Cloud strategy · 8 min read

Single cloud or
multi-cloud?

Multi-cloud is often presented as the mature answer. For many organisations, a well-designed single-cloud strategy is the smarter place to start.

There is nothing inherently wrong with multi-cloud. In the right context, it can be a deliberate, valuable choice. But it is not a free source of flexibility, and it is not automatically a stronger negotiating position. It is an operating model with real cost, capability and complexity attached to it.

The case for starting single-cloud

A single-cloud strategy lets an organisation focus its investment. Teams can build deeper platform knowledge, establish one coherent security and identity model, and create reusable patterns for data, networking, observability and delivery. That concentration matters, especially for organisations that are still strengthening their cloud foundation.

The result is often faster progress with less architectural duplication. Instead of maintaining several versions of the same landing zone, integration pattern, tooling stack and governance model, teams can make one foundation dependable and usable.

Does multi-cloud improve negotiating power?

It is often argued that the ability to move workloads gives an organisation more leverage with the large cloud providers. In my experience, the size and consistency of the committed volume usually have a much greater impact on commercial conversations than the theoretical option to move a workload elsewhere.

Large providers are open to negotiation, particularly where an organisation has a clear roadmap and predictable demand. There are also established programmes that can reduce spend without introducing a second cloud: reservations, savings plans and longer-term commitments can all be effective when they are matched to stable usage.

The hidden operating cost

Making workloads portable across clouds is not simply a deployment choice. It means maintaining skills, security controls, pipelines, monitoring, data patterns and operational procedures across more than one platform. For many organisations—particularly those that are not born cloud-native—that is a significant additional cost.

Portability can be useful for specific services or regulatory needs. But making every workload movable often creates a broad tax on engineering time, reliability and focus. The question is not whether portability is possible. It is whether it creates enough value for the organisation to justify what it takes to sustain it.

Choose the strategy that fits

The right strategy starts with the organisation: its business model, risk profile, capabilities, data needs and appetite for operating complexity. For most companies, a disciplined single-cloud foundation can deliver the resilience, scale and commercial control they need. Multi-cloud should be earned by a clear business case—not adopted as a default.

We will explore the practical choices behind this in more depth in a future insight.