Cloud Migration Decision Framework

Cloud migration is not a technology project — it is a business transformation that happens to involve technology. This framework provides a structured approach to evaluating, planning, and executing Azure migrations for enterprise workloads, with decision criteria that help teams avoid the most common and costly mistakes.

Workload Assessment Checklist

Before migrating anything, classify each workload using the 6R framework:

1. Rehost (Lift-and-Shift) — Move as-is to Azure VMs. Fastest, lowest effort, but no cloud-native benefits.
2. Replatform — Minor modifications (e.g., move SQL Server to Azure SQL Managed Instance). Moderate effort, significant operational benefit.
3. Refactor — Redesign for cloud-native (containers, serverless, managed services). Highest effort, highest long-term value.
4. Repurchase — Replace with SaaS (e.g., on-prem CRM to Dynamics 365). Eliminates maintenance burden entirely.
5. Retire — Decommission. The most cost-effective migration strategy for workloads that should not exist.
6. Retain — Keep on-premises. Some workloads have regulatory, latency, or cost reasons to stay.

Decision Criteria: Lift-and-Shift vs. Refactor

The most consequential decision in any migration is where to draw the line between rehost and refactor. Use these criteria:

• Lift-and-shift when: The application has a limited remaining lifespan (< 3 years), has no scaling requirements, or when speed-to-cloud is the primary goal (datacenter exit deadline).
• Refactor when: The application is strategic, has growing scale requirements, would benefit from managed services (auto-scaling, serverless), or when operational cost reduction is a priority.

FinOps: Cost Optimization from Day One

Cloud cost overruns are the most common post-migration complaint. Build FinOps practices into the migration plan:

• Right-size VMs before migration — analyze actual CPU/memory utilization, not allocated capacity
• Use Reserved Instances for predictable workloads (1- or 3-year commitments save 30–60%)
• Implement auto-scaling for variable workloads — pay for what you use
• Set budget alerts and cost anomaly detection from the first deployed resource
• Tag every resource with cost center, environment, and owner for chargeback visibility

Phased Rollout Strategy

Migrate in waves, not all at once:

1. Wave 0 — Foundation: Landing zone, networking, identity (Azure AD/Entra ID), governance policies. No workloads yet.
2. Wave 1 — Non-critical workloads: Dev/test environments, internal tools. Build operational muscle.
3. Wave 2 — Business applications: Production workloads with established rollback procedures.
4. Wave 3 — Core systems: ERP, databases, mission-critical apps. Maximum preparation, minimum risk.

The most expensive migration is the one you have to redo. Invest in the assessment phase, validate with a pilot wave, and scale only when operational confidence is established.