject-area roadmap is architectural planning for enterprise data, not day-to-day database administration.

Question 25

Topic: Data Security

A retail analytics team needs a customer dataset for self-service dashboards showing purchasing trends by region and age band. The source contains classified sensitive attributes, including names, email addresses, and tax identifiers. Analysts do not need those attributes, but they need stable customer counts and segmentation. Governance policy requires least-privilege access and reusable controls. What is the best professional decision?

Options:

• A. Encrypt the sensitive columns in place

• B. Grant read access after analyst approval

• C. Publish a governed masked analytical view

• D. Remove all customer-level records from analytics

Best answer: C

Explanation: Data classification and access control should align access to the business need. Here, analysts need useful analytical data, not direct sensitive attributes. A governed analytical view can expose approved fields such as region and age band, mask or suppress direct identifiers, and apply role-based access consistently. This supports least privilege while keeping the dataset reusable for dashboards and trend analysis.

Encryption protects data at rest or in transit, but it does not by itself decide who may see decrypted values. Broad read access with approval still overexposes classified data. Removing all customer-level records may protect privacy, but it can also eliminate the counts and segmentation needed for analysis.

• Approval alone does not reduce the data exposed to the minimum needed for the analytics task.
• Encryption in place protects storage or transmission, but authorized users could still see decrypted sensitive values.
• Full record removal may be overly restrictive because the team still needs stable counts and segments.

Questions 26-50

Question 26

Topic: Master and Reference Data Management

A regional insurance company finds that each claims office maintains its own loss-cause codes. Several local values overlap, some have different meanings by office, and enterprise reports cannot reliably group claims by loss category. Which reference data governance action best addresses the problem?

Options:

• A. Increase storage capacity for the claims data repository

• B. Assign each claim a unique master record identifier

• C. Establish an enterprise code set with controlled change approval

• D. Create a new dashboard filter for each claims office

Best answer: C

Explanation: Reference data governance manages valid values, code sets, classifications, and their meanings across the organization. When local offices create overlapping or conflicting codes, the needed action is to define an authoritative enterprise code set, assign stewardship, document definitions and mappings, and control future changes through an approval process. This addresses both the current inconsistency and the lifecycle problem that allowed uncontrolled values to spread. Reporting can then group claims consistently because each value has an agreed business meaning. Dashboard filters may hide variation but do not standardize the code set, and master identifiers address entity identity rather than reference values.

• Report filtering may help users view office-specific data, but it leaves conflicting code meanings unresolved.
• More storage addresses capacity, not inconsistent classification values or code governance.
• Master identifiers support matching and identifying core entities, but they do not define approved reference values.

Question 27

Topic: Metadata Management

A financial services company is building a governed customer analytics platform. Business teams disagree on the definition of “active customer,” integration teams cannot trace fields from CRM and billing into the warehouse, security needs PII classifications before broader access is approved, and duplicate customer records are delaying master data decisions. Which action is the best professional decision?

Options:

• A. Implement an MDM matching tool before documenting definitions, lineage, or stewardship accountability

• B. Redesign the warehouse schema first so all reporting tables use the same physical naming standards

• C. Create an enterprise metadata catalog with governed definitions, lineage, classifications, mappings, quality rules, and stewardship roles

• D. Ask each department to publish its own metric glossary for use in its dashboards

Best answer: C

Explanation: Metadata supports multiple data management activities when it describes data meaning, structure, origin, movement, ownership, controls, and fitness for use. In this scenario, the organization needs shared business definitions for governance, lineage for integration and warehousing, classifications for security, quality rules for trust, and stewardship accountability for master data decisions. A governed metadata catalog or repository can connect these views so different teams work from consistent, traceable information. Physical design changes or isolated glossaries may help locally, but they do not address the cross-domain need to manage customer data as a governed enterprise asset.

• Warehouse redesign addresses physical structure and naming, but it does not resolve business definitions, lineage, PII classification, or master data accountability.
• MDM tool first may improve matching later, but duplicate resolution needs agreed definitions, identifiers, rules, and stewardship context.
• Department glossaries can document local meanings, but separate glossaries risk reinforcing conflicting definitions instead of governing shared metadata.

Question 28

Topic: Data Modelling and Design

A data modelling review finds that the same business concept is represented as Customer, Client, and Account Holder across related logical models. The approved business glossary has one definition for Customer, but the models do not show links to glossary terms. Which review action best improves model quality and traceability?

Options:

• A. Create a new report formatting standard

• B. Move the models to a shared repository

• C. Require glossary-linked entity and attribute definitions

• D. Tune physical indexes for the affected tables

Best answer: C

Explanation: Model governance and design standards help ensure that data models use consistent names, definitions, and relationships that reflect business meaning. When several model elements represent the same concept under different names, the quality issue is semantic alignment, not storage performance or presentation. A review standard that requires entities and attributes to be tied to approved business glossary definitions supports consistency, reuse, and traceability from requirements through models to implementation. A repository can help manage artifacts, but it does not by itself resolve inconsistent business meaning.

• Physical tuning addresses database performance after implementation, not inconsistent logical business definitions.
• Report formatting may improve dashboard usability, but it does not govern model semantics.
• Shared repository can support control and versioning, but it needs definition-linking standards to fix the issue.

Question 29

Topic: Data Modelling and Design

Two active projects are designing customer data for separate applications. One model treats Customer as an individual person only; the other treats Customer as either a person or an organization. Both projects claim their model is correct for the same enterprise customer subject area. What is the best next action?

Options:

• A. Escalate to model governance to agree the shared subject-area definition

• B. Allow each project to keep its model independently

• C. Resolve the conflict during physical database design

• D. Choose the model from the project with the larger budget

Best answer: A

Explanation: When projects define incompatible models for the same business subject area, the issue is not primarily a physical design or project-priority decision. It is a model governance and design standards issue. A shared enterprise or subject-area view should clarify the business meaning of Customer, approved definitions, allowed subtypes, naming standards, and relationships. Data owners, stewards, and model governance participants should resolve the semantic conflict so project-level logical and physical models can align where reuse and consistency are required. The key is to govern the meaning first, then let implementation models specialize within agreed standards.

• Independent models may meet local needs, but they preserve conflicting enterprise meanings for the same subject area.
• Budget-based selection uses a project management factor rather than business semantics or governed design standards.
• Physical design resolution is too late because table structures should implement an agreed logical meaning, not decide it.

Question 30

Topic: Data Architecture

A regional insurer is replacing several policy administration systems over three years. Claims, billing, and customer service teams need consistent customer and policy data for reporting and integration, but each program is currently defining its own data structures. Governance exists but is lightweight, with only part-time data stewards. Which architectural response is the best professional decision?

Options:

• A. Wait until all systems are replaced before harmonizing data

• B. Establish enterprise data architecture principles and shared domain models

• C. Focus only on selecting a single integration platform

• D. Let each replacement program define its own canonical structures

Best answer: B

Explanation: Enterprise data architecture alignment provides cross-domain direction so programs can make compatible decisions over time. In this scenario, the main risk is that separate replacement programs will create inconsistent definitions and structures for customer and policy data. Establishing architecture principles, common subject-area views, and shared domain models supports reuse, integration, reporting consistency, and governance without requiring a fully centralized redesign. Lightweight governance can still review and maintain these core artifacts through part-time stewardship. A platform may help move data, but it does not by itself align meaning, ownership, or structure across domains.

• Program autonomy increases speed locally but preserves inconsistent customer and policy definitions across the enterprise.
• Delayed harmonization allows divergence to become embedded in new systems, making later correction more expensive.
• Platform-only focus addresses connectivity, not shared business meaning, governance alignment, or reusable architectural standards.

Question 31

Topic: Data Governance

A company is creating a shared customer data standard. Sales, billing, and support disagree on which business unit has authority to approve the official definition of “active customer” and the rules for resolving conflicting customer attributes. Which action best fits this need?

Options:

• A. Ask the database administrator to update the customer table schema

• B. Tell data entry staff to choose the most complete customer record

• C. Assign decision rights to a data governance council and accountable data owner

• D. Have the project manager add the issue to the delivery schedule

Best answer: C

Explanation: Data governance establishes authority and accountability for decisions about data as an enterprise asset. When multiple business areas disagree on the meaning of a shared term or on rules for resolving conflicts, the issue requires formal decision rights, not only technical work or task coordination. A governance council, data owner, or similar accountable governance role can approve the business definition, standard, and escalation path. IT administration may implement approved changes, project management may coordinate work, and operational staff may follow procedures, but they should not independently decide enterprise-wide business meaning or stewardship policy.

• Schema update treats the issue as a technical implementation task before the business authority and rules are approved.
• Delivery scheduling may track the work, but it does not establish who has authority to decide the standard.
• Record selection leaves enterprise policy to operational staff and risks inconsistent conflict resolution.

Question 32

Topic: Metadata Management

A data governance council is resolving a reporting dispute over the term “active customer.” Business teams need an approved, plain-language definition, accountable owner, and usage notes that can be reused across reports. They do not need column names, table layouts, system inventory details, or source-to-target flow diagrams.

Which metadata artifact best fits this need?

Options:

• A. Business glossary

• B. Lineage map

• C. Data dictionary

• D. Data catalog entry

Best answer: A

Explanation: A business glossary is the best fit when the need is to standardize business meaning. It captures terms such as “active customer” in plain language, along with business ownership, approved definitions, and usage guidance. A data dictionary is more technical and describes data elements such as columns, data types, constraints, and valid values. A data catalog entry helps users discover and understand data assets, often combining technical, operational, and governance metadata. A lineage map shows how data moves and transforms across systems. The key distinction is that the glossary governs meaning, while the other artifacts describe structures, assets, or movement.

• Column-level detail belongs in a data dictionary, not in an approved business-term definition.
• Asset discovery is the main purpose of a data catalog entry, even if it can link to glossary terms.
• Data movement is shown by lineage, which explains sources, transformations, and destinations rather than business meaning.

Question 33

Topic: Document and Content Management

An enterprise content repository is used for supplier contracts. An audit finds the same signed contract stored in several team folders, inconsistent search results, retention jobs skipping many files because record class is blank, and permissions inherited from locally created folders. Which governance weakness most directly explains the issue?

Options:

• A. No enforced content taxonomy, metadata, and lifecycle ownership standards

• B. Limited full-text indexing across contract PDFs

• C. No vendor master data hub for supplier identifiers

• D. Insufficient repository storage tiering for older documents

Best answer: A

Explanation: Content governance defines how documents are classified, described, owned, secured, retained, and disposed of across their lifecycle. In this case, duplicate contract copies, inconsistent discoverability, blank retention classification, and ad hoc folder permissions all indicate that repository rules are not being governed or enforced consistently. A taxonomy and required metadata support findability and retention; ownership and lifecycle rules support disposition and access accountability.

A search-indexing problem might explain poor search results, but it would not explain skipped retention processing or uncontrolled permissions.

• Vendor master data could improve consistent supplier identification, but it does not govern document classification, retention, or folder-based access.
• Storage tiering addresses cost and performance for stored content, not duplicate control, metadata completeness, or retention classification.
• Full-text indexing may improve search, but it does not resolve missing record classes or locally inherited permissions.

Question 34

Topic: Data Governance

A data governance council reviews the quarterly scorecard for the customer data domain and wants to identify the primary governance problem.

Which problem does the scorecard most clearly indicate?

Options:

• A. Stewardship capacity

• B. Control effectiveness

• C. Policy adoption

• D. Issue backlog

Best answer: C

Explanation: Governance scorecards should be interpreted by comparing each measure with its target and then linking the gap to the governance capability it represents. Here, the only significant miss is policy attestation by data users: 42% against a 95% target. That points to weak adoption or awareness of the approved policy. Stewardship capacity appears adequate because named steward coverage exceeds the target. Issue backlog is not the main concern because aging issues are below the threshold. Control effectiveness is also not the primary problem because audit exceptions are within tolerance. The scorecard is signaling a change-management and adoption gap, not a lack of roles, unresolved issues, or failed controls.

