Data Architect

What a Data Architect does

Data Architects design how data flows, is stored, secured, and made useful across an organization. You define standards and patterns, make platform choices, and guide teams to implement reliable, scalable, and compliant data systems.

• Day-to-day: shape data models and integration patterns, review designs, set governance rules, guide build vs. buy decisions, and align stakeholders.
• Typical deliverables: architecture blueprints, conceptual/logical/physical models, data contracts, governance policies, lineage and catalog standards, performance and cost guidelines, and reference implementations.

Who this is for

• Engineers or analysts who enjoy systems thinking, data modeling, and long-term platform strategy.
• People who like clarifying ambiguity, documenting decisions, and aligning diverse teams.
• Those who value reliability, data quality, security, and measurable business outcomes.

Prerequisites

• Comfort with databases (relational and/or cloud data warehouses) and SQL basics.
• Understanding of batch and streaming integration concepts.
• Ability to read/write technical documentation and communicate trade-offs clearly.

Hiring expectations by level

Salary and growth

• Junior/Associate: ~$80k–$120k
• Mid-level: ~$110k–$160k
• Senior/Lead/Principal: ~$150k–$220k+
• Manager/Head of Data Architecture: ~$170k–$250k+
• Contract/Freelance: ~$90–$160/hr

Varies by country/company; treat as rough ranges.

Where you can work

• Industries: finance, healthcare, retail, tech, manufacturing, government, SaaS
• Teams: data platform, analytics engineering, data governance, security, enterprise architecture
• Work modes: product companies, consultancies, agencies, and internal platform teams

Skill map

As a Data Architect, you’ll combine strategy, modeling, integration, governance, security, and delivery skills. The skills below match the Skills section of this page.

• Data Architecture Strategy: principles, roadmaps, and decision frameworks
• Conceptual and Logical Data Modeling: business entities and relationships
• Physical Data Modeling and Storage Design: schemas, indexing/partitioning
• Dimensional Modeling for Analytics: star/snowflake, facts, slowly changing dimensions
• Integration Architecture (ETL/ELT): batch vs. streaming, CDC, orchestration
• Data Governance and Stewardship: policies, roles, compliance, lifecycle
• Security and Privacy Architecture: PII/PHI controls, encryption, access models
• Data Quality and Observability: rules, SLAs/SLOs, monitoring and alerting
• Metadata and Lineage Architecture: catalogs, lineage, data contracts
• Performance and Scalability: workload management, cost optimization
• Architecture Delivery and Communication: ADRs, roadmaps, stakeholder alignment

Practical portfolio projects

1. Data Platform Blueprint
   Deliverables: current-state vs. target-state diagrams, principles, NFRs, platform choices, ADRs, and a 3–6 month roadmap.
2. Analytics Warehouse with Dimensional Modeling
   Deliverables: star schema for a business domain (e.g., Orders), SCD handling, data contracts, and sample queries with performance notes.
3. Streaming Integration (CDC to Analytics)
   Deliverables: CDC design, idempotent processing, schema evolution policy, replay strategy, and end-to-end lineage diagram.
4. Governance and Catalog Rollout
   Deliverables: data classification scheme, ownership (stewards), glossary terms, catalog structure, and steward workflow.
5. Performance and Cost Optimization Plan
   Deliverables: workload profiling, partitioning/clustering plan, storage tiering, and a before/after cost-performance comparison.
6. Quality and Observability Framework
   Deliverables: tiered SLAs/SLOs, data tests, alert thresholds, escalation paths, and dashboard mockups.

Interview preparation checklist

• Explain conceptual vs. logical vs. physical models with a simple example.
• Design a star schema and justify grain, facts, and dimensions.
• Describe your approach to CDC, schema evolution, and idempotency.
• Discuss PII handling: classification, masking, key management, and auditability.
• Show how you measure data quality and define SLAs/SLOs.
• Walk through an ADR you wrote and the trade-offs involved.
• Estimate and optimize a workload for performance and cost.
• Describe lineage and catalog strategies for self-serve analytics.
• Communicate a migration plan with milestones and rollback steps.

Common mistakes (and how to avoid them)

• Big-bang designs with no delivery proof: favor iterative delivery with reference implementations.
• Over-modeling early: start conceptual, validate with real queries, then refine.
• Ignoring data quality and contracts: define tests and contracts before integrating consumers.
• One-size-fits-all tooling: select patterns per workload (OLTP vs. analytics vs. streaming).
• Neglecting security/privacy early: classify data and define controls up front.
• Unclear ownership: assign stewards and producers/consumers with clear responsibilities.

Learning path

Recommended sequence that balances fundamentals and delivery:

1. Conceptual and Logical Data Modeling
2. Dimensional Modeling for Analytics
3. Integration Architecture (ETL/ELT)
4. Data Quality and Observability
5. Metadata and Lineage Architecture
6. Security and Privacy Architecture
7. Performance and Scalability
8. Data Governance and Stewardship
9. Physical Data Modeling and Storage Design
10. Data Architecture Strategy
11. Architecture Delivery and Communication

Next steps

• Pick one portfolio project and deliver a thin slice within two weeks.
• Document everything: models, ADRs, contracts, test plans, runbooks.
• Practice the interview checklist aloud and refine your artifacts.

Pick a skill to start — see the Skills section below.