Master data requires identifying and developing a trusted instance version for each instance of a conceptual entity and maintaining the currency of that version

The challenge with master data is entity resolution, which is the process of identifying and managing associations between data from disparate systems and processes

The entities and instances represented by each row of master data have different expressions in different systems.

Master data management is to eliminate these differences so that individual entities and instances can be consistently identified in different environments.

It should be noted that this process must be continuously managed to keep the identities of these master data entities and instances consistent.

Both provide important contextual information for the creation and use of transactional data (reference data also provides context for primary data) in order to understand the meaning of the data

Both are shared resources managed at the enterprise level

Having multiple instances of the same reference data will reduce efficiency and inevitably lead to inconsistencies between instances. Inconsistency will lead to ambiguity, and ambiguity will bring risks to the organization.

Reference data does not change easily, and its data set is usually smaller, less complex, and has fewer columns and rows than a transaction data set or a master data set.

Reference data management does not include the challenges of entity resolution

Different management priorities

One challenge in reference data management is who leads or is responsible for the definition and maintenance of reference data

Identifying responsibilities for acquiring data and managing updates is part of reference data management

Lack of maintenance accountability creates risks as discrepancies in reference data may lead to misunderstanding of data context

list

Cross-reference data list

taxonomy

Ontology

Many organizations create reference data to support internal processes and applications

Used to describe a data set created and maintained by an industry association or government agency rather than by an organization to provide a common standard for encoding important concepts

For example, the International Classification of Diseases codes (ICD) provide a common way to classify health conditions and treatments.

Can be classified or analyzed based on geographic information

For example, Census Bureau reports on population density, translation of historical meteorological information into strict geographic categories

Many commercial activities rely on the use of some common, continuously calculated data

For example, foreign exchange calculations rely on well-managed, up-to-date exchange rate tables

The main difference between computational reference data and other types of reference data is how often it changes

Like other data, reference data changes over time

Because it is commonly used in a variety of organizations, it is important to maintain key data in reference data sets to ensure that their lineage and circulation processes are understood and maintained.

When different versions of a "truth" may exist, it is necessary to distinguish between them

In order to do this, it is necessary to know where the data comes from or is accessed, and the specific use and purpose for which it was prepared.

A system of record is an authoritative system that creates, captures, and maintains data using a defined set of rules and expectations

A reference system is also an authoritative system. Data consumers can retrieve reliable data from the reference system to support transactions and analysis, even if the information does not originate from the reference system.

Master data management applications MDM, data sharing center DSH and data warehouse DW are often used as reference systems

Trusted sources are considered the “best version of the truth”

Among trusted sources, the record that represents the most accurate data for an entity or instance can be called a golden record

Master data management can only ensure the consistent use of the most accurate and timely data for core business entities across systems by controlling master data values ​​and identifiers.

Goals include ensuring the availability of accurate, up-to-date values ​​while mitigating the risk of ambiguous identifiers

step

Data model management

data collection

Data validation, standardization and data enrichment

Entity resolution and identifier management

is data about individuals, organizations and their roles in business relationships

Includes data about business units, cost centers, profit centers, general ledger accounts, budgets, plans, and projects

Includes data regarding contracts, regulations and other legal matters

Product Lifecycle Management PLM

Product Data Management PDM

Enterprise Resource PlanningERP

Manufacturing Execution System MES

Customer Relationship Management CRM

Provides the ability to track and share geographic information and create hierarchies or maps based on geographic information

A reference catalog is the authoritative list of master data entities (companies, people, products, etc.) that organizations can purchase and use as the basis for transactions

The registry is an index to master data records in various systems of record

Record the master data local to the system management application and access the master data based on the master index

A registry is relatively easy to implement because it requires few changes to the system of record

In this approach, each application interacts with a central system to access and update master data.

Master data exists within Trading Center and not in any other application

Transaction Center is the system of record for master data

Hybrid mode is a mix of Registry and Transaction Center

Record master data local to system management applications

Master data is consolidated in a common repository and shared through a data sharing center, eliminating the need for direct access from systems of record