CDMP
Chapter: Reference and Master Data
By: Budoor Ahmed
Telegram: @Budoor2654
Email: Budoor2654@[Link]
"This course was crafted during the mysterious hours when even the owls were asking,
'Who's up learning Reference and Master Data now?' "
Outlines
1. Reference and Master Data Introduction
1.1 Business Drivers
1.2 Goals and Principles
1.3 Essential Concepts
2. Reference and Master Data Activities
3. Tools and Techniques
4. Implementation Guidelines
5. Organisation and culture change
6. Reference and Master Data Governance
By: Budoor Ahmed Reference and Master Data
Reference and Master Data- DAMA
By: Budoor Ahmed Reference and Master Data
Introduction
• Reference and Master Data: Crucial data shared across all business areas, processes, and systems.
• Examples: Customer lists, location codes, business units, delivery options, part lists, cost center codes, tax
codes.
• Benefits:
• Organizational Efficiency: Streamlines operations and decision-making.
• Customer Satisfaction: Enhances service quality and responsiveness.
• Goal:
• Promote a unified data environment where reference and master data are consistently maintained and
accessible to all relevant parties.
By: Budoor Ahmed Reference and Master Data
Business Drivers
Business Drivers:
1. Meeting organizational data requirements - Data Consistency: Ensures uniform access to accurate and
current data across the organization.
2. Quality Control: Mitigates errors and decision-making risks by maintaining high data quality.
3. Integration Efficiency: Lowers costs and complexities in merging new data sources.
4. Risk Management: Simplifies data architecture to reduce operational and compliance risks.
• Impact:
• Critical for success in sales growth, cost reduction, and compliance efforts.
• Enhances IT efficiency, decision confidence, and customer satisfaction.
By: Budoor Ahmed Reference and Master Data
Goals and Principles
• Goals of Reference & Master Data Management (MDM):
• Data Integrity: Ensure complete, consistent, and current Master and Reference Data across the organization.
• Data Sharing: Enable cross-enterprise sharing of Master and Reference Data.
• Efficiency and Simplification: Reduce costs and complexity in data usage and integration.
• Guiding Principles
• Shared Data: Managed for organization-wide accessibility.
• Ownership: Owned by the organization, not confined to specific departments.
• Quality and Stewardship: Requires continuous Data Quality monitoring and governance by Business Data Stewards.
• Controlled Change:
• Master Data changes must be accurate, current, and reversible.
• Reference Data alterations follow a strict, approved process.
• Authority: Master Data replicated only from the system of record to ensure accuracy.
• Establish a framework for managing Master and Reference Data that supports organizational efficiency, data
quality, and shared data principles.
By: Budoor Ahmed Reference and Master Data
Essential Concepts
Differences Between Master and Reference Data
• Master Data:
• Definition: Aggregation of Reference, enterprise structure, and transaction structure data.
• Examples: Customer identifiers, products, vendors.
• Purpose: Provides context for business transactions, ensuring a single version of truth for key business entities.
• Management Challenges: Entity resolution, maintaining consistency across systems.
• Reference Data:
• Definition: Data used to classify or characterize other data, often less volatile and simpler than Master Data.
• Examples: Code tables, status codes, classification schemes.
• Purpose: Provides context for both transactional and Master Data, enabling meaningful data interpretation.
• Management Focus: Ensuring complete, accurate, and current domain values.
By: Budoor Ahmed Reference and Master Data
Essential Concepts
Differences Between Master and Reference Data
• Shared Goals:
• Enable meaningful use of transactional data by providing critical context.
• Managed at the enterprise level to ensure consistency and reduce ambiguity.
• Key Management Principles:
• MDM (Master Data Management): Ensures consistent use of accurate and timely business entity data.
• RDM (Reference Data Management): Manages domain values and definitions for organizational clarity.
By: Budoor Ahmed Reference and Master Data
Essential Concepts
Reference Data
• What Is Reference Data?
• Data used to classify or relate other data within or beyond an organization.
