Enterprise Data Modelling¶
The organizational practice of creating and governing a shared conceptual model of enterprise data—its business entities, meanings, relationships, rules, and authoritative vocabulary—across systems and projects.
Core Idea¶
Enterprise data modelling is semantic coordination at organizational scale. It identifies what the business means by its key data and how those concepts relate before technology-specific schemas fragment the vocabulary.
The model becomes useful through governance and mapping. Diagrams and dictionaries must connect to accountable stewards, decision processes, integration contracts, physical systems, and managed evolution.
Scope of Application¶
- Data governance. Assigns definitions, ownership, and change authority.
- Systems integration. Maps heterogeneous applications to shared semantics.
- Analytics. Stabilizes dimensions, measures, and lineage.
- Architecture planning. Guides platform and domain boundaries without dictating one implementation.
Clarity¶
State organizational scope, modelling level, concepts and definitions, identifiers, relationships and cardinalities, business rules, notation and repository, source systems, local-to-enterprise mappings, steward and decision rights, extension policy, versioning, lineage, review cadence, quality measures, and explicit exclusions. Inclusion test: Require a deliberately shared, governed model of data semantics and relations across an enterprise or comparably broad organization, with traceable links to local systems. Exclusion test: Exclude one database's physical schema, an application object model, an ungoverned diagram, a list of data fields, a data catalog containing metadata but no integrated conceptual relations, and enterprise architecture with no data model. Nearest boundary: A canonical integration model standardizes exchange representations; an enterprise data model can inform it but also defines enduring business semantics independent of one interface. Exit condition: The model loses enterprise standing when scope, definitions, ownership, and mappings are not maintained, or when one system's implementation details are mistaken for organization-wide truth. Common misclassifications: It is not merely one database schema. It is not a diagram with no governance. A data catalog is not necessarily an enterprise model. One canonical XML schema cannot capture every conceptual and physical layer. Nearest named distinctions: Physical data model: Specifies storage details for a particular implementation. Data catalog: Inventories assets and metadata but may not reconcile concepts. Enterprise architecture: Covers capabilities, applications, and technology beyond data semantics. Ontology: Can provide formal semantics but is not identical to the organizational practice.
Manages Complexity¶
Large organizations contain overlapping terms, duplicated identities, local rules, acquisitions, and legacy schemas. A shared model must reconcile enough meaning for interoperability without becoming either an abstract poster or a centralized bottleneck.
Abstract Reasoning¶
- Set enterprise scope and priority decisions the model must support.
- Elicit business concepts and rules from processes, records, and stakeholders.
- Reconcile identities, synonyms, and conflicting definitions.
- Represent relationships and map them to local systems without copying implementation accidents.
- Govern ownership, versions, extensions, and measurable adoption over time.
Knowledge Transfer¶
Conceptual modelling and governance transfer across organizations, but entities, authority, regulation, and system mappings are enterprise-specific. A template industry model becomes authoritative only through local validation and stewardship.
Relationships to Other Abstractions¶
Current abstraction Enterprise Data Modelling Domain-specific
Parents (1) — more general patterns this builds on
-
Enterprise Data Modelling presupposes Schema Prime
Enterprise Data Modelling presupposes Schema: the parent's defining role is necessary to the child's frozen mechanism or criterion.
Hierarchy path (1) — routes to 1 parentless root
- Enterprise Data Modelling → Schema → Abstraction
Neighborhood in Abstraction Space¶
Enterprise Data Modelling sits in a crowded region of the domain-specific corpus (31st percentile for distinctiveness): several abstractions share nearly its structure, so a description that fits it tends to fit its neighbors too.
Family — Organizational Patterns & Management Concepts (29 abstractions)
Nearest neighbors
- Virtual Design and Construction — 0.92
- Broadbanding — 0.88
- Commons-Based Peer Production — 0.88
- Translative Case — 0.88
- Mushroom Management — 0.88
Computed from structural-signature embeddings · 2026-10-08