• Steward coverage is not the issue because critical data elements already exceed the target for named stewards.
• Issue aging is not the issue because open items older than 30 days are below the threshold.
• Access controls are not the issue because audit exceptions remain within the stated tolerance.

Question 35

Topic: Metadata Management

A data catalog has high search adoption, and analysts can locate customer and policy datasets quickly. However, regulatory reporting teams still misuse fields because they do not understand business meaning, source-to-report lineage, known quality issues, or whether a dataset is approved for a specific report. Stewardship capacity is limited, so the improvement must add the most decision value without rebuilding the platform. What is the best professional decision?

Options:

• A. Add more technical column names and physical storage locations

• B. Move the reporting datasets into a separate data mart

• C. Enrich catalog entries with governed business, lineage, quality, and usage metadata

• D. Create a new searchable index across all data stores

Best answer: C

Explanation: The problem is not data discovery; users can already find datasets. The missing value is contextual metadata that helps users decide whether data is understandable, trustworthy, traceable, and appropriate to use. In DAMA-aligned metadata management, a catalog becomes more useful when technical metadata is connected to business glossary definitions, lineage, data quality indicators, ownership or stewardship, classifications, and approved-use guidance. With limited stewardship capacity, focusing enrichment on high-risk regulatory reporting datasets is more practical than replacing the tool or redesigning the data architecture. The key distinction is searchability versus interpretability and governed use.

• Technical inventory only fails because storage locations and column names do not explain business meaning, quality, lineage, or permitted use.
• More search coverage fails because users already find datasets; the gap is understanding and trust.
• Separate data mart may support reporting delivery, but it does not by itself provide governed definitions, lineage, quality context, or usage approval.

Question 36

Topic: Data Governance

A retailer repeatedly finds duplicate customer records, conflicting definitions of “active customer,” and unresolved disputes about which team may change customer status codes. Each department applies local fixes, but the same issues return during onboarding and reporting. Which governance action best addresses the pattern?

Options:

• A. Establish governed ownership, standards, and escalation for customer data

• B. Ask database administrators to restrict all status-code changes

• C. Run a one-time customer data cleansing project

• D. Create another report to track duplicate customer records

Best answer: A

Explanation: Repeated data issues that cross teams usually indicate a governance gap, not only a technical or cleanup problem. DAMA-aligned data governance should clarify decision rights, assign business accountability, approve common definitions and standards, and provide an issue-management path for escalation and resolution. Because the same defects recur in onboarding and reporting, the response should also integrate the agreed controls into the business process so the issue is prevented rather than repeatedly corrected after the fact. A cleansing effort or monitoring report may help temporarily, but it does not resolve unclear ownership or inconsistent standards.

• One-time cleanup treats symptoms but leaves ownership, definitions, and onboarding controls unresolved.
• Technical restriction may control a field but does not decide business accountability or shared definitions.
• Additional reporting improves visibility but does not create standards, decision rights, or escalation.

Question 37

Topic: Data Modelling and Design

A retailer is preparing a data model for a customer 360 reporting initiative. The business has already agreed that Customer, Order, Product, and Promotion are in scope. No database platform has been selected. During review, sales and finance disagree about whether an Order can include multiple Promotions and how Promotion relates to Order Line. Incorrect handling could double-count campaign revenue. What is the best professional decision?

Options:

• A. Resolve logical relationships and cardinalities before physical design

• B. Handle the issue only in dashboard calculations

• C. Defer the issue to physical table and index design

• D. Reopen conceptual scope for the customer 360 model

Best answer: A

Explanation: Conceptual models establish the major business subject areas and high-level entities in scope. Logical models refine those entities into relationships, cardinalities, optionality, and business rules without depending on a specific database technology. Here, the entities are already agreed and no platform has been selected. The unresolved question is how Promotion relates to Order and Order Line, with a reporting risk if the relationship is modeled incorrectly. That makes it a logical modelling issue, not a conceptual scope issue or a physical implementation decision. The key takeaway is to resolve business relationship rules before translating the model into tables, keys, indexes, or reports.

• Conceptual scope is not the main issue because the relevant entities have already been agreed as in scope.
• Physical design is premature because no database platform has been selected and the relationship rules are still unclear.
• Dashboard calculations would mask the modelling problem rather than establish a trusted shared structure for reporting.

Question 38

Topic: Data Warehousing and Business Intelligence

The sales VP and CFO disagree about Monthly Recurring Revenue on an executive dashboard. Sales uses CRM subscription status, Finance uses billed amounts excluding credits and trials, and the warehouse transformation rules are not documented. Executives want one trusted KPI for decision support. What is the best next action?

Options:

• A. Govern the KPI definition, sources, transformations, and lineage

• B. Use the Finance report because it supports accounting close

• C. Build a new dashboard directly from the CRM source

• D. Average the two reported values for executive reporting

Best answer: A

Explanation: When executives disagree about a KPI, the core issue is usually metric governance, not visualization. BI decision support depends on consistent business definitions, approved calculation rules, known sources, and documented transformations. The best action is to bring the accountable business stakeholders and data stewards together to define the KPI, document its source-to-target lineage, and publish the approved definition in the BI metadata or metric catalog. This creates traceability and repeatable interpretation across dashboards. Choosing one department’s report or creating another dashboard may be faster, but it does not resolve conflicting definitions or undocumented transformations.

• Finance-first shortcut may align with accounting needs, but it does not establish an enterprise KPI definition across stakeholders.
• Averaging values hides the disagreement and produces a number with no defensible business meaning.
• CRM-only reporting changes the source but still leaves exclusions, transformations, and metric ownership unresolved.

Question 39

Topic: Master and Reference Data Management

A regional sales report shows different totals for the same product category depending on which source system feeds the report. Investigation finds that one system maps code A1 to “Accessories,” while another maps A1 to “Add-ons.” Both mappings are used in production reporting. What is the best response?

Options:

• A. Establish an governed reference code set and approved mappings

• B. Ask each report owner to rename the category locally

• C. Create a new dashboard that hides the source-code values

• D. Treat the issue as a duplicate master data record

Best answer: A

Explanation: Reference data management controls code sets, valid values, meanings, and mappings used to classify business data. When reports disagree because the same code is mapped to different meanings, the durable fix is to define and govern the reference code set and its cross-system mappings. That includes business-approved definitions, ownership, change control, and distribution to consuming systems. Local report changes may mask symptoms but do not resolve the shared semantic inconsistency. The key takeaway is that inconsistent classification codes require governed reference data standards, not cosmetic reporting changes or master data deduplication.

• Local renaming fails because each report would still maintain its own interpretation instead of using a controlled enterprise mapping.
• Hiding codes fails because presentation changes do not correct the inconsistent meaning used in aggregation.
• Master data duplicate fails because the problem concerns classification values and mappings, not duplicate records for business entities.

Question 40

Topic: Master and Reference Data Management

A retail company maintains a reference hierarchy for sales regions. During a reorganization, the data team overwrites the existing region hierarchy with the new structure instead of adding effective dates and preserving the prior version. Current operational systems use the same new hierarchy, but prior-month BI reports now group historical sales differently each time they are refreshed.

Which effect best describes this lifecycle failure?

Options:

• A. Inconsistent operational processing between systems

• B. Loss of regulatory traceability for submitted filings

• C. Reduced reporting comparability across periods

• D. Duplicate creation of master customer records

Best answer: C

Explanation: Reference data lifecycle management should preserve controlled changes, including versioning and effective dating when classifications change over time. In this case, operational systems are aligned on the current hierarchy, so the main failure is not inconsistent transaction processing. The visible business impact is that historical sales are regrouped under the latest hierarchy when BI reports refresh, making prior-month and current-month results difficult to compare on a like-for-like basis. Effective-dated reference data would allow reports to apply the classification that was valid for each reporting period. The key distinction is unstable analytical comparison, not duplicate master data or a regulatory audit problem stated in the scenario.

• Operational inconsistency does not fit because the current systems are described as using the same new hierarchy.
• Regulatory traceability would require a compliance filing or audit trail concern, which is not stated.
• Duplicate master records relates to master data identity resolution, not hierarchy version control.

Question 41

Topic: Data Governance

A data governance council is reviewing a quarterly scorecard for customer data. The business goal is to improve trusted cross-sell reporting before a new campaign. Governance policies were approved last year, and executive sponsors want the first improvement action to address the most limiting governance problem.

Scorecard excerpt:

What is the best professional interpretation of the scorecard?

Options:

• A. Policy adoption is the primary weakness.

• B. The scorecard shows no material governance problem.

• C. Stewardship capacity is constraining governance execution.

• D. Control effectiveness is below expectations.

Best answer: C

Explanation: Governance scorecards should be interpreted by comparing each measure with its target and then linking the gap to a governance capability. In this case, policy adoption is close to target because standard definition use is 88% against a 90% target. Control effectiveness is meeting target because access-control remediation is 96% against a 95% target. The largest capability gap is steward availability: only 5 stewards have protected time against a need for 12. The high number of aging issues is a likely symptom of that capacity constraint, because unresolved governance issues often require steward analysis, decision preparation, and escalation. The first improvement action should therefore focus on stewardship capacity rather than rewriting policies or tightening controls.

• Policy adoption gap is tempting because definitions matter for reporting, but the adoption measure is only slightly below target.
• Control weakness does not fit because access-control exception remediation is exceeding the stated target.
• No material problem ignores the severe steward shortfall and the large aged issue backlog.

Question 42

Topic: Master and Reference Data Management

A regional insurer is consolidating claims data from three business units. Each unit uses different local claim-status codes, and regulatory reports require a consistent enterprise classification. Local operations still need to see their legacy values during a transition. What is the best professional decision?

Options:

• A. Ask IT to rename the fields to a common column name

• B. Create a governed enterprise code set with mappings and change control

• C. Load all local codes into the warehouse without standardization

• D. Let each unit keep its codes and reconcile only report totals

Best answer: B

Explanation: Reference data governance establishes approved code sets, definitions, ownership, stewardship, mapping rules, and change control for values used across processes and reports. In this situation, the problem is not just different field names or reporting layout; it is conflicting classification schemes that affect regulatory reporting and integration. The strongest action is to define an enterprise claim-status code set, assign accountable governance roles, map each local value to the standard, and control future changes. This supports consistent reporting while allowing local legacy values to remain visible during transition. Technical consolidation alone does not resolve semantic inconsistency.

• Reconciling totals hides code-level inconsistencies and does not create an authoritative classification for future reporting.
• Renaming fields addresses syntax, not the meaning or permitted values of claim-status codes.
• Loading all codes preserves inconsistency and pushes interpretation problems into every downstream report or integration.

Question 43

Topic: Data Security

A regional health insurer is creating a shared analytics workspace for care-quality reporting. Analysts need member-level claims data to validate quality measures, but the workspace will also include protected health information. The governance policy requires role-based access, quarterly access reviews, and documented business justification for sensitive data use. Which action is the best professional decision?

Options:

• A. Let project managers approve ad hoc access

• B. Grant analysts only the PHI fields required for validation

• C. Give all analysts full claims-table access

• D. Mask all member identifiers for every user

Best answer: B

Explanation: Least privilege means users receive only the access necessary to perform an approved business activity. Need-to-know adds that sensitive or regulated data should be limited to people with a justified role in using it. In this case, analysts have a valid need to validate care-quality measures, but that does not justify unrestricted claims-table access. The appropriate control is role-based access to only the PHI fields needed, supported by documented justification and periodic review as required by policy.

Over-restricting all identifiers may prevent valid measure validation, while broad access or informal approval weakens governance and increases privacy risk.

• Full-table access fails because it exceeds the analysts’ stated need and exposes more regulated data than necessary.
• Universal masking may block legitimate validation work if member-level identifiers are required for the approved purpose.
• Ad hoc approval bypasses the required role-based process, business justification, and quarterly review controls.