• Types of Reference Data:
• Simple Lists: Pair code values with descriptions (e.g., Country Codes).
• Cross-Reference Lists: Translate between different code sets representing the same concept.
• Taxonomies: Capture information at varying levels of specificity (e.g., ZIP codes within towns and states).
• Ontologies: Manage complex taxonomies and content classification for web content.
• Management Goals:
• Ensure consistency and currentness across functions.
• Make data accessible and usable organization-wide.
• Challenges:
• Ownership: Identifying responsible parties for external or cross-departmental data.
• Complexity: Balancing detail with usability to meet diverse user needs.
• Importance of Metadata:
• Tracks source, version, and updates to ensure lineage and currency.
By: Budoor Ahmed Reference and Master Data
Essential Concepts
Reference Data
By: Budoor Ahmed Reference and Master Data
Essential Concepts
Master Data
• Master Data: Authoritative data about key business entities such as customers, products, employees, and
locations.
• Importance:
• Provides context for transactions and analysis.
• Ensures authoritative, accurate data for business operations.
• Systems of Record and Reference:
• System of Record (SoR): Original source of data creation and maintenance.
• System of Reference: Authoritative source for data consumption.
• Trusted Source and Golden Record:
• Trusted Source: Best version of truth for Master Data.
• Golden Record: Consolidated, accurate record of an entity.
By: Budoor Ahmed Reference and Master Data
Essential Concepts
Master Data
• Management Challenges:
• Entity resolution: Identifying and managing data across systems.
• Data stewardship: Maintaining data quality and integrity.
• Goals:
• Ensure consistent, accurate, and timely data across systems.
• Support efficient business operations and decision-making.
• Master Data Domains:
• Parties: Individuals and organizations.
• Products and Services: Offerings managed across lifecycle.
• Financial Structures: Accounts, cost centers, budgets.
• Locations: Geographic data and addresses.
By: Budoor Ahmed Reference and Master Data
Essential Concepts
Master Data
• Strategy:
• Start with specific data domains.
• Implement governance and quality controls.
• Utilize technology for data integration and consistency.
By: Budoor Ahmed Reference and Master Data
Essential Concepts
Master Data
By: Budoor Ahmed Reference and Master Data
Essential Concepts
Data Sharing Architecture
• Systems of Record: Specific to Master Data subject areas (e.g., HR for employee data, CRM for customer
data).
• Architectural Approaches:
1. Registry:
1. Index pointing to Master Data across systems.
2. Pros: Easy to implement, minimal changes required.
3. Cons: Complex queries, semantic differences handling.
2. Transaction Hub:
1. Central system where Master Data is accessed and updated.
2. Pros: Enhances governance, provides consistent Master Data source.
3. Cons: High cost to modify existing systems, centralized business rules management.
3. Consolidated:
1. Hybrid approach; Master Data managed locally but consolidated for reference.
2. Pros: Enterprise view with minimal impact on existing systems.
3. Cons: Data replication, potential latency issues.
By: Budoor Ahmed Reference and Master Data
MDM Activities
Define MDM Drivers and Requirements
• Identifying MDM Drivers:
• Improvement Goals: Enhance customer service, operational efficiency.
• Risk Mitigation: Address privacy and compliance concerns.
• Defining Requirements:
• Within Applications: Relatively straightforward.
• Across Applications: More challenging, requiring standardization.
• Strategic Approach:
• Focus: Target one Master Data subject area or entity at a time.
• Prioritization: Based on cost/benefit analysis and complexity.
• Learning: Start with simpler categories to refine the process.
By: Budoor Ahmed Reference and Master Data
MDM Activities
Evaluate and Assess Data Sources
• Assess the structure, content, and creation processes of current application data.
• Goals of Data Assessment:
• Metadata Improvement: Enhance Metadata through quality assessment.
• Completeness: Determine data completeness for Master Data attributes.
• Definitions and Granularity: Clarify attribute definitions and granularity.
• Semantic Reconciliation: Align on attribute naming and definitions across the business.
