Enterprise Architecture and Emerging Technology
This domain tests whether you can use architecture to connect business capabilities to technology standards and evaluate emerging technology without hype. Expect questions on TOGAF, architecture domains, capability mapping, cloud responsibility, and AI governance.
TOGAF and architecture domains
TOGAF's Architecture Development Method runs from the Preliminary phase through A Architecture Vision, B Business Architecture, C Information Systems Architectures, D Technology Architecture, E Opportunities and Solutions, F Migration Planning, G Implementation Governance, and H Architecture Change Management, with Requirements Management at the center. Architecture domains are business, data, application, and technology. Deliverables include principles, capability maps, reference architectures, building blocks, and an architecture repository. Good EA reduces duplication, guides portfolio rationalization, and makes change faster; it is not shelfware.
Emerging technology evaluation
Evaluate emerging technology against business capability fit, risk, total cost, skills, integration, lock-in, and regulatory obligations. Cloud uses shared responsibility: the provider secures the cloud, while the customer secures what it puts in the cloud. AI governance covers data provenance, model risk, bias testing, human oversight, monitoring, and disclosure. Use pilots with success criteria, architecture decision records, and exit plans.
Worked example
A service desk proposes a generative AI assistant to reduce handle time by 15%. Phase A defines the outcome and measures the baseline. Phase B identifies affected capabilities and process changes. Phase C classifies the data and designs integration with the knowledge base. Phase E compares build, buy, and partner options. An architecture decision record captures the choice, assumptions, and consequences. The pilot has guardrails: no customer personal data in prompts, human review for regulated answers, and a rollback plan. The pilot is judged on handle time, accuracy, and escalation rate - not on demo quality.
Traps that cost marks
- Treating EA as documentation rather than decisions and standards.
- Skipping the baseline architecture and designing only the target.
- Choosing technology before defining the capability and outcome.
- Ignoring data quality and integration cost in an AI business case.
- Assuming cloud shifts all security responsibility to the provider.
- Adopting emerging technology without an exit plan or lock-in analysis.