Cloud adoption: A once-in-a-decade decision

With insights from

Giulio Rognini
New Business Lead

A payback calculation shows affordability but says nothing about value. That missing lens is consequential. You are leaving value on the table by treating this once-in-a-decade architectural decision as a procurement exercise.

This article is written for those approaching cloud adoption who suspect that the decision process has stalled at the cost level. What often eludes the conventional cloud adoption business case is that cost is the floor, not the ceiling, of the assessment. Organizations that extract the most from the cloud – that achieve company transformation through cloud adoption – ask a different, harder question: what can we gain now, and what design decisions will create future opportunities and value?

Where cloud adoption ROI stops short

A rigorous cloud adoption assessment begins with precision about what is being built. “Move to the cloud” is a slogan, whereas the underlying intention of cloud adoption is very specific. In cloud adoption for MedTech, you are enhancing one or more data and processing workloads, whether building from the ground up to connect a fleet of medical devices, run analytics on patient data, and host a regulated software product, or migrating existing processes, such as an enterprise back-office. Each workload carries different requirements and architecture.

The standard approach surveys these workloads to define a current baseline and a target state. Infrastructure costs, data center expenses, licensing fees, maintenance overhead, and staffing are mapped to cloud equivalents, such as managed compute, storage, networking, and operational services. KPIs are then defined to reflect the needs met by the adoption and their value to the target state. These might include cost per workload, uptime, deployment frequency, or incident response time. The exercise produces a well-structured case for or against adoption.

Three blind spots of cloud adoption

This is necessary work for every credible cloud adoption project, but when it is the only lens applied, three blind spots emerge. The third blind spot is what makes the decision framing based on cost alone genuinely risky. Not wrong, but incomplete in a way that quietly closes opportunities before you know they existed.

The assumption that cloud adoption is always cheaper does not hold.

Infrastructure savings can be offset by data egress costs, managed service fees, licensing changes, and the ongoing effort of cloud optimization. The total cloud ownership cost is consistently more complex than the initial model suggests.

The human dimension is routinely underestimated.

Habits, workflows, and organizational structures travel with cloud adoption. Changing how teams think, build, and operate in a cloud environment has time, effort, and friction costs that often exceed those of compute.

The real value of cloud adoption rarely arrives at go-live.

Instead, it compounds over the next three to ten years, as the chosen architecture either opens or forecloses capabilities you did not know you would want. Go-live is a milestone, not a destination.

Shaping factors: what conventional approaches miss

Conventional assessments consider cost, scalability, and reliability. These are constraints. While you can decide the extent to which they impact your decision, they are always a restriction. There is another category of adoption decisions that appear to be constraints but behave very differently in practice. We call these shaping factors.

A shaping factor is a design-level decision about how your data is structured and governed, how your regulatory posture is set up, and how your platform is architected. Together, they define the business depth and breadth you can build on top of the cloud over the following five to ten years.

Uniquely, shaping factors are not one-directional. Cost is a one-way constraint: it can only restrict you. Shaping factors define guardrails, but they also open doors. Carefully planned and handled, shaping factors become a competitive advantage.

Two further properties distinguish them from ordinary project decisions.

Shaping factors are horizontal: a single architectural choice about data governance ripples across every workload on the new platform, not just the one you migrated first.
Shaping factors are time-loaded: the upshot of a shaping decision rarely surfaces immediately, but three, five, or ten years later, when a choice made on day one quietly reveals an option or a feature you did not know you wanted.

Thinking through shaping factors expands your cost analysis from a two-year payback exercise to a ten-year growth question. With this lens, your cloud adoption ceases to be a cost-reduction project and becomes a cloud adoption strategy conceptualized for long-term value creation.

To illustrate the impact of shaping factors, consider how Linde Healthcare, a global medical gas supplier, digitized its gas cylinder tracking process. The initial goal was to replace paper-based ordering with a cloud-based monitoring system that would give hospital teams reliable visibility of cylinder locations and stock levels. A clear efficiency play. But when the shaping factors were applied, unanticipated opportunities emerged.

Five shaping factors that define your cloud adoption

Each of the five shaping factors below serves a clear, anticipated purpose. Designed in concert, however, they harbor a second quality that conventional cloud adoption business cases never reach: they create the conditions for opportunities to surface as new applications, efficiencies, markets, or business models.

Shaping factor 1: Data sovereignty and governance

The anticipated benefit of well-designed data governance is clarity. For your Chief Data Officer and Data Protection Officer, knowing what you hold, where it can legally reside, and under what conditions it can be used allows the organization to operate confidently in regulated markets, attract clients who require clear infrastructure as a condition of partnership, and build verifiable data products. Cloud provider and model selection follow from governance and cloud sovereignty at regional and global levels, and organizations that design for complexity across markets can scale out without rebuilding.

The deeper value of that clarity is what it reveals over time. Well-classified and managed data grant transparency to make unforeseen connections. Patterns emerge across patient populations that suggest new clinical products. Partners approach you with collaboration proposals that build on a verifiable data foundation your competitors cannot offer. More than telling you where or how you operate, good governance shows you what you can build to drive growth.

Precise data sovereignty and governance meant that performance data aggregated across Linde’s entire deployed cylinder fleet was classified, accessible, and governed for the first time. Usage patterns emerge across clinical settings that no field report had ever captured: which hospital divisions run through cylinders fastest, at what intervals, and with what variation. That data informed the next generation of Linde’s offering: an advisory service for hospitals to visualize their own cylinder fleet performance.

