Lift and shift — migrating workloads to the cloud without significant modification — has become the default first move for organisations beginning their cloud journey. It’s fast, relatively low-risk, and gets infrastructure off ageing hardware. For those goals, it works.
But treating lift and shift as a destination rather than a starting point is where organisations leave significant value on the table.
What You’re Getting vs What You’re Paying For
When you lift and shift a workload to Azure, you’re running your existing architecture on Microsoft’s infrastructure. You benefit from managed hardware, improved uptime, and geographic redundancy. What you’re not getting is the cloud-native capability that actually changes economics.
A VM running 24/7 in Azure because it was running 24/7 on-premises is not delivering cloud value — it’s just outsourced hosting, often at a higher unit cost than the hardware it replaced.
The operational cost reductions that cloud vendors cite — the 30–40% reduction in infrastructure spend that appears in their case studies — come from refactoring, not rehosting.
The Three Layers of Cloud Value
Layer 1: Rehosting. Move the workload. Benefit: hardware management elimination, basic resilience improvement.
Layer 2: Re-platforming. Modify the workload to take advantage of managed services without rewriting the application. A database moved to Azure SQL Managed Instance instead of SQL Server on a VM. A job scheduler replaced with Azure Functions. This layer starts delivering meaningful operational cost savings by eliminating the management overhead of supporting infrastructure.
Layer 3: Refactoring / Re-architecting. Redesign workloads to be cloud-native — auto-scaling, event-driven, serverless where appropriate. This is where 20–30% operational cost reductions become achievable, not because cloud compute is cheaper, but because you’re only consuming resources when the workload actually requires them.
Where the 30% Number Comes From
Operational cost reduction of this magnitude typically comes from a combination of:
- Right-sizing — eliminating over-provisioned resources that were sized for peak demand and run at 10% utilisation the rest of the time
- Auto-scaling — matching compute to actual workload patterns
- Managed services — eliminating the operational overhead of patching, maintaining, and monitoring infrastructure components that Azure manages for you
- Reserved instances — committing to baseline usage at significant discounts once consumption patterns are understood
None of these require rewriting applications from scratch. Many are achievable within 6–12 months of initial migration through targeted re-platforming.
What a Realistic Roadmap Looks Like
We typically structure cloud value realisation in three phases:
- Migrate — get workloads off ageing on-premises infrastructure, prioritising those with clear business continuity risk
- Optimise — within 90 days of migration, conduct a FinOps review and implement right-sizing, auto-scaling, and managed service replacements where they deliver clear ROI
- Innovate — use the operational savings and freed engineering capacity to build capabilities that weren’t possible on-premises: AI, advanced analytics, event-driven architectures
The organisations that capture genuine cloud value are the ones that plan phase 2 at the same time as phase 1 — not as an afterthought 18 months later.
If your organisation has migrated to the cloud but isn’t seeing the economic benefits you expected, let’s talk. We frequently help organisations audit their existing cloud environments and identify where re-platforming investments will deliver the clearest return.