Data Architecture Domain
Data Architecture Domain
Starting From What the System Needs to Know, Not From What It Already Has
The Data Architecture Domain governs the structural design of the data required by the marketing system, including its definition, availability, quality, integration, governance, and translation into usable intelligence. It establishes the informational foundation on which measurement, technology, AI, and most strategic decisions depend.
This domain begins with requirements rather than available data, which is an important distinction. Organizations frequently build reporting and analytical environments around whatever their existing platforms happen to capture. Data Architecture reverses that logic by first determining what the system must know, what decisions that information must support, and what data structures are therefore required.
Data Quality as a Design Output, Not a Cleanup Task
This domain covers data design, availability architecture, integration architecture, and requirements translation. Data quality is treated as a structural design output rather than simply a cleansing problem to be addressed after the fact, reflecting the MABOK's principle that data architecture should be designed from intelligence requirements downward, not from data availability upward.
A mature Data Architecture produces consistent definitions, governed access, interoperable information, and data suited to the decisions it supports. Because AI and measurement can never exceed the integrity of the data underlying them, weaknesses in this domain become constraints that propagate throughout the rest of the marketing architecture.
What a Mature Data Architecture Domain Looks Like
An organization with mature Data Architecture has documented, consistent definitions for its core metrics across every team that uses them, and can trace any reported number back to a governed source rather than to a spreadsheet someone built once and everyone kept copying.