Question 44

Topic: Data Quality

A data quality analyst profiles customer onboarding data against the rule, “Legal name must be populated before account activation.” The initial profile shows a 22% failure rate.

Exhibit: Failure pattern

What is the best next action?

Options:

• A. Correct the source-system capture process

• B. Escalate the issue to the governance council

• C. Refine the quality rule scope

• D. Send the exceptions to data stewards for review

Best answer: C

Explanation: Profiling evidence should be used to distinguish real data defects from misapplied or incomplete data quality rules. Here, most failures are prospects or pending accounts, while the business definition says activation means Status = Active. The source behavior also matches the business process: legal name is captured when the contract is signed. The apparent defect is therefore caused by applying the rule too broadly, not by bad source data or unresolved accountability. The next action is to refine the rule filter or condition so measurement reflects the intended population. Remediation should wait until the rule correctly separates valid exceptions from true defects.

• Source correction would fit if active accounts were missing legal names because the capture process failed.
• Stewardship review would fit if individual exceptions required business judgment or manual resolution.
• Governance escalation would fit if definitions or accountabilities conflicted across domains and could not be resolved operationally.

Question 45

Topic: Data Quality

A data quality team found that a source-to-target mapping truncates product category codes during a nightly warehouse load. Several months of sales records now contain invalid category values. The business needs the defect corrected, not merely reported or bypassed. Which action best represents corrective remediation?

Options:

• A. Fix the mapping and reload the affected records

• B. Add a dashboard showing invalid category counts

• C. Tell analysts to exclude invalid categories temporarily

• D. Require future category codes to pass validation

Best answer: A

Explanation: Corrective remediation addresses an identified data quality defect by fixing the cause and correcting the affected data. In this case, the mapping error created invalid values in historical warehouse records, so remediation should both repair the mapping and reload or correct the impacted records. A preventive control helps stop future defects before they occur. Detective monitoring identifies or alerts on defects. A temporary workaround reduces immediate business disruption but leaves the underlying defect unresolved. The key distinction is whether the action actually removes the known problem and restores trustworthy data.

• Monitoring counts is detective because it reveals the size or trend of invalid categories but does not correct them.
• Future validation is preventive because it blocks new bad values but does not repair the existing warehouse records.
• Analyst exclusion is a workaround because it avoids using bad data temporarily while the defect remains.

Question 46

Topic: Data Integration and Interoperability

A retailer wants near-real-time inventory availability on its website after sales occur in stores and online. The source sales systems must remain operational during business hours, and the website only needs a small set of product and stock fields rather than full transactional history. Which integration pattern is the best professional decision?

Options:

• A. Streaming event-based integration

• B. Nightly batch file transfer

• C. Full database replication

• D. Manual API polling by the website

Best answer: A

Explanation: Streaming event-based integration is suited to continuously moving small, time-sensitive changes, such as sales or stock adjustments, from operational systems to consuming applications. It supports low-latency availability while avoiding the need to copy full transaction histories. In this situation, the website needs current inventory fields, not complete sales records, and the sales systems should not be heavily queried during operating hours. A controlled event stream or message-based feed can publish relevant changes as they occur and allow downstream services to update product availability. Batch transfer is simpler but introduces delay, while full replication moves more data than needed and can increase operational and governance complexity.

• Nightly batch transfer fails because it cannot meet the near-real-time inventory requirement.
• Full replication moves unnecessary transactional data and adds avoidable complexity for a narrow website use case.
• Manual API polling can increase source-system load and is less reliable than publishing changes as events.

Question 47

Topic: Big Data

A retailer is launching a customer insight platform that must ingest clickstream events in near real time, combine them with daily point-of-sale files and product reviews, and support trusted executive reporting. Early profiling shows frequent spikes in event volume, changing review formats, and inconsistent customer identifiers across sources. Which professional decision best addresses these big data characteristics?

Options:

• A. Load all sources into a single relational table and standardize fields after executive reports are published

• B. Design a governed ingestion and processing approach with scalable storage, streaming capability, metadata, quality rules, and stewardship

• C. Restrict ingestion to point-of-sale files because they are structured and easier to reconcile

• D. Prioritize dashboard design first and defer source profiling until business users identify reporting gaps

Best answer: B

Explanation: Big data characteristics affect both technology choices and data management controls. High and spiking volume points to scalable storage and processing. Near-real-time clickstream data requires ingestion and processing patterns that can handle velocity. Reviews, files, and events introduce variety and changing formats, so metadata and adaptable integration rules are needed. Inconsistent customer identifiers raise veracity and quality concerns, requiring quality rules, matching controls, and stewardship. A professional response balances ingestion, storage, processing, quality, and governance rather than treating big data as only a storage or reporting problem.

• Single-table loading ignores velocity and variety, and fixing definitions after reports are published undermines trust.
• Dashboard-first work treats symptoms rather than managing profiling, metadata, and quality at ingestion.
• Restricting sources avoids complexity but fails the business goal of combining behavioral, sales, and review data.

Question 48

Topic: Data Governance

A data governance council reports strong participation: 95% attendance, weekly meetings, and a growing issue log. However, customer and product definitions remain disputed across business units, and the same quality defects recur in regulatory reports. The executive sponsor wants evidence that governance is improving business outcomes, not just showing activity. Which improvement is the best professional decision?

Options:

• A. Increase meeting frequency and require more status updates

• B. Move definition decisions to the database administration team

• C. Add outcome-based governance measures and accountable issue resolution

• D. Track only attendance and number of issues submitted

Best answer: C

Explanation: Governance metrics should show whether governance is producing value, not merely whether governance bodies are busy. High attendance and frequent meetings are activity measures; they do not prove that definitions are being agreed, ownership is clear, or recurring data quality defects are being eliminated. A better improvement is to add outcome-focused measures such as time to approve critical definitions, percentage of issues resolved by accountable owners, recurrence rate of priority defects, and impact on regulatory reporting reliability. These measures should be paired with clear stewardship and escalation paths so decisions actually close. The key is shifting from participation tracking to decision effectiveness and measurable reduction of business risk.

• More meetings can increase activity without resolving definition conflicts or recurring defects.
• Technical ownership shift confuses custodianship with business accountability for meanings and quality expectations.
• Activity-only tracking repeats the current weakness because submitted issues and attendance do not show resolution or value.

Question 49

Topic: Document and Content Management

A finance department stores signed supplier contracts in a shared content repository. The documents contain payment terms and personal contact details, and they may be needed as evidence after a contract ends. The repository currently has no consistent metadata or disposal process. Which lifecycle action best fits the need?

Options:

• A. Let each contract owner delete files when no longer needed

• B. Convert the contracts to a reporting data mart

• C. Archive all contracts permanently in low-cost storage

• D. Classify the contracts and assign a retention schedule

Best answer: D

Explanation: Document lifecycle management uses classification and retention rules to control content from creation or capture through storage, use, archival, and disposition. Signed supplier contracts are business records with evidentiary value and sensitive content, so they should be classified, tagged with appropriate metadata, protected according to that classification, and assigned a retention schedule based on legal, regulatory, and business requirements. Retention does not mean keeping everything forever; it defines how long content must be retained and when it can be reviewed, archived, or disposed of defensibly. The key is consistent governance over the document type, not individual preference or storage convenience.

• Reporting data mart fails because contracts are managed as content records, not primarily as analytical facts.
• Permanent archiving fails because indefinite retention can increase cost, risk, and privacy exposure.
• Owner deletion fails because disposition should follow approved retention rules, not ad hoc user decisions.

Question 50

Topic: Data Warehousing and Business Intelligence

The COO and CFO disagree about the reported “active customer” KPI. The sales dashboard counts CRM accounts with an opportunity in the last 12 months, while the finance dashboard counts billed customers with a paid invoice in the last 18 months. Each dashboard also applies different exclusion rules for test accounts. What is the best action to resolve the disagreement?

Options:

• A. Build a golden customer record before changing the KPI

• B. Redesign the dashboards to display both values

• C. Govern a shared KPI definition, lineage, and transformation rules

• D. Adopt the finance source as the single KPI source

Best answer: C

Explanation: Conflicting executive KPIs usually require metric governance, not a dashboard or technology fix. A governed KPI should have an agreed business definition, named ownership, approved source systems, transformation rules, exclusions, calculation logic, and lineage from source to report. In this case, the dashboards differ on population, time window, and exclusion handling, so selecting a single source without agreement would only hide the disagreement. Publishing the governed definition enables consistent decision support and clarifies when different metrics are intentionally separate.

• Single source shortcut fails because choosing finance does not resolve the business definition, time window, or exclusion-rule disagreement.
• Dashboard redesign may make the discrepancy visible, but it does not establish a trusted KPI.
• Golden customer record helps with master data identity, but it does not define “active” or standardize calculation rules.

Questions 51-75

Question 51

Topic: Data Governance

A data steward is investigating why the enterprise customer dashboard shows a different “active customer” count than the Finance monthly report. Sales defines active as “any purchase in the past 12 months,” while Finance defines it as “open account with billable activity in the current quarter.” The metric is used in executive reporting, no approved enterprise definition exists, and both business units claim ownership. Which action is the best professional decision?

Options:

• A. Have the steward select Finance’s definition for consistency

• B. Let each unit keep its definition with separate report labels

• C. Ask IT to change the dashboard to the Sales definition

• D. Escalate for an enterprise definition decision and standard

Best answer: D

Explanation: Governance decision rights determine whether an issue is resolved by a steward, delegated to technical custodians, standardized through agreed rules, or escalated to an accountable governance body. Here, the issue is not a simple data defect or implementation task. It is a business-definition conflict across two units, affects executive reporting, and lacks an approved enterprise standard. The steward should prepare the facts, impacts, and candidate definitions, then escalate to the appropriate data owner, data council, or governance forum for a decision. After approval, the definition can be documented in the business glossary and applied consistently in reports. Technical changes come after the business decision, not before it.

• Technical shortcut fails because IT should not choose the business meaning of an enterprise metric.
• Local definitions may be useful for unit-specific reporting, but they do not resolve the enterprise reporting conflict.
• Steward-only decision exceeds stewardship authority when ownership is disputed and the impact is enterprise-wide.

Question 52

Topic: Data Integration and Interoperability

Three regional policy systems must send customer and policy status data to a new claims analytics service. The business needs comparable status reporting within 6 months. The source systems cannot be replaced, status values such as “active” and “closed” have different meanings by region, and auditors need traceability of transformations. Which decision best addresses the interoperability problem?

Options:

• A. Build separate point-to-point transformations for each source feed.

• B. Rename all feed fields to match the analytics service.

• C. Implement a governed canonical exchange standard with shared definitions and code mappings.

• D. Store raw source extracts and let analysts interpret status values.

Best answer: C

Explanation: Semantic interoperability requires systems to exchange data with a shared understanding of meaning, not just compatible file formats or field names. In this situation, the main risk is inconsistent interpretation of status values across regions. A governed canonical exchange standard, supported by business definitions and approved code mappings, gives each source a traceable way to translate local meanings into common enterprise meanings. It also fits the constraint that existing systems cannot be replaced. Point-to-point transformations may move data, but they are harder to govern consistently and audit across multiple feeds. The key is to standardize meaning and transformation rules, not merely centralize or rename data.

• Point-to-point logic may connect the systems, but it risks inconsistent mapping rules and weak reuse across regions.
• Raw extracts preserve source detail, but they shift semantic interpretation to analysts and do not create comparable reporting.
• Field renaming improves superficial consistency, but it does not resolve different business meanings for the same status values.

Question 53

Topic: Data Architecture

A bank is launching a digital lending strategy. Executives want a shared view that links the strategy to the lending capabilities, origination and approval processes, supporting applications, key data stores, and integration patterns between systems. Which enterprise data architecture artifact best fits this need?