• Data Quality Assessment:
• Identify Quality Issues: Assess data quality and address root causes.
• Disparity Between Sources: Tackle structural and value representation differences.
• Assumption: Expect data to be imperfect and verify its quality.
By: Budoor Ahmed Reference and Master Data
MDM Activities
Define Architectural Approach
• Align MDM with organizational goals.
• Models:
1. Transaction Hub: Best for centralized, manageable environments.
2. Registry: Index-based, links multiple data sources without centralizing.
3. Consolidated: Mixes hub and registry, suitable for diverse, siloed organizations.
• Decision Factors:
• Source System Variety: Influences integration strategy.
• Organizational Scale: Impacts choice between hub, registry, or consolidated model.
• Implementation:
• Tailored to organization's size, geographic spread, and data integration needs.
• Involves stakeholders from business domains, data architecture, and enterprise architecture.
• Choose MDM architecture that facilitates efficient data management, integration, and quality across the
organization.
By: Budoor Ahmed Reference and Master Data
MDM Activities
Model Master Data
• MDM as a process for integrating data across the organization.
• Models help integrate new sources as the organization grows.
• Standardization: Establishes enterprise-wide definitions for entities and attributes.
• Canonical Model.
By: Budoor Ahmed Reference and Master Data
MDM Activities
Define Stewardship and Maintenance Processes
• Need for Stewardship: Essential for addressing process exceptions and enhancing data quality.
• Ongoing Quality Support: MDM projects must allocate resources for sustained Master Data quality.
• Continuous Improvement: Involves analyzing records, improving source systems, and refining MDM
algorithms.
By: Budoor Ahmed Reference and Master Data
MDM Activities
Establish Governance Policies to Enforce Use of MD
• Initial Master Data efforts are challenging but essential.
• Governance Benefits: Operational efficiency, quality improvement, enhanced customer service.
• Roadmap for system adoption of master values and identifiers.
By: Budoor Ahmed Reference and Master Data
Reference Data Activities
Define Drivers and Requirements
• Drivers: Operational efficiency and improved data quality.
• Cost Effectiveness: Central management reduces costs compared to decentralized maintenance.
• Risk Reduction: Minimizes inconsistency risks between systems.
• Priority: Complex Reference Data sets require more effort than simple ones.
• System Requirements: Defined by the most important Reference Data sets.
• Maintenance: Scheduled upkeep of existing Reference Data sets.
By: Budoor Ahmed Reference and Master Data
Reference Data Activities
Assess Data Sources
• Standard Reference Data: Obtainable from creating/maintaining organizations; some free, some paid.
• Internal Reference Data: Challenges in identifying and managing sources within the organization.
• Central Management: Requires understanding and support from data owners for enterprise-wide
stewardship.
By: Budoor Ahmed Reference and Master Data
Reference Data Activities
Define Architectural Approach
• Consider Requirements: Account for the specific needs and challenges of managing Reference Data, including
data volatility and update frequency.
• Plan for integrating data purchased from vendors, considering delivery and integration methods.
• Ensure the system allows for easy manual updates with data entry rules to maintain data integrity, especially
for hierarchical data.
• Implement tools that enable Data Stewards to update data ad-hoc with automated approval workflows.
• Inform data consumers about updates, especially when changes affect programming logic.
• Evaluate the potential impacts of Reference Data changes before implementation.
By: Budoor Ahmed Reference and Master Data
Reference Data Activities
Model Reference Data Sets
• Reference Data includes complex information like ZIP codes with state, county, and geopolitical attributes.
• Creates data models for Reference Data sets to clarify relationships and establish quality rules.
By: Budoor Ahmed Reference and Master Data
Reference Data Activities
Define Stewardship and Maintenance Processes
• Stewardship Role: Ensures Reference Data is accurate, complete, and current. Stewards may directly update
data or facilitate consensus among business units.
• Metadata Documentation: Capture essential details of each Reference Data set, including stewardship,
update frequency, and usage processes.
