1.1. The Three Questions Framework

💡 First Principle: Every data architecture decision comes down to three fundamental questions. Master these, and you can evaluate any Azure service, even ones that didn't exist when you studied. Think of these questions as a diagnostic checklist—just like a doctor asks about symptoms before prescribing treatment, a data professional asks these questions before recommending services.

What happens when you skip this framework? You end up storing images in a SQL database (expensive and pointless), running real-time fraud detection with batch processing (too slow), or using a global database for a single-region app (overkill and costly). The framework prevents these mismatches.

The Three Questions:

1. What SHAPE is your data? (Structured → Semi-structured → Unstructured)
2. What is your INTENT? (Transactional → Analytical → Archival)
3. What SCALE do you need? (Single-region → Multi-region → Global)

⚠️ Exam Trap: stopping after the first question. Shape alone never picks a service — structured data belongs in Azure SQL Database for a transactional workload and in Synapse for an analytical one. All three answers together narrow the field; any one of them on its own leaves several options standing.