Options:

• A. Dashboard design guide for lending reports

• B. Physical database indexing standard

• C. Loan application data quality rulebook

• D. Enterprise data architecture roadmap and capability map

Best answer: D

Explanation: Enterprise data architecture provides the organizing view that aligns data assets and flows with business direction. For the digital lending initiative, the useful artifact must show how business strategy depends on capabilities, how those capabilities are executed through processes, which applications and data stores support them, and how data moves across the environment. A roadmap and capability-oriented architecture view is appropriate because it supports planning, gap analysis, prioritization, and communication across business and technology stakeholders. More detailed artifacts may be needed later, but they do not provide the cross-enterprise alignment requested.

• Indexing standard focuses on physical database performance and implementation consistency, not strategic alignment across capabilities and systems.
• Quality rulebook helps define and monitor valid lending data, but it does not map strategy to applications, stores, and integrations.
• Dashboard guide supports BI presentation consistency, but it does not describe the enterprise data landscape behind the lending capability.

Question 54

Topic: Data Quality

A data quality dashboard shows a rising duplicate-customer rate that is distorting compliance reports. Operations can merge records, but Sales and Compliance disagree on the business rule for when two customer records represent the same party. What response best integrates the issue with data quality governance?

Options:

• A. Ask report developers to filter duplicate records from compliance reports

• B. Have operations merge the duplicates and close the dashboard exception

• C. Log the issue with steward ownership, metric targets, remediation status, and governance escalation

• D. Restrict access to customer records until the duplicate rate improves

Best answer: C

Explanation: Data quality governance integrates issue management with accountability, measurement, and decision rights. A duplicate-customer problem is not only an operational clean-up task when business areas disagree on the matching rule. The issue should be logged, assigned to the appropriate data steward or data owner, measured with agreed quality rules and targets, tracked through remediation, and escalated to the governance body when cross-functional policy or definition conflicts cannot be resolved locally. This creates a controlled path from detection to resolution and prevents hidden fixes that mask the root cause. Technical correction is useful, but governance is needed to approve the rule and sustain quality over time.

• Operational cleanup only fails because merging records without an agreed matching rule can repeat or worsen the defect.
• Report filtering hides symptoms in one output but does not resolve the governed customer definition or source data issue.
• Access restriction treats the problem like a security exposure, not a quality issue requiring ownership, metrics, and remediation tracking.

Question 55

Topic: Data Warehousing and Business Intelligence

A retailer has an enterprise data warehouse with conformed customer, product, and sales history. The marketing department wants a governed, marketing-focused store containing only campaign, customer segment, and promotion performance data for self-service analysis and recurring dashboards. Which architecture component best fits this need?

Options:

• A. Operational data store

• B. BI delivery layer

• C. Data mart

• D. Staging area

Best answer: C

Explanation: A data mart is used to package analytical data for a specific business area, subject, or community while retaining governed definitions and integration with enterprise data where appropriate. In this scenario, marketing needs a curated subset of historical, conformed warehouse data for analysis and dashboards. That aligns with a dependent data mart or similar subject-focused analytical store. A staging area is temporary and supports extraction, cleansing, and loading. An operational data store supports integrated, current or near-current operational reporting. A BI delivery layer presents and distributes information through reports, dashboards, and semantic access, but it is not itself the focused analytical data store being requested.

• Staging area is temporary processing space for load preparation, not a governed departmental analytical store.
• Operational data store emphasizes current operational integration, not historical marketing analysis from warehouse data.
• BI delivery layer supports presentation and access, but the request is for the curated analytical store behind that delivery.

Question 56

Topic: Master and Reference Data Management

A company is implementing customer master data management. The hub detects likely duplicate customer records, combines confirmed duplicates into one golden record, and routes low-confidence cases to a data steward. For confirmed duplicates, it must decide whether the email address from CRM or billing should be kept in the golden record.

Which MDM activity specifically addresses that attribute-level decision?

Options:

• A. Hierarchy management

• B. Stewardship review

• C. Matching

• D. Survivorship

Best answer: D

Explanation: In master data management, matching identifies records that may represent the same real-world entity, and merging consolidates confirmed duplicates. Survivorship is the rule-driven or steward-assisted process that decides which attribute values survive into the golden record when sources conflict, such as choosing the preferred email address from CRM or billing. Stewardship review is used when automated rules cannot confidently decide, and hierarchy management maintains relationships such as customer-to-household or product-to-category structures.

The key distinction is that survivorship works at the attribute-value selection point after duplicate candidates have been identified and consolidated.

• Matching is about finding likely duplicates, not choosing which conflicting attribute value to keep.
• Hierarchy management manages parent-child or grouping relationships, not source-value preference for a golden record.
• Stewardship review handles exceptions or low-confidence decisions, but the described attribute-selection rule is survivorship.

Question 57

Topic: Metadata Management

A data catalog has been populated by automated scanners that captured table names, column names, data types, and technical lineage. Business users still cannot tell what key data elements mean, who is accountable for them, whether they may be used for marketing, or which quality issues affect reports. Which action best addresses the gap?

Options:

• A. Restrict catalog access to technical teams

• B. Ask database administrators to rename unclear columns

• C. Increase scanner frequency for all source systems

• D. Create a stewardship workflow to add business metadata

Best answer: D

Explanation: Automated metadata capture is effective for technical metadata such as schemas, data types, and lineage, but it usually cannot determine business meaning, accountability, acceptable use, or known data quality caveats. Those elements require metadata governance and stewardship. Data stewards, data owners, and subject matter experts should curate glossary definitions, ownership assignments, classifications, usage guidance, and quality notes through an approved workflow. This makes the catalog useful for business decision making and controlled data use, not just technical discovery.

• More scanning may refresh technical metadata, but it will not supply human business context or accountable ownership.
• Column renaming can improve clarity in a database, but it does not govern definitions, acceptable use, or quality context across assets.
• Technical-only access reduces the catalog’s value because business users need governed metadata to interpret and use data responsibly.

Question 58

Topic: Metadata Management

A retailer has conflicting sales and finance reports because teams use different meanings for “active customer.” The chief data steward wants a business-approved definition that nontechnical users can understand, including common synonyms, ownership, and business rules. The organization is early in metadata management and is not yet ready to redesign source systems. What artifact should the steward establish first?

Options:

• A. Data dictionary record

• B. Lineage map for customer reporting

• C. Catalog entry for the customer table

• D. Business glossary entry

Best answer: D

Explanation: A business glossary is the best fit when the main problem is inconsistent business meaning across stakeholders. It defines business terms, approved definitions, synonyms, accountable owners or stewards, and business rules in plain language. That addresses the retailer’s immediate need to align sales and finance before changing technical structures. A data dictionary is more technical, usually describing fields, data types, formats, and constraints. A catalog entry helps users find and understand data assets, often linking to glossary terms, classifications, and metadata. A lineage map shows how data moves and transforms across systems. The key distinction is that the glossary governs the meaning of the term, while the other artifacts describe technical assets, discovery metadata, or data flow.

• Data dictionary is tempting because it documents data, but it focuses on technical elements such as columns, data types, and constraints.
• Catalog entry helps with asset discovery, but the need is to agree on the business meaning of “active customer.”
• Lineage map would help trace report differences later, but it does not establish the approved business definition.

Question 59

Topic: Data Quality

A retailer’s customer analytics team spends each month correcting invalid email addresses and duplicate customer records before campaign reporting. The business goal is to improve trust in segmentation within two quarters, the CRM system is the source for new customer capture, and stewardship resources are limited. What is the best professional decision for sustainable improvement?

Options:

• A. Analyze source-process causes and remediate capture controls

• B. Archive duplicate records after campaign reporting is complete

• C. Increase monthly cleanup staff for reporting cycles

• D. Create a dashboard that flags invalid emails after extraction

Best answer: A

Explanation: Root-cause analysis is central to sustainable data quality improvement. In this situation, the recurring defects appear before reporting, at customer capture in the CRM source system. Repeated downstream cleanup may protect one reporting cycle, but it does not stop invalid emails or duplicate records from being created again. With limited stewardship capacity and a two-quarter trust goal, the better action is to identify process, rule, training, validation, or ownership causes at the point of origin and implement preventive controls. Downstream detection can still be useful for monitoring, but it should support remediation rather than become the permanent quality process.

• More cleanup capacity treats symptoms and consumes stewardship effort without reducing future defect creation.
• Post-extraction flagging improves visibility but still allows poor-quality data to enter and persist in the source.
• Late archiving reduces visible duplicates after reporting, but it does not address capture rules, matching practices, or ownership.

Question 60

Topic: Data Governance

Sales, Finance, and Customer Support use different meanings for “active customer,” producing conflicting executive reports. Data stewards have documented each definition and its impact, but the term is enterprise-wide and no single business unit has authority to decide. Which governance role or body should resolve the definition conflict?

Options:

• A. Enterprise data governance council

• B. Department data steward

• C. Report developer

• D. Data custodian

Best answer: A

Explanation: Data governance operating models assign decision rights based on scope and accountability. A business glossary conflict that affects multiple business units and executive reporting is not mainly a technical or report-formatting problem. Stewards should document candidate definitions, impacts, and usage, but an enterprise-level governance body is typically needed to arbitrate and approve a shared definition when no single domain or department has decision authority. The key distinction is decision authority: stewardship supports and facilitates, while the governance council resolves cross-functional standards, policies, priorities, and escalated conflicts.

• Technical custody fails because a custodian manages systems and controls, not enterprise business meaning.
• Report development fails because changing a dashboard does not establish an approved shared definition.
• Department stewardship fails because a single department steward cannot impose an enterprise-wide definition across business units.

Question 61

Topic: Data Security

A marketing analytics team needs customer data to study campaign response by age band and region. The source table includes email addresses, phone numbers, full dates of birth, and government identifiers. The team does not contact customers, and the data governance policy requires least-privilege access for sensitive attributes. What is the best professional decision?

Options:

• A. Grant full table access after the team signs a confidentiality reminder

• B. Deny all access until a new enterprise privacy policy is written

• C. Copy the source table to a separate analytics database without identifiers removed

• D. Provide a masked, minimized dataset approved for the stated analysis

Best answer: D

Explanation: Privacy protection and security governance should apply least privilege and data minimization. The marketing team needs analytical attributes such as age band and region, not direct identifiers or full sensitive values. A governed dataset can replace full dates of birth with age bands, remove or mask contact details and government identifiers, and provide access only to approved users for the stated purpose. This preserves business value while reducing privacy and security risk.

A confidentiality reminder may support awareness, but it does not replace appropriate access controls or masking. The key takeaway is to enable legitimate use with only the data necessary for that use.

• Full table access exposes direct identifiers that are not needed for the analysis.
• Denying all access blocks a legitimate business need when a controlled subset can meet it.
• Separate database copy changes location but does not reduce sensitivity or enforce minimization.

Question 62

Topic: Master and Reference Data Management

An insurer maintains a claim-status code set used by claims processing, customer portals, and BI reports. Application teams sometimes add or retire codes in local copies without review, creating duplicate statuses, obsolete values, and inconsistent downstream reporting. Which lifecycle control best addresses the problem?

Options:

• A. Periodic profiling of downstream tables for invalid status values

• B. Governed change workflow with approval, versioning, and effective dates

• C. Business glossary definition for the claim-status concept

• D. Source-to-target mapping for each application interface

Best answer: B

Explanation: Reference data lifecycle management should control how code sets are requested, approved, changed, retired, versioned, and distributed. In this scenario, the core failure is not just poor detection or unclear wording; teams are independently changing local copies. A governed change workflow with stewardship approval, version control, effective dates, and publication rules prevents unauthorized additions, manages obsolete values, and gives downstream systems a consistent approved set to use. Profiling can reveal defects after they occur, and mappings can help interfaces transform values, but neither is the primary lifecycle control for controlling changes to the reference data itself.

