Business, Data, Application and Technology Architecture

The four EA domains (BDAT), what each describes, key artifacts, how they connect, and practical examples from a retail bank and a manufacturer.

The four domains

Domain Describes Key artifacts
Business architecture Strategy, capabilities, value streams, organization, processes, information concepts Capability map, value stream map, organization map, business process models
Data architecture Data entities, ownership, flows, lifecycle, quality, governance Conceptual/logical data models, data lineage, master data design, data catalog
Application architecture Applications, their functions, interfaces and interactions Application portfolio catalog, application communication diagram, interface catalog
Technology architecture Platforms, infrastructure, networks, standards supporting applications Technology standards, platform reference architectures, deployment views
flowchart TB
  S[Business strategy & goals] --> BA[Business architecture<br/>capabilities, value streams]
  BA --> DA[Data architecture<br/>entities, ownership, flows]
  BA --> AA[Application architecture<br/>apps supporting capabilities]
  DA --> AA
  AA --> TA[Technology architecture<br/>platforms, cloud, network]

Traceability

The value of EA comes from linking these: goal → capability → applications supporting it → data used → technology hosting it. This answers questions like "which applications support Customer Onboarding, which of them run on end-of-life technology, and what customer data do they hold?".

Example: retail bank

  • Business: capability "Customer Onboarding" scored low on maturity; value stream "Open account" takes 5 days.
  • Data: Customer data mastered in three systems with conflicting addresses; need a customer MDM and golden record.
  • Application: onboarding uses a legacy branch app + a separate KYC tool + manual steps; target: digital onboarding platform with API integration to KYC and core.
  • Technology: legacy app runs on unsupported middleware; target platform is the bank's Kubernetes and API management standard.

Data architecture essentials

  • Data ownership and stewardship per domain (data mesh principles where appropriate).
  • Master data management (customer, product, supplier) and reference data.
  • Data classification and privacy controls.
  • Integration patterns: APIs for operational access, events/CDC for propagation, lakehouse for analytics.
  • Data quality dimensions: accuracy, completeness, consistency, timeliness, validity, uniqueness.

Interview questions and answers

How do you connect business strategy to technology decisions?

Map strategic goals to capabilities, assess capability maturity and importance, identify which applications and data underpin the priority capabilities, then prioritize technology investments that close those gaps, documenting the trace for funding decisions.

What is the difference between solution architecture and enterprise architecture?

Enterprise architecture spans the whole organization and long horizons (portfolio, standards, roadmaps). Solution architecture designs a specific solution for a specific problem within those guardrails, in detail, for delivery.