Master data model (federated data service)

1. Do you maintain a global directory or database of data objects, with global identifiers? Or do you have mechanisms for mapping these objects between different databases/directories? Are these mechanisms electronic or manual? Are all such objects mapped, or is this done only for certain applications and sets of objects? Are these mappings undertaken automatically by the infrastructure?
2. Do you have mechanisms for looking up global objects by searching on their characteristics?
3. Is there difficulty in moving data from one application to another? For all applications? For only some applications?

1. Is there a common data model across all applications?
2. Are there independent data models for different applications?
3. Are mapping rules used to convert between different data models?
4. Does your organization have or are you developing a Business Information Model to standardize data and message formats and concepts across the enterprise?

1. Are the data models defined by a language that includes taxonomies, ontologies, or other high-level logical representations?

Silo
(Level 1)

Application-specific Data Solution

Information is replicated and redundant. Conceptual enterprise information model is absent.

Integrated
(Level 2)

LOB or Enterprise-specific

Information is shared across some applications using Extraction, Transformation, Load, Manipulate (ETLM) or message-oriented technologies. Initial data vocabularies are beginning to emerge.

Componentized
(Level 3)

Canonical Models

Formal SOA methods and practices are employed by multiple groups within the enterprise.

The organization has a loosely defined enterprise architecture supported by limited tooling and governance practices.

Services
(Level 4)

Information as a Service

Multiple business units are using meta-data relationships.

Business data vocabularies are standardized within a business unit or process area.

Business data can be shared within a business unit and with business partners in a consistent manner. Interfaces are defined using common message data vocabularies.

Composite Services
(Level 5)

Business Data Dictionary and Repository

Information services are in place such as data validation, data cleansing, data transformation, partner data integration, or others.

Master data services are in place and are utilized across the enterprise.

Business data vocabularies are standardized for use across the enterprise.

Virtualized Services
(Level 6)

Virtualized Information Services

Business data vocabularies can be expanded and enhanced as required to support new services, external partners, and business process reconfiguration.

A registry with metadata is used to manage enterprise service assets.

A formal enterprise-wide business information model has been developed and deployed.

Dynamically Re-Configurable Services
(Level 7)

Semantic Data Vocabularies

Business data vocabularies can easily be expanded and enhanced as required to support new services, external partners, and business process reconfiguration.

Business data is defined using semantic web constructs, or ontologies (e.g., UN/CEFACT Core Components, ISO 11179).

A formal enterprise business information model has been designed and implemented that includes both enterprise and external relationship entities.