• Profiling after use can identify invalid or duplicate values, but it does not prevent unapproved code-set changes.
• Glossary definition improves shared understanding, but it does not manage approvals, retirement, versioning, or distribution.
• Interface mapping supports integration, but it does not govern the lifecycle of the approved reference values.

Question 63

Topic: Data Modelling and Design

A bank maintains a central model repository, but project teams often upload logical data models after development is complete. Several models define Customer differently, lineage links to source systems are missing, and analysts cannot assess which reports would be affected by a proposed definition change. The bank wants better reuse and impact analysis without slowing active projects unnecessarily. What is the best professional decision?

Options:

• A. Ask each project team to keep its own model inventory

• B. Create a new reporting data mart for customer analysis

• C. Establish model repository governance with required ownership, status, definitions, lineage links, and change review

• D. Add more physical database details to every model

Best answer: C

Explanation: A model repository supports reuse and impact analysis only when its contents are governed, not merely stored. The facts point to weak control over model ownership, approved definitions, lifecycle status, lineage, and change review. Requiring these elements in the repository gives teams a shared source for finding reusable models, understanding where data comes from, and assessing downstream effects when a definition changes. The governance approach can be scaled to avoid unnecessary delay, for example by requiring minimum metadata and review at defined model lifecycle checkpoints. More technical detail or another data mart may help other problems, but it does not fix uncontrolled definitions and missing lineage in the repository.

• Local inventories would fragment knowledge further and reduce enterprise reuse and impact analysis.
• Physical details may support implementation, but they do not establish approved definitions, ownership, or lineage control.
• New analytical storage might serve a reporting need, but it bypasses the repository governance weakness causing the issue.

Question 64

Topic: Metadata Management

A company is implementing a metadata catalog to support impact analysis for regulatory reporting. The modelling tool contains logical and physical data structures, the integration platform contains source-to-target mappings and lineage, the BI platform contains report metrics, and the governance portal contains glossary terms, owners, and classifications. Which approach best supports repository integration?

Options:

• A. Recreate all modelling, lineage, and glossary metadata directly in the catalog

• B. Use the BI platform as the single system of record for all metadata

• C. Manually upload quarterly spreadsheets from each tool into the catalog

• D. Define authoritative metadata sources and automate controlled exchange with the catalog

Best answer: D

Explanation: Metadata repository integration should connect specialized tools while respecting where different metadata is created and governed. In this scenario, data models, integration mappings, BI metrics, and governance definitions each originate in different repositories. A catalog can provide discovery, relationship navigation, and impact analysis, but it should not replace every source tool. The stronger pattern is to define authoritative sources for each metadata type, agree on synchronization rules, and use automated interfaces or harvesting to keep repository content aligned. This supports lineage, glossary linkage, stewardship, classifications, and report impact analysis without uncontrolled duplication. Manual uploads and wholesale recreation increase staleness and inconsistency.

• Quarterly spreadsheets create stale metadata and make impact analysis unreliable when models, mappings, or reports change.
• BI as record source overstates the role of reporting metadata and would not govern modelling structures, integration lineage, or glossary ownership.
• Recreating metadata duplicates work and creates competing versions instead of integrating repositories with clear ownership.

Question 65

Topic: Data Quality

A customer data quality rule requires every active customer to have a valid tax identifier. A one-time cleansing effort corrected most existing records, but the same validation failures reappear each week in newly created customer records. Which quality improvement best addresses the issue?

Options:

• A. Update the data quality scorecard threshold

• B. Schedule a larger monthly data cleansing cycle

• C. Add validation to the customer creation process

• D. Create a duplicate-customer matching rule

Best answer: C

Explanation: Recurring failures after cleanup indicate that remediation has corrected symptoms but not the source of the defect. In data quality management, cleansing can improve current records, but root cause remediation should prevent the same error from being introduced again. A validation control in the customer creation process is a preventive improvement because it stops records that violate the tax identifier rule before they become operational data. Monitoring and scorecards remain useful, but they do not by themselves change the process that creates bad data. The key distinction is corrective cleanup versus preventive process control.

• More cleansing may improve existing records again, but it leaves the weekly creation of invalid records unchanged.
• Scorecard adjustment changes measurement or tolerance, not the underlying defect source.
• Duplicate matching addresses entity resolution, but the stated rule failure is a missing or invalid tax identifier.

Question 66

Topic: Metadata Management

A retailer maintains separate metadata repositories for its data catalog, ETL mappings, BI reports, and business glossary. The same field is called Customer Status in the glossary, Acct_Status in integration mappings, and Client Type in several reports, with no shared identifiers or synchronization. What is the primary data-management risk in this environment?

Options:

• A. Storage capacity planning will become unreliable

• B. Reference data codes will stop being valid

• C. Transaction tables will lose referential integrity

• D. Users may misinterpret lineage and business meaning

Best answer: D

Explanation: Metadata repository integration reduces the risk that different tools describe the same data asset in incompatible ways. When glossary terms, ETL mappings, catalog entries, and BI report definitions are disconnected, users cannot reliably connect technical metadata to business metadata or trace how a data element is transformed and used. The visible symptom is not just naming variation; it is the absence of shared identifiers and synchronization across repositories. That creates inconsistent meaning, weak lineage, and lower trust in reports and impact analysis. The core issue is fragmented metadata architecture, not a direct failure of database constraints, code-set validity, or infrastructure planning.

• Referential integrity is a physical or logical data constraint issue, not the main risk from unsynchronized metadata repositories.
• Reference code validity concerns governed code sets; the scenario shows conflicting metadata names and meanings, not expired or invalid codes.
• Capacity planning depends on storage and workload metrics, which are not affected directly by glossary and lineage fragmentation.

Question 67

Topic: Data Security

A health insurer shares customer data with analytics teams. Encryption, backups, and role-based access are already in place. The data governance council now needs evidence of who used sensitive data, when it was used, and whether the use matched an approved business purpose. Which security practice best addresses this gap?

Options:

• A. Nightly backup and recovery testing

• B. Audit logging with purpose-of-use review

• C. Standardized reference data codes

• D. Database encryption at rest

Best answer: B

Explanation: Privacy protection and security governance require more than preventing unauthorized access. When the gap is evidence of who used sensitive data and whether that use was appropriate, the relevant practice is auditable monitoring of access and use. Logs, retained and reviewed under governance procedures, support accountability, investigations, compliance evidence, and responsible data use. Encryption protects confidentiality, and backups support availability, but neither shows whether a legitimate user’s activity matched an approved purpose. The key distinction is between preventing or recovering from security events and producing reliable evidence of data use.

• Encryption at rest protects stored data from unauthorized disclosure, but it does not show whether approved users used data appropriately.
• Backup testing supports availability and recovery, but it does not provide user-level accountability for sensitive data access.
• Reference codes improve consistency of coded values, but they are not a security governance control for monitoring data use.

Question 68

Topic: Data Ethics

A bank wants to combine customer service transcripts, account activity, and demographic data to identify customers who may be under financial stress and target them with premium credit offers. The data was originally collected for service and account administration. Which practice best supports ethical data use and accountability before launching the initiative?

Options:

• A. Limit access to the analytics team

• B. Measure only model accuracy before deployment

• C. Conduct an ethical data impact assessment

• D. Proceed because the data is already internal

Best answer: C

Explanation: Ethical data management requires more than permission to access data. A responsible review should assess whether the new purpose is compatible with the original collection purpose, whether customers were informed or gave appropriate consent, whether the analysis could create unfair outcomes, and whether targeted offers could cause unintended harm. It should also define accountability for decisions, transparency to affected customers, and controls for acceptable use. Security and analytics quality matter, but they do not replace an ethical assessment of purpose, fairness, consent, and impact.

• Internal data assumption fails because internal availability does not automatically make a new use ethical or consistent with consent.
• Access restriction only reduces exposure but does not address fairness, transparency, purpose limitation, or harm.
• Accuracy focus only misses ethical risks such as discriminatory targeting, manipulation, and incompatible secondary use.

Question 69

Topic: Data Modelling and Design

A data modeler has produced a normalized logical data model that follows enterprise standards and resolves known structural issues. During validation, several business data owners say they do not understand the model and are reluctant to approve it. What is the best next response?

Options:

• A. Ask database administrators to approve the model instead

• B. Proceed because the model is technically correct

• C. Translate the model into business terms and review examples with the owners

• D. Replace the logical model with a physical database design

Best answer: C

Explanation: Model communication and stakeholder validation are not satisfied by technical correctness alone. A logical data model must be understandable to the business people accountable for the meaning and use of the data. The modeler should use business terminology, examples, definitions, relationship narratives, and possibly simplified views to help owners validate whether the model reflects business reality. Approval from technical teams can confirm implementability or standards compliance, but it cannot replace business acceptance of definitions, entities, relationships, and rules. The key distinction is between a technically sound model and a validated business representation.

• Technical correctness only fails because standards compliance does not prove that business meaning has been understood and accepted.
• Technical approval substitute fails because database administrators are custodians or implementers, not the business owners of meaning.
• Physical redesign fails because changing design level does not address the communication and validation problem.

Question 70

Topic: Data Modelling and Design

During validation of a logical data model for order fulfillment, operations stakeholders say a single corporate order can be split across several ship-to sites. The business goal is accurate fulfillment reporting by destination. The model is still logical, before physical design, but the current CRM stores only one ship-to address per order. What is the BEST professional decision?

Options:

• A. Keep the model unchanged because the CRM supports only one address.

• B. Revise the relationship cardinality and note the CRM constraint separately.

• C. Defer the feedback as outside the model’s fulfillment scope.

• D. Create a definition-gap issue for the term ship-to address.

Best answer: B

Explanation: Stakeholder validation should test whether the model represents real business rules, not merely the current system design. Here, stakeholders are not disputing the meaning of ship-to address; they are saying the relationship between order and destination is wrong for corporate orders. Because the model is still logical, the model should represent the required cardinality, such as one order having multiple ship-to sites when split fulfillment occurs. The CRM limitation should be captured as an implementation constraint or integration issue for later design work. Treating the current CRM structure as the business rule would preserve a known reporting defect.

• CRM-first modeling fails because a logical model should capture the business relationship before accepting a physical system limitation.
• Definition gap is not the main issue because the feedback concerns how many destinations an order may have, not what the term means.
• Scope deferral fails because fulfillment reporting by destination is an explicit business goal.

Question 71

Topic: Data Integration and Interoperability

A retailer integrates daily sales data from point-of-sale systems into a staging area, then loads a warehouse fact table used by finance reports. Finance requires row counts and sales totals to agree from the source extract through reporting. Where should reconciliation controls be placed?

Options:

• A. At each major data movement handoff

• B. Only between the operational source systems

• C. Only during the initial warehouse migration

• D. Only in the final finance reports

Best answer: A

Explanation: Reconciliation in data integration checks that data remains complete and consistent as it moves between environments and uses. When integrated data must agree across source, staging, target, and reports, controls should be applied at the key handoffs: source-to-extract, extract-to-staging, staging-to-target, and target-to-reporting. Typical checks include record counts, control totals, balancing rules, rejected-record logs, and exception review. Limiting reconciliation to one layer can hide defects introduced earlier or later in the flow. The key takeaway is that reconciliation follows the movement of data, not just the final consumer view.

• Final reports only can detect visible reporting differences but may not identify where data was lost or transformed incorrectly.
• Source-only checks confirm operational agreement but do not validate extraction, staging, warehouse loading, or reporting transformations.
• Initial migration only treats reconciliation as a one-time activity, but recurring data movement needs recurring controls.

Question 72

Topic: Data Integration and Interoperability

A retailer is redesigning order-status integration for customer service and mobile self-service. Agents need updates within seconds when an order is packed, shipped, or delayed. Source application teams will not allow direct database reads, and the status mapping must be reused by several downstream consumers with retry and audit capability. Which approach is the best professional decision?

