FinOps named a real problem: cloud costs that nobody owns. But most programmes treat it as a finance function or a tooling problem, when it is really a change in how engineering decisions get made.
Kubernetes has crossed the enterprise threshold. For most large organisations the constraint is no longer the platform but the skills, security, observability, and delivery model around it.
Multi-cloud promises portability, resilience, and vendor leverage. Most enterprises arrive at it by drift rather than design, and inherit operational complexity instead of strategic optionality.
Cloud migration plans fund infrastructure, tooling, and architecture, and almost nothing for the organisational change that makes the technical work stick. That gap, not technical complexity, is what stalls most programmes.
Most enterprises moved to the cloud without redesigning how IT delivers value. The cloud operating model, not the migration, is the decision that determines whether the investment compounds or stalls.