• Stewards may help define common values for shared concepts across different business units.
• Utilize Reference Data Management tools for streamlined review and approval workflows.
• Identify organizational roles accountable for Reference Data content and maintenance.
By: Budoor Ahmed Reference and Master Data
Reference Data Activities
Establish Reference Data Governance Policies
• Central Management: Ensures organizational value through centrally managed Reference Data.
• Policies must mandate use of Reference Data to guarantee consistency and quality.
• Direct & Indirect Use: Reference Data should be utilized either directly from the repository or indirectly via a
system of reference.
By: Budoor Ahmed Reference and Master Data
Reference Data Activities
Establish Reference Data Governance Policies
• Central Management: Ensures organizational value through centrally managed Reference Data.
• Policies must mandate use of Reference Data to guarantee consistency and quality.
• Direct & Indirect Use: Reference Data should be utilized either directly from the repository or indirectly via a
system of reference.
By: Budoor Ahmed Reference and Master Data
3. Tools and Techniques
• Master Data Management (MDM) requires specialized tools for effective identity management.
• MDM can be implemented using a variety of tools including:
1. Data Integration Tools
2. Data Remediation Tools
3. Operational Data Stores (ODS)
4. Data Sharing Hubs (DSH)
5. Specialized MDM Applications
• Many vendors offer solutions covering different Master Data subject areas.
By: Budoor Ahmed Reference and Master Data
Implementation Guidelines
• MDM and RDM are forms of data integration with applicable principles of data interoperability.
• Implementations are not overnight; they require specialized knowledge and are best done incrementally
through a roadmap, prioritized by business needs.
• Success in MDM programs hinges on proper governance. Data governance professionals must understand
MDM and RDM challenges and assess organizational readiness.
• Establishing a proper reference architecture is crucial for managing and sharing Master Data.
• Monitor data flow to understand sharing, identify lineage, assist in issue analysis, evaluate integration
techniques, and assess data latency and business rule validity.
• Reference Data changes must be controlled, requiring organizational willingness to give up local control.
• Establish agreements to define data sharing conditions, supported by Data Governance programs and
involving various stakeholders.
• Ensure data quality through SLAs, metrics, and processes to address data issues.
By: Budoor Ahmed Reference and Master Data
Implementation Guidelines
By: Budoor Ahmed Reference and Master Data
Organization and Cultural Change
• MDM and RDM necessitate that individuals relinquish control over some data and processes to foster shared
resources.
• Local data managers may view MDM/RDM efforts as complicating their workflows, despite the risks of locally
managed data.
• Most recognize the benefit of having a single, accurate, and complete view of data (e.g., customer
information) over multiple partial views.
• Enhancing reference and Master Data quality and availability will require changes to existing procedures and
practices, tailored to the organization's current readiness and aligned with its mission and vision.
• A significant cultural shift involves establishing clear accountability for decisions among Data Stewards,
Architects, Managers, Executives, and ensuring collaborative decision-making by data stewardship teams,
program steering committees, and the Data Governance Council.
By: Budoor Ahmed Reference and Master Data
Reference and Master Data Governance
• Governance and stewardship are crucial for managing shared Reference and Master Data effectively.
• Role of Governance:
• Identifies data sources for integration.
• Enforces data quality rules.
• Sets conditions for data use.
• Determines monitoring activities and frequency.
• Prioritizes data stewardship efforts and response levels.
• Defines representation of information for stakeholder needs.
• Establishes standard approval processes for RDM and MDM deployment.
• Governance processes include compliance, legal stakeholders, and information consumers to mitigate
organizational risks through privacy, security, and data retention policies.
• Data governance must be an ongoing process that can adapt to new requirements and changes, making
principles, rules, and guidelines known to users of Reference and Master Data.
By: Budoor Ahmed Reference and Master Data
Congratulations, Reference and Master Data wizards! You've successfully navigated the data realms,
and now you're certified nocturnal data guardians. Until our next moonlit adventure, stay data-
driven and dream in hexadecimal! #GoodNightData