Options:

• A. Manual daily CSV exports from each source team

• B. Direct read-only database access for each consumer

• C. Managed event-based integration with reusable canonical mappings

• D. Nightly ETL from source databases into the warehouse

Best answer: C

Explanation: The visible requirements point to an integration pattern for operational interoperability, not just batch reporting. Near-real-time status updates favor event-based or message-oriented movement. A managed integration layer can preserve source application control by avoiding direct database reads, apply reusable canonical mappings, and provide retry, monitoring, and audit features. This also supports multiple consumers without each one creating its own point-to-point transformation logic. Batch ETL is useful for analytical consolidation, but it would not meet the timeliness requirement for customer service interactions.

• Nightly ETL is too slow for updates needed within seconds and is oriented more toward analytical loading.
• Direct database reads undermine source control and create tight coupling between consumers and operational schemas.
• Manual CSV exports lack timeliness, reliability controls, and reusable governed transformation logic.

Question 73

Topic: Data Storage and Operations

A retailer runs an operational order database that must remain available for customer service during a storage-system failure. The business can tolerate only a few seconds of data loss, and the standby environment must be able to take over quickly. Which lifecycle action best fits this need?

Options:

• A. Configure near-real-time replication to a standby system

• B. Create nightly backups of the order database

• C. Define a seven-year retention rule for order records

• D. Move closed orders to a long-term archive

Best answer: A

Explanation: Replication is the best fit when operational data must remain both durable and immediately usable after an infrastructure failure. It maintains a current or near-current copy of data on another system so operations can fail over quickly. Backups protect against loss and support restore to a point in time, but they usually require recovery steps before the data is usable. Archival moves data for long-term preservation or lower-cost storage, often for inactive data. Retention defines how long data must be kept or disposed of; it does not by itself create an available standby copy. The key distinction is operational continuity versus preservation or later restoration.

• Nightly backups reduce loss risk, but restoring them may take time and may lose changes made after the last backup.
• Long-term archive fits inactive historical records, not rapid use during a storage-system failure.
• Retention rule governs how long records are kept, not how to maintain a usable failover copy.

Question 74

Topic: Document and Content Management

A regulated insurer stores policy documents in an enterprise content repository. The legal team needs a control that ensures policy documents are kept for 10 years, prevents disposal during active litigation, and provides evidence that disposition followed approved rules. Which content management control best addresses this need?

Options:

• A. Check-in/check-out version control

• B. Full-text indexing and metadata tagging

• C. Records retention schedule with legal hold and disposition controls

• D. Digital signatures on approved documents

Best answer: C

Explanation: Records management controls support retention and legal defensibility across the document lifecycle. A retention schedule defines how long records must be kept, disposition controls manage approved destruction or transfer, and legal hold suspends normal disposition when litigation or investigation requires preservation. Together, these controls create evidence that content was retained and disposed of according to authorized rules. Search indexing improves findability, version control manages revisions, and digital signatures support authenticity, but none of those controls alone governs retention periods and legally defensible disposition.

• Search controls help users find documents, but they do not determine how long records are kept or when disposal must stop.
• Version control protects revision history, but it does not provide retention scheduling or legal hold management.
• Digital signatures support authenticity and approval evidence, but they do not manage retention periods or disposition rules.

Question 75

Topic: Data Architecture

A retail company is investigating conflicting “ship date” values in operational reports and a warehouse dashboard. The data architecture team is asked to show where the field originates, how it moves through integration jobs, and which transformations occur before it reaches downstream reporting. Which architecture artifact best fits this need?

Options:

• A. Enterprise subject-area model

• B. Data ownership matrix

• C. Data lineage view

• D. Physical database schema

Best answer: C

Explanation: A data lineage view is used when stakeholders need to trace data from its source through movement, transformation, and consumption. In this scenario, the uncertainty is not simply what the field means or who owns it; the key need is to understand how the value changes as it passes from operational systems through integration processes into reporting. Lineage helps identify where defects, timing differences, or transformation rules may create inconsistent results. It also supports impact analysis and auditability across the data lifecycle.

The closest alternative, an ownership matrix, can identify accountability, but it will not show the path and transformations that explain conflicting reported values.

• Enterprise scope is addressed by a subject-area model, but that does not trace a field through systems and transformations.
• Ownership accountability is addressed by an ownership matrix, but it does not explain how values are derived downstream.
• Physical implementation is shown by a database schema, but it is too narrow for cross-system lineage analysis.

Questions 76-100

Question 76

Topic: Data Integration and Interoperability

A regional health network is integrating patient referral data across hospitals, clinics, and an external care-management partner. Each system can send files successfully, but the receiving teams still disagree on what “active patient” means, use different patient identifiers, and apply different code sets for referral status. Which practice best supports interoperability for this initiative?

Options:

• A. Add more dashboard filters for referral reports

• B. Move all data into one operational database

• C. Increase the frequency of file transfers between systems

• D. Create shared semantic standards and governance for exchanged data

Best answer: D

Explanation: Interoperability is more than technical connectivity. In this scenario, the files already move between systems, but the organizations do not share definitions, identifiers, code sets, or decision rights for the exchanged data. A shared semantic and standards approach addresses those gaps by defining common business meanings, accepted data formats, identifier matching rules, controlled vocabularies, interface expectations, and governance for changes or disputes. This allows each system and organization to interpret referral data in the same way. Simply improving transport or reporting does not resolve inconsistent meaning.

• Transfer frequency improves timeliness but does not fix conflicting definitions, identifiers, or code sets.
• Single database may reduce some integration complexity, but it is often unrealistic and does not replace cross-organization standards and governance.
• Dashboard filters may help users slice reports, but they do not create semantic consistency across source systems.

Question 77

Topic: Data Storage and Operations

A records management team must store finalized contract documents for 7 years. Documents are accessed often during the first 90 days, rarely after that, and must remain retrievable for audits. Storage costs are increasing on the current high-performance platform. Which storage operations response best fits the requirement?

Options:

• A. Delete contracts after 90 days of inactivity

• B. Apply lifecycle-based tiering with retention controls

• C. Rely only on daily backup copies

• D. Keep all contracts on high-performance storage

Best answer: B

Explanation: Storage operations should match data placement and handling to business requirements for performance, availability, recoverability, retention, scalability, and usage. Here, the documents have high early access, low later access, and a fixed retention obligation. Lifecycle-based tiering moves less-used content to lower-cost storage while retention controls prevent premature deletion and support audit retrieval. This balances cost with the need to preserve records. Backup alone supports recovery, but it is not a managed retention and access strategy for records that must remain available over time.

• High-performance storage meets access needs but over-serves rarely used records and does not address rising cost.
• Inactivity deletion conflicts with the 7-year retention requirement.
• Daily backups help recover from loss but are not a substitute for governed record storage and retrieval.

Question 78

Topic: Data Warehousing and Business Intelligence

A finance data warehouse feeds a monthly revenue dashboard due for executive review tomorrow. Lineage analysis shows that invalid country values are being introduced in a CRM field, and the same field also feeds customer service exports and a regulatory report. The organization has named data owners and stewards, but country-value quality rules are not yet standardized. What is the best professional decision?

Options:

• A. Exclude records with invalid countries from executive reporting

• B. Patch only the warehouse transformation for the dashboard

• C. Delay all reporting until the CRM application is redesigned

• D. Apply a controlled warehouse fix and initiate governed source remediation

Best answer: D

Explanation: Warehouse remediation can protect a time-sensitive BI deliverable, but it should be controlled, documented, and traceable because it does not correct the originating defect. Since the invalid values are created in CRM and reused by multiple downstream consumers, the durable fix belongs at the source and should be managed through data owners and stewards. Enterprise data quality management should define the country-value standard, assign accountability, monitor recurrence, and coordinate remediation across impacted uses. A short-term warehouse rule may be justified for tomorrow’s dashboard, but only as part of a governed response that addresses lineage, ownership, and reuse risk.

• Warehouse-only patch leaves the CRM defect and other downstream uses exposed.
• Full reporting delay overreacts to the immediate dashboard need when a controlled temporary remediation is possible.
• Record exclusion may distort revenue reporting and hides the quality issue instead of managing it through governance.

Question 79

Topic: Data Warehousing and Business Intelligence

A BI team must certify a sales performance dashboard before the quarterly sales review. The sales governance council has approved the definitions for Revenue, Quota, and Attainment %, and source-to-target lineage has been documented. Users report that the KPI is credible at total sales level but becomes misleading when filtered by product.

Exhibit: KPI design summary

Which management issue best explains the dashboard problem?

Options:

• A. Unclear business definition of revenue

• B. Insufficient dashboard access control

• C. Incompatible grain between revenue and quota

• D. Poor source-to-target lineage documentation

Best answer: C

Explanation: The issue is an incompatible grain in the KPI components. Revenue is captured at order-line level, including product, while quota exists only by salesperson and month. When users apply a product filter, the numerator changes but the denominator has no product-level meaning. This can repeat, allocate, or suppress quota in ways that make Attainment % misleading. Approved definitions and documented lineage reduce ambiguity about what the fields mean and where they came from, but they do not solve a mismatch in the level of detail used for calculation. The data management response should define an appropriate allocation rule, restrict unsupported filters, or redesign the metric grain before certification.

• Revenue definition is not the main issue because the definitions were already approved by governance.
• Lineage documentation is not the main issue because the source-to-target path is already documented.
• Access control does not explain why the KPI changes meaning when filtered by product.

Question 80

Topic: Data Governance

A bank is integrating customer data for an enterprise reporting program. The Retail and Commercial domains cannot agree on the definition of “active customer,” and the disagreement changes regulatory exposure counts. Domain stewards have already reviewed the issue twice without resolution, and the report deadline is approaching. What is the BEST professional response?

Options:

• A. Postpone the report until both domains independently agree

• B. Escalate the issue to the data governance council with impact analysis

• C. Allow the reporting project manager to choose the definition

• D. Use the Retail definition because it has more records

Best answer: B

Explanation: Governance escalation is appropriate when a data issue crosses domain boundaries, has material business or compliance impact, and cannot be resolved at the stewardship level. The next step is not to let a project team or one domain decide unilaterally. A governance body, such as a data governance council or decision forum with accountable data owners, should review the documented issue, business impact, alternatives, and recommendation. This keeps the decision transparent, authoritative, and reusable for future reporting and integration work. The key is to escalate with enough evidence for a decision, not merely to hand off an unclear disagreement.

• Project convenience fails because a project manager usually does not have enterprise accountability for cross-domain business definitions.
• Record count logic fails because the larger data set does not determine the authoritative definition.
• Indefinite postponement fails because unresolved material issues should move through governance decision channels, not wait passively.

Question 81

Topic: Data Security

A retailer is defining access controls for three datasets. Customer profiles contain email, phone number, and date of birth. Pricing strategy is marked internal confidential. A product availability feed has no personal or confidential fields but is essential to online checkout. Which access-control decision best reflects these classifications?

Options:

• A. Apply the same access rule to all three datasets for consistency

• B. Restrict or mask customer profiles, limit pricing access, and protect feed availability

• C. Classify the availability feed as confidential because checkout depends on it

• D. Give analysts full customer profiles because sales analysis has business value

Best answer: B

Explanation: Data classification guides the type and strength of access control. Personal data in customer profiles requires privacy-aware controls such as need-to-know access, masking, or minimization. Internal pricing strategy requires confidentiality controls because disclosure could harm the business. The product availability feed is business-critical, but the stated facts do not make it sensitive or confidential; its main control focus is reliable, authorized operational use, including integrity and availability. Business criticality can increase protection requirements, but it does not automatically mean the data should be treated as private or confidential.

• Business value alone does not justify broad access to personal data; analysis should use minimized or masked data where possible.
• Operational criticality affects availability and integrity needs, but it does not by itself create confidentiality classification.
• Uniform access rules ignore different sensitivity, privacy, confidentiality, and criticality drivers across datasets.

