Open Framework, Information Management Strategy & Collaborative Governance | Data & Social Methodology - MIKE2.0 Methodology
Wiki Home
Collapse Expand Close

Collapse Expand Close

To join, please contact us.

Improve MIKE 2.0
Collapse Expand Close
Need somewhere to start? How about the most wanted pages; or the pages we know need more work; or even the stub that somebody else has started, but hasn't been able to finish. Or create a ticket for any issues you have found.

Conceptual Definition of Reference and Master Data

From MIKE2.0 Methodology

Jump to: navigation, search



Definition of Reference and Master Data

DataType Pyramid.jpg
As a conceptual model there are common differentiation of master and reference data. The main differentiation dimensions are the source and the time aspects of the data. Frequently also the usage scope is added as a differentiation factor. There is a discussion on where exactly the border between the different types of data is or how much value add the differentiation provides. However, for the master data strategy and blueprint discussion and clarification of scope and interdependencies the model has proven to be helpful.

Even though this model implies a strict hierarchical structure of data, distinction between the different types are blurry in reality and need to be clarified during detailed design. The scope of metadata, reference data and master data definition is also dependant on the maturity level of the information governance. It may be wise to start with limited metadata definitions and reference standards as long as the conceptual and strategic implication of the relationship between the different types of data are recognized.


Data about data, Definition and description of data objects, entities, attributes. Exists frequently in enterprise as data dictionary, but needs to be extended beyond the static definition to include business and governance rules. Metadata has the highest abstraction level and must be persistent over time. The source of metadata is internal to the enterprise. The scope of metadata is acorss all domains and foundational.
Examples: Technical and Conceptual standard on how to capture a customer address. Date and time formats and conversion rules for different regions and purposes (DDMMYYYY, dd. mm. yy etc…)

MIKE2.0 provides a detailed Defintion for Metadata

Reference Data

Structured Lists of Values that are used as lookup values by other (master) data. They serve to structure and organize data records and make the data usable in transactions. They are frequently defined by external standard institutions or agencies to facilitate cross enterprise exchange. However this is not a limiting condition, because especially in larger enterprises it is necessary to define shared internal reference data (e.g. organizational entities, valuation classes, product hierarchies etc.). Reference Data is static in nature, but has an evolutionary character. the scope of reference data is across different domains, entities and systems. The standardization of the values is a crucial exercise in all data harmonization and consolidation efforts. The maintenance of reference data is a crucial exercise, as the distribution of the reference data across all relevant systems is a prerequisite to maintain the interoperability between systems and processes.

Examples: ISO country codes, Custom Tariff numbers, DUNS numbers, etc.

Relation to Metadata: where the metadata defines the standard (eg: what is the correspondence language and how it is captured) the Reference Data is the list of the possible values with the respective name/value pairs (en > English, de > German, etc.)
Relation to Master and Transactional data: Reference Data ensures that master data and transactional data can be related to each other and business processes can be automated eg: the Definition of the Correspondence language must be mapped to the language of product descriptions and to the language of forms used during quoting, sales, and logistics transactions.

Master Data

Complex records that represent an external or internal entity that is the basis for business processes. While the master data often represents external entities, it is always specific to the enterprise. A company or a person exists independently of the enterprise, but as master data it exists as business partner, customer, supplier or employee in relationship to the enterprise. From a time perspective, master data is also stable, but it has an organic lifecycle. The lifecycle is organic as each record contains creation, change and obsoletion in correspondence with the development of the lifecycle status of the entity it represents. (for example: an employee master record gets created and changed according to the status of the employee in the enterprise (hire > transfer > terminate). The scope of master data is the enterprise.

Examples: Material, Product, Customer, Supplier, Employee, Chart of accounts, Asset etc.

Dependant Master Data

Dependant Master data is a persistent combination of several master data records and it represents logical relationships between entities. A bill of material for example describes which materials are part of a product, or a price is a combination of a product with an organizational Unit and a specific group of customers. Dependant master data have generally a shorter lifecycle than the referenced Master data and are frequently with clearly defined validity period (the price is valid “from to”). the scope of dependant master data is generally not the enterprise but the operational unit or the system (different units may have different prices for the same product, a service unit may have a different bill of material for the same product than a production unit). Due to this character, the overall volume of dependant master data is substantially larger than the volume of master data

Examples: List Prices, Contracts, Bills of Materials

Transaction Data

Transaction data is always created in relationship to the execution of a specific business process that it represents. The validity period of this data is restricted to the cycle time of the specific business process, even though the resulting document data may have a long retention period for reporting, tracking and compliance reasons. The scope of transaction data is always restricted to the operational unit. Due to the process execution focus the volumes of this data is generally the highest as it represents everything that happens with the entities described with master data.

Examples: Quotes, Sales Orders, Deliveries, Service Orders

Wiki Contributors
Collapse Expand Close
  • Zschmid
    Article Score:12
    Total Score:88
  • Bsomich
    Article Score:6
    Total Score:4351
  • Mickey72
    Article Score:3
    Total Score:81

View more contributors