Beyond Cloud Migration: Why the real challenge has only just begun

Back in 2020, I, like many others, believed the hardest part of cloud adoption was almost behind us and opportunities laid ahead if only we could get our cloud strategy right. I even wrote a blog post about it (The Cloud Arms Race) which highlighted some of the key issues I saw. Five years on… Have we now got things right?

I predicted that cloud migration was just the beginning, with cloud optimisation becoming the next big battleground. I suggested that success would depend on strong DevOps strategies, skilled talent, and avoiding vendor lock-in. I anticipated growth in open infrastructure solutions and a fragmentation of the cloud market leading to a price war. I also highlighted that most organisations would struggle to unlock the next level of value due to significant skill shortages and poor architectural choices early in the development process.

I think my predictions were scarily accurate. However, in hindsight, I significantly underestimated just how complex, fragmented, and critical the next phase of cloud maturity would become and how slowly organisations would move.

What I didn’t predict was the explosion in platform complexity, the rise of multi-cloud, compliance-by-design, security, automation, AI ops, and developer experience as a business differentiator. These have created a new reality where the cloud isn’t just infrastructure, it’s a product that needs to be managed like one.

Where we got it wrong.

I encouraged organisations to invest in DevOps to unlock efficiency and cost savings. They didn’t, and I don’t believe they have identified this as a genuine problem as they muddle on. Instead, the patchwork of tools, sprawling environments, and misaligned teams have created silos, operational drift, and spiralling costs. Most organisations don’t need more cloud, they need a cohesive platform strategy.

Why should you care now?

• CEOs: Cloud and platform engineering are now strategic enablers of innovation and business agility. If your teams can’t ship quickly, scale reliably, and manage costs, your growth slows. The cloud is not just a cost centre, it’s where your business value is created or lost.

• CFOs: Cloud sprawl and poor platform governance are killing your budgets. The price war hasn’t happened, vendor lock-in is real and costs are spiralling out of control. You need visibility, predictability, and operational controls baked into delivery pipelines to bring financial accountability back into the engineering process.

• CTOs/CIOs: The challenge isn’t migrating workloads, it’s enabling teams to build, run, and secure services at scale. You need consistent standards, automation, and governance across hybrid and multi-cloud environments as part of a unified platform architecture without building more technical debt and increasing costs.

• Heads of Engineering: The cloud has added more complexity than we ever imagined. Workloads, skills gaps, burnout, shadow ops, and DIY platform stacks have created inefficiencies and technical debt at a scale far bigger than we could have ever imagined. Do your teams sleep comfortably at night?

So how did we get into this unholy mess?

Simple. Many problems started with a “lift and shift” migration where legacy systems were moved to the cloud without redesign. While this was seen as a quick win, it often led to long-term inefficiencies with long-term consequences. Without effective cloud optimisation, organisations face much higher costs from underutilised resources, miss out on key cloud-native benefits like scalability and automation, and struggle with increased operational complexity. Additionally, these systems often fail to meet modern cloud security and compliance standards, introduce operational risk and continue to build technical debt instead of delivering the promised agility and savings.

Why “Lift and Shift” without cloud optimisation has failed:

1. Wasted Costs — Cloud pricing is based on usage and efficiency. Moving bloated, legacy systems into the cloud without re-architecting them means you’re likely paying more than you did before, especially for idle resources, underutilised VMs or over-provisioned services.

2. No Cloud-Native benefits — You miss out on scalability, elasticity, and automation because the service still behaves like it’s in an on-prem data centre.

3. Operational complexity — Legacy tools and architectures often don’t integrate well with modern cloud services, causing teams to manually patch and monitor systems, increasing risk and technical debt.

4. Security & Compliance — Cloud environments require new security models, like identity-based access and zero trust. Lift-and-shift workloads rarely align with these models and can introduce vulnerabilities.

5. Skills gaps — The shift to platform engineering requires a fundamental change in mindset from manually managing infrastructure to building automated, reusable platforms with a developer-first approach. It demands new skills in infrastructure as code, CI/CD, container orchestration, and close collaboration with engineering teams, moving from GUI-driven tasks to code-driven, scalable systems

In short, lift and shift was never “transformation”, and platform engineering is not “sysadmin 2.0”. It’s a fundamentally different discipline designed for speed, scale, and developer empowerment, and organisations are paying a heavy price for failing to adapt to this new reality.

Today, the real question isn’t whether you’re in the cloud, it’s whether your cloud is working for you or against you.