Question 82

Topic: Master and Reference Data Management

A data governance council has approved retiring the reference data code PROD_CAT = LEGACY for future transactions. Historical orders must keep their original value for reporting, and downstream systems must know when to stop accepting the code. Which lifecycle action best fits this need?

Options:

• A. Delete the code from all reference tables immediately.

• B. Audit current code usage without changing published reference data.

• C. Retire the code with an effective end date and publish guidance.

• D. Create a replacement master record with the same identifier.

Best answer: C

Explanation: Reference and master data lifecycle management controls how shared data assets move from creation and approval through publication, use, maintenance, retirement, and audit. In this case, the code has already been approved for retirement, but it still has historical meaning in existing orders. The right lifecycle action is to retire it in a controlled way: set an effective end date, mark it inactive for new transactions, keep it available for historical interpretation, and publish the change to downstream users and systems. Audit records should show who approved the retirement and when it became effective. Immediate deletion would break reporting and lineage.

• Immediate deletion fails because historical orders still need the retired code to remain interpretable.
• Replacement master record confuses master data with a reference-data code-set change.
• Audit only records evidence of use but does not stop future use or notify consumers.

Question 83

Topic: Metadata Management

Analysts can locate datasets in the enterprise catalog, but they often misinterpret fields, cannot see where values came from, do not know whether data is fit for reporting, and are unsure which uses are approved. Which metadata improvement best addresses this gap?

Options:

• A. Publish a physical data model for each source database

• B. Improve catalog search ranking and keyword synonyms

• C. Create a master data hub for shared customer identifiers

• D. Add business definitions, lineage, quality indicators, and usage guidance

Best answer: D

Explanation: Metadata management creates business value when it helps people understand, trust, and use data appropriately. In this scenario, discovery is not the problem because users can already find datasets. The gap is interpretability and governed use: meaning, lineage, quality, and approved usage. Improving the catalog with business definitions, lineage, data quality indicators, classifications, ownership, and usage guidance directly addresses those needs. Search tuning helps with discovery, and a physical model helps technical implementation, but neither gives the full business context needed for safe, consistent use.

• Search tuning fails because the datasets are already discoverable; the missing information is context and trust.
• Master data hub targets consistent shared master data, not general catalog understanding across datasets.
• Physical models describe database structures but usually do not cover approved use, business meaning, lineage, or quality status.

Question 84

Topic: Master and Reference Data Management

A manufacturer uses a shared reference list of product status codes in ERP, CRM, ecommerce, and BI reporting. Marketing wants to add a new status for prelaunch products within two weeks, but Finance warns that status changes affect revenue reporting and historical trend comparisons. The organization has a small data governance council and no enterprise MDM platform yet. What is the best professional decision?

Options:

• A. Let Marketing add the status in CRM first

• B. Submit the code through the reference data change process

• C. Ask each application team to maintain its own status list

• D. Delay the change until an enterprise MDM platform is implemented

Best answer: B

Explanation: Shared reference data needs lifecycle governance because it is reused across business processes and reporting. Adding a product status is not just local application maintenance; it can affect definitions, downstream mappings, report logic, historical comparability, and ownership. The practical response is to use a controlled reference data change process: assess business meaning and impacts, approve the new value, define effective dates and mappings, communicate to consuming systems, and monitor implementation. A full MDM platform can help later, but governance responsibilities exist even with simple tools and a small council.

The key distinction is that local maintenance changes one application, while lifecycle governance controls shared data from proposal through retirement across all consumers.

• CRM-first change fails because it treats a shared code as a local Marketing configuration and risks inconsistent downstream reporting.
• Separate status lists fail because independent maintenance undermines semantic consistency across ERP, CRM, ecommerce, and BI.
• Wait for a platform fails because tooling is not required before applying governance, impact assessment, and approval controls.

Question 85

Topic: Data Management Process

A finance division is launching a data quality improvement effort after several reports used different definitions of “active customer.” Business leaders agree that one definition should be used across reporting, but no group has authority to approve or maintain that definition. What action best fits the need?

Options:

• A. Buy a profiling tool for customer tables

• B. Measure the defect rate in customer records

• C. Assign stewardship accountability for the term

• D. Write a step-by-step report refresh procedure

Best answer: C

Explanation: DAMA-DMBOK process thinking separates different kinds of data management responses. A shared business definition problem is not first a tooling, procedural, or measurement problem. When a term such as “active customer” affects cross-functional reporting and no group has authority to approve or maintain it, the needed action is to establish stewardship accountability. Stewards coordinate business meaning, resolve definition conflicts, maintain glossary content, and escalate decisions through governance when needed.

Technology can support cataloging or profiling, procedures can standardize repeatable tasks, and measurements can monitor quality. None of those creates the business authority needed to define and govern a shared term.

• Tool-first thinking fails because profiling may reveal data patterns but does not decide the approved business definition.
• Procedure focus fails because report refresh steps do not resolve ownership of shared terminology.
• Measurement focus fails because defect rates are useful after criteria are defined, not before accountability exists.

Question 86

Topic: Data Quality

A data quality review finds that the enterprise customer dashboard shows conflicting counts of “active customers.” Root cause analysis shows that two source systems use different meanings: one uses an open contract, and the other uses any login in the last 90 days. The integration job is technically loading both fields as designed. Which remediation action best addresses the defect?

Options:

• A. Ask report users to choose their preferred count

• B. Reload the dashboard data from both source systems

• C. Approve one business definition and update integration rules

• D. Increase the integration job frequency to hourly

Best answer: C

Explanation: When a data quality defect is caused by poor or conflicting definitions, the durable remediation is not a technical reload or faster processing. The organization should resolve the business meaning through governance, assign or involve the appropriate data owner or steward, approve the definition, and then update data quality rules, mappings, and reporting logic to use that definition consistently. In this scenario, the integration job is working as designed, but it is moving semantically inconsistent data into a shared dashboard. The key takeaway is to fix the root cause: unclear business definition and inconsistent interpretation.

• Reloading data does not solve the disagreement over what “active customer” means.
• Faster integration may make inconsistent counts appear sooner, but it does not improve quality.
• User preference shifts the decision to consumers and leaves the shared metric uncontrolled.

Question 87

Topic: Data Warehousing and Business Intelligence

A company’s self-service analytics program has reduced report turnaround time, but teams now publish dashboards with different definitions of “active customer” and “monthly recurring revenue.” Executives are making conflicting decisions from these measures. Which action best preserves analytical speed while reducing unmanaged decision risk?

Options:

• A. Require IT to build every analytical report centrally

• B. Redesign warehouse tables to improve dashboard performance

• C. Establish BI governance for shared metrics and certified datasets

• D. Cleanse source records that fail data quality rules

Best answer: C

Explanation: Self-service BI needs governance when inconsistent measures affect business decisions. The issue is not mainly report speed, source-data defects, or query performance; it is uncontrolled analytical meaning and use. BI governance establishes how important metrics are defined, approved, certified, documented, and stewarded. It can include a business glossary, governed semantic layer, certified datasets, ownership, and an escalation path for metric conflicts. This approach preserves user agility while adding controls for trusted, comparable decision making. A fully centralized reporting model may reduce inconsistency, but it sacrifices the self-service value that the organization wants to keep.

• Centralized report building overcorrects by removing self-service rather than governing the measures that create decision risk.
• Source cleansing helps when records are inaccurate, but the visible problem is inconsistent metric definition and certification.
• Warehouse redesign may improve performance, but faster dashboards can still publish conflicting business measures.

Question 88

Topic: Document and Content Management

A sales operations team stores procedures, pricing notes, and customer approval templates across several content repositories. Analysts are using different versions of the same document to support business decisions, and many files are redundant, outdated, or trivial. Which control best addresses the risk?

Options:

• A. Increase repository storage capacity

• B. Create more dashboard filters

• C. Apply content lifecycle and records controls

• D. Encrypt all documents at rest

Best answer: C

Explanation: Redundant, outdated, trivial, and conflicting documents are a content governance and records-control problem. The appropriate control is to manage content through its lifecycle: assign ownership, capture key metadata such as status and effective date, identify the authoritative version, apply retention and disposition rules, and remove or archive content that should no longer guide decisions. This reduces ambiguity about which document is current and trustworthy.

Security, reporting, and storage controls may be useful in other situations, but they do not resolve conflicting document authority or obsolete content. The key takeaway is to control content as a managed business asset, not just store more files.

• More storage preserves the same redundant and obsolete material rather than controlling its lifecycle.
• Dashboard filters affect report presentation, not document authority, retention, or disposition.
• Encryption at rest protects confidentiality but does not identify current, authoritative, or obsolete documents.

Question 89

Topic: Data Architecture

A global insurer has separate policy, claims, and billing domains. Each domain has valid differences in terminology and process, but duplicate definitions of Customer and Product are causing inconsistent integration designs. Which architecture response best reduces fragmentation while preserving legitimate domain differences?

Options:

• A. Use dashboard metric definitions as the architecture view

• B. Mandate one physical schema for every domain application

• C. Create an enterprise conceptual model with governed domain extensions

• D. Let each domain define core entities independently

Best answer: C

Explanation: Data architecture should provide shared enterprise views and principles without erasing real business-domain variation. An enterprise conceptual model or subject-area model can define common entities, relationships, and naming principles for concepts such as Customer and Product. Domain logical or physical models can then extend those concepts for policy, claims, or billing needs. This reduces semantic fragmentation, improves integration consistency, and avoids forcing every system into the same implementation design. The key distinction is between standardizing shared meaning and over-standardizing local implementation.

• Single physical schema overreaches because applications may have valid process and technology differences that do not need identical storage structures.
• Independent core definitions preserves autonomy but leaves the integration and semantic fragmentation problem unresolved.
• Dashboard metrics support reporting consistency, but they are not a substitute for enterprise data architecture views of core business concepts.

Question 90

Topic: Data Storage and Operations

A retailer’s order database supports online sales. The business states that after an outage, order processing must resume within 4 hours and no more than 15 minutes of committed orders can be lost. The current process performs one full backup each night. Which operational action best fits the stated need?

Options:

• A. Move orders older than 3 years to archive storage

• B. Add disk capacity to the database server

• C. Add transaction log backups and test recovery procedures

• D. Create a new dashboard for backup job status

Best answer: C

Explanation: Recovery operations should be designed around the business impact of downtime and data loss. The 4-hour restart need is a recovery time objective, and the 15-minute loss limit is a recovery point objective. A nightly full backup alone can lose up to a full day of transactions, so it does not meet the stated data-loss tolerance. Adding transaction log or other frequent incremental backups reduces potential data loss, while documented and tested restore procedures confirm that recovery can be completed within the required time. Capacity expansion, archiving, and monitoring may be useful operational practices, but they do not by themselves satisfy the recovery requirement.

• Archiving old orders helps retention and storage management, but it does not reduce data loss after an outage.
• Adding disk capacity may address storage growth, but it does not establish a recoverable backup point every 15 minutes.
• Backup dashboards improve visibility, but monitoring alone does not prove recovery can meet time and data-loss targets.

Question 91

Topic: Data Governance

A data steward finds that Sales and Finance use different definitions of “active customer” in monthly executive reports. The difference changes revenue-retention results, and both departments insist their definition is authoritative. The current business glossary has no approved definition for the term. What should the steward do next?

Options:

• A. Escalate to the governance council for a decision

• B. Adopt the Sales definition for all reports

• C. Ask IT to change the Finance report logic

• D. Record both definitions as valid glossary terms

Best answer: A

Explanation: A stewardship process should resolve routine issues within the steward’s authority, delegate technical work to custodians, and standardize terms when business agreement already exists. Here, the issue crosses business areas, affects executive reporting, and lacks an approved glossary definition. Because both departments claim authority, the steward should not unilaterally choose a definition or send the problem directly to IT. The governance operating model exists to define decision rights for conflicts like this, often through a data governance council, data owner decision, or similar escalation path. Once the decision is made, stewardship can update the glossary, standards, and reporting rules.

• Unilateral selection fails because choosing Sales bypasses agreed decision rights for a cross-functional business term.
• Technical delegation fails because IT can implement report logic but should not decide the business meaning of “active customer.”
• Dual glossary terms fails because documenting both as valid would preserve inconsistent executive reporting rather than standardize the term.

Question 92

Topic: Data Integration and Interoperability

A finance data mart is loaded nightly from two billing systems. Month-end reports must agree with the source systems, the staging tables, the curated data mart, and the BI semantic layer. Where should reconciliation controls be placed to best support this requirement?

Options:

• A. Only after users report a discrepancy

• B. At each major handoff in the data flow

• C. Only before data is extracted from sources

• D. Only in the BI reports used by finance

Best answer: B

Explanation: Reconciliation in data integration checks that data remains complete, accurate, and consistent as it moves between environments and uses. When the requirement names source systems, staging, target structures, and reporting consumption, controls should be placed at the key transfer and transformation points across that flow. This may include record counts, control totals, balancing rules, exception logs, and source-to-target comparisons. Placing checks only at the report layer may confirm a visible mismatch, but it does not show where the defect occurred or prevent downstream propagation. The key point is to reconcile across the integration lifecycle, not only at one endpoint.

• Report-only checking can reveal a final variance, but it misses earlier handoffs and weakens root-cause analysis.
• Source-only checking verifies extraction readiness, but it does not validate staging transformations, target loads, or reporting logic.
• Reactive checking depends on users finding problems, which is not an effective integration control.

Question 93

Topic: Data Warehousing and Business Intelligence

A finance team distrusts a new sales warehouse because monthly revenue totals differ from the billing system and analysts cannot explain how discounts and returns are handled. Which practice best addresses the warehouse data quality issue?

Options:

• A. Increase warehouse refresh frequency and tune the load jobs

• B. Redesign the dashboard layout and add warning notes to revenue charts

• C. Load billing data unchanged and prevent all analyst-created calculations

• D. Define source-to-target rules, reconciliation checks, lineage metadata, and stewardship ownership

Best answer: D

Explanation: Warehouse data quality is not established only by moving data into an analytical platform. The team must understand source-system meaning, document transformation and business rules, reconcile warehouse results to trusted sources, capture metadata and lineage, and assign stewardship for definitions and issue resolution. In this scenario, the unexplained treatment of discounts and returns is a transformation and definition problem, while the mismatch with billing totals requires reconciliation controls. Lineage and stewardship make the result explainable and governable over time. Presentation changes or faster loads may improve usability or timeliness, but they do not prove that revenue is complete, consistently defined, or traceable.

• Dashboard focus fails because chart notes do not resolve source, transformation, or reconciliation defects.
• Raw loading only fails because warehouses often need governed transformations and shared metric definitions, not just unchanged source copies.
• Performance tuning fails because faster refreshes can reproduce the same quality and lineage problems more quickly.

Question 94

Topic: Data Governance

A bank is standardizing customer reporting across marketing, risk, and compliance. Marketing counts any customer with campaign activity in the last 24 months as “active,” while risk counts only customers with an open account. The difference affects an executive dashboard and a regulatory attestation. The operating model says stewards document approved definitions, custodians implement system changes, and enterprise conflicts are escalated for prioritization and decision rights. Which governance role or body should take the lead in resolving the issue?

Options:

• A. Marketing report owner

• B. Data Governance Council

• C. Data custodian for the CRM system

• D. Customer data steward

Best answer: B

Explanation: Data governance operating models assign decision rights based on the type and scope of the issue. A conflicting business definition that affects multiple functions and regulated reporting is not just a documentation task or a system implementation issue. It needs an authorized governance body to resolve the conflict, approve the enterprise definition, and prioritize the resulting changes. Data stewards can prepare analysis and maintain the glossary after approval, and custodians can implement technical changes once the decision is made. A report owner may raise the issue, but should not unilaterally define an enterprise term used by risk and compliance.

• Steward documentation is important, but stewards should not independently decide a contested enterprise definition.
• Technical custody fails because the CRM custodian implements approved changes rather than setting business meaning.
• Single report ownership is too narrow because the term affects multiple functions and a regulatory attestation.

Question 95

Topic: Data Ethics

A retailer collected precise location data for fraud detection and described that purpose in its customer privacy notice. The analytics team now wants to use the same raw location histories to infer income level and send targeted partner offers. Customers have not been told about this new purpose. What is the most appropriate response?

Options:

• A. Proceed because the data was collected by the retailer

• B. Escalate the proposed use for governance review before proceeding

• C. Limit the analysis to recent customers only

• D. Remove customer names and use the raw histories

Best answer: B

Explanation: Ethical data use considers more than technical access or internal ownership. A new use of precise location data to infer income and support partner marketing is materially different from fraud detection. It may affect customer expectations, fairness, transparency, and privacy. Governance or ethics review should assess purpose compatibility, legal and policy obligations, communication or consent needs, data minimization, and appropriate controls before the team proceeds. Simply having collected the data does not make every later use acceptable.

• Internal collection fails because ethical accountability still applies when data is reused for a new purpose.
• Name removal fails because raw location histories can remain sensitive and potentially identifying.
• Recent customers only reduces scope slightly but does not address purpose change, transparency, or oversight.

Question 96

Topic: Data Warehousing and Business Intelligence

A finance team, sales team, and operations team each build monthly management reports from separate spreadsheet extracts taken from operational systems. Executives see different values for revenue and customer counts depending on which report they open. Which warehouse improvement best addresses the inconsistency?

Options:

• A. Create a governed data mart with conformed definitions

• B. Redesign the dashboard colors and layout

• C. Increase extract frequency for each spreadsheet

• D. Allow each team to publish its own metric glossary

Best answer: A

Explanation: Inconsistent BI metrics often result from unmanaged analytical data flows: separate extracts, local transformations, and team-specific definitions. A governed warehouse or data mart layer provides a controlled path from source data to reporting. It can standardize transformations, define measures such as revenue and customer count, and use conformed dimensions so reports aggregate data consistently across business areas. The improvement is architectural and governance-oriented, not mainly a presentation or refresh-speed issue. The key takeaway is to move repeated reporting from isolated extracts into a managed analytical environment with shared definitions and stewardship.

• More frequent extracts may make data fresher, but it does not resolve different calculations or unmanaged transformations.
• Separate glossaries preserve local interpretations instead of creating shared, governed metric definitions.
• Dashboard redesign improves usability, but it does not correct inconsistent data flows or measure logic.

Question 97

Topic: Metadata Management

A sales leader asks the data catalog team to make Net Revenue understandable to nontechnical stakeholders. They need the approved business definition, accountable data owner, permitted use, and current quality status. They do not need table structures, job logs, or code-set values. Which metadata type best addresses the request?

Options:

• A. Business metadata

• B. Operational metadata

• C. Reference data

• D. Technical metadata

Best answer: A

Explanation: Business metadata helps stakeholders understand data in business terms: what a data element means, who is accountable for it, how it may be used, and whether it is fit for purpose. In this scenario, the need is not to inspect physical implementation or processing behavior; it is to make Net Revenue understandable and governable for business users. Technical metadata would be more appropriate for schemas, mappings, lineage details, and transformations. Operational metadata would describe processing events, usage statistics, or job execution. Reference data is managed data content, not metadata.

• Technical detail is too implementation-focused for a request centered on business definition, ownership, and permitted use.
• Operational activity fits job runs, access counts, and processing status rather than shared business meaning.
• Reference data contains controlled values such as codes or classifications, not descriptive context about a metric.

Question 98

Topic: Master and Reference Data Management

A regional sales platform is still using last quarter’s territory codes, while the customer master in the CRM has direct updates to customer ownership that were not reviewed by data stewards. Downstream reports now disagree on sales responsibility. What is the best action?

Options:

• A. Ask each consuming system to correct its local mappings

• B. Implement governed approval and publication for master and reference changes

• C. Rebuild the reporting data mart from the CRM source

• D. Restrict all downstream access until data defects stop

Best answer: B

Explanation: Master and reference data lifecycle management requires controlled creation, approval, publication, and distribution of shared data. Customer ownership in a customer master should not be changed directly without stewardship review and governance approval. Territory codes are reference data, so consuming systems need an authorized, current version through a managed publication or synchronization process. The strongest action combines both controls: steward approval for master data changes and controlled distribution of approved reference data to downstream systems. Local fixes or rebuilding reports may mask symptoms, but they do not prevent recurrence.

• Local corrections create inconsistent copies and bypass the authoritative lifecycle for shared data.
• Data mart rebuild refreshes reporting but does not stop unapproved master changes or stale reference-code use.
• Access restriction is disproportionate and does not establish approval, versioning, or distribution controls.

Question 99

Topic: Data Architecture

A company is replacing several regional customer systems. Executives want customer and product data to be shared across business units for future initiatives, not only for the current migration. Data definitions differ by region, stewardship roles are informal, and integration teams have been building point-to-point mappings. What is the best professional recommendation?

Options:

• A. Prioritize a reporting data mart for executive dashboards

• B. Let each application team define its own customer model

• C. Build temporary point-to-point mappings for each regional system

• D. Create an enterprise data architecture view with standards and stewardship assignments

Best answer: D

Explanation: Durable data sharing depends on architecture artifacts that describe enterprise subject areas, common definitions, standards, integration principles, and stewardship accountability. In this situation, the core problem is not only moving data for a migration; it is inconsistent meaning and fragmented ownership across regions. An enterprise data architecture view gives teams a shared target for customer and product data and creates a basis for reusable integration patterns and metadata. Stewardship assignments are important because standards need accountable business participation, not only technical mapping work. A reporting mart may help analytics, but it does not resolve the enterprise sharing and reuse problem by itself.

• Temporary mappings solve immediate connectivity but reinforce point-to-point complexity and inconsistent meaning.
• Reporting focus may satisfy dashboards but does not establish reusable enterprise data structures or stewardship.
• Local models preserve regional autonomy but increase semantic inconsistency and reduce enterprise reuse.

Question 100

Topic: Data Management Process

A regional insurer is launching a program to reduce claim processing errors, improve executive loss-ratio reporting, and demonstrate better control over customer data used in regulatory submissions. Which data management activity best fits this need?

Options:

• A. Coordinate governed data definitions, quality rules, lineage, and stewardship

• B. Create a marketing campaign using customer segments

• C. Redesign the claims application user interface

• D. Increase server capacity for the reporting database

Best answer: A

Explanation: The data management function supports the organization by making data reliable, understood, protected, and usable across business activities. In this scenario, the same data affects operational claim handling, analytical reporting, regulatory risk, and executive decisions. Coordinating definitions, data quality rules, lineage, and stewardship addresses those needs directly: it clarifies what the data means, improves fitness for use, shows where data comes from, and assigns accountability for resolving issues. Technology and business initiatives may depend on this work, but they do not replace the data management function’s role in governing and improving data as an organizational asset.

• User interface focus may improve workflow usability, but it does not establish trusted definitions, lineage, or accountability for data.
• Server capacity can help performance, but faster reporting does not make the underlying data fit for operational or regulatory use.
• Marketing segmentation uses data for one business purpose, but it does not address cross-functional control, quality, and decision support.

Continue in the web app

Use IT Mastery for interactive DAMA CDMP Data Management Fundamentals practice with mixed sets, timed mocks, topic drills, explanations, and progress tracking.

Try DAMA CDMP Data Management Fundamentals on Web

Focused topic pages

• Data Management Process
• Big Data
• Data Architecture
• Document and Content Management
• Data Ethics
• Data Governance
• Data Integration and Interoperability
• Master and Reference Data Management
• Data Modelling and Design
• Data Quality
• Data Security
• Data Storage and Operations
• Data Warehousing and Business Intelligence
• Metadata Management