Shaping factor 2: Security by design

Embedding security from the first architectural decision rather than adding it afterward builds trust. For your Chief Information Security Officer, R&D, and Quality teams, that trust is a commercial credential. It accelerates regulatory approvals, shortens sales cycles, and opens conversations that are simply not possible with organizations whose security posture cannot be demonstrated.

A less obvious benefit is access. Consider cutting-edge strategies developed in federated learning networks, cross-institutional research consortia, and clinical data-sharing platforms. Accessing them requires a strong security foundation. Organizations that build that foundation early are eligible for a category of partnerships, data access, and innovation that excludes less-prepared competitors. Thus, the security architecture you build to protect your current workloads determines your stake in tomorrow’s leading practices.

With a demonstrable security posture that hospital IT teams understood, Linde’s cylinder monitoring system could be integrated into hospital operations. Deployed cylinders shifted from tracked inventory to connected infrastructure, embedded in clinical workflows and accessible to nurses, porters, and administrators through role-appropriate dashboards. That integration opened a category of institutional partnership that is structurally unavailable to vendors without verifiable security.

Shaping factor 3: AI strategy

Speed is often the reason to align your adoption architecture with your AI ambition. You want to deploy models faster, train them on real-world patient data, and launch new capabilities without platform retrofits. Engaging your Chief Technology Officer and Chief AI Officer to make decisions at adoption time about data flows and processing pipelines advances AI advantages faster than treating AI as a future phase, when cloud architecture becomes a ceiling rather than a launchpad.

But a more powerful quality of a well-designed AI architecture is that it keeps generating discoveries. Real-world patient data flowing continuously through a system that learns from it surfaces clinical patterns that no algorithm anticipated. Those patterns suggest new product features, new intervention points, and new service models. Thus, the architecture you built for known AI goals becomes the engine that produces the next set of goals. This is not a side effect of getting your AI strategy right; it’s the whole point of a strategy.

Coupling Linde’s cloud adoption with an AI strategy turned continuous cylinder data streams into a learning system. Usage signatures surfaced that predicted depletion rates before they became shortfalls. Early indicators of maintenance needs created conditions for predictive maintenance contracts, a new revenue stream that emerged from data that had always flowed but had never been captured before cloud adoption.

Shaping factor 4: Partner ecosystem

Business innovation emerges from combined expertise, and a well-chosen partner ecosystem secures execution quality. Choices by your teams responsible for partner and vendor strategy are of utmost importance. A transformation partner brings strategic foresight and governance structure. An adoption partner makes architectural decisions that stand the test of time. A post-adoption partner drives new business value. A managed services partner ensures continued performance as the business grows.

The right partners also bring perspective. They can look at what you have built and see what you have not yet considered. Armed with insights from experience, other industries, and other cloud adoption journeys, they approach your architecture with a different lens. Shaping the right partner ecosystem is not just about gaining cloud capabilities but about discovering what it makes possible.

Examining Linde’s incoming cylinder data revealed that a significant share of high-value consumables was being lost because cylinders were not being returned under rental programs. Solving the loss required a supply chain adjustment and produced a new pricing model that boosted business value. In this case, a post-adoption optimization partner unlocked that value, which had been invisible before connectivity, and made it actionable.

Shaping factor 5: Organizational transformation

Development, operations, security, and commercial leadership must move together in a cloud adoption. To deliver on the intended value of adoption, work with your Chief Executive Officer and Chief Operating Officer to ensure that the underlying talent base matches and commits to the new operating environment.

The true sign of that commitment is a shift in the questions that your organization starts asking. When data genuinely drives decisions, teams begin interrogating assumptions they had never questioned. They ask what customers need rather than what they request; they examine where operational inefficiencies quietly compound; they follow data-driven suggestions for product direction. As such, true organizational change does not lead to a fixed destination but keeps generating new ones.

The organizational shift that followed at Linde was consequential. Linde's sales teams could move from selling cylinders to selling outcomes like fleet uptime, replenishment reliability, and procurement intelligence. Service shifted from reactive dispatch of representatives to predictive management driven by live data. And with real-world usage data flowing continuously from deployed cylinders, product development gained an evidence base it had never had before.

The transformational power of these five factors compounds from their coordinated design. Well-governed data creates the foundation for AI that generates new insights. Security by design earns access to ecosystems that expand what those insights can do. The right partners help you see what the architecture makes possible. An organization that has transformed around data is structured to act on what it discovers. Each factor reinforces the others, and together they shape continued growth opportunities years after go-live.

As for the Linde case, cloud adoption delivered more than just a reduction in the cost of manual cylinder management. When the shaping factors were designed as growth levers rather than compliance checkboxes, Linde gained a new business model, a new revenue stream, a new competitive position, a new digital product category, and a new way of working. This kind of transformation is not simple, and the change is significant. But when operating in a market where connectivity redefines customer expectations, change is no longer optional.

The questions worth asking for a cloud adoption business case

So, if a cloud adoption decision lands on your desk framed as a payback calculation, the right response is to ask more questions. Cost analysis is where the conversation starts, not ends. The true cloud adoption ROI is not captured in a two-year payback model.

Your aim should be to amplify the value the investment creates through operational efficiency, novel cost-effective operations, and new revenue that could not be supported before adoption. Capture those opportunities by deliberately exploring the five shaping factors for each workload to be enhanced with cloud capability and designing them as growth levers.

That exploration changes what gets put on your desk. You no longer work with a cost model, but with the architecture of tomorrow: the data platform that reveals new products, the security posture that opens new partnerships, the AI infrastructure that keeps learning, the partners who open new perspectives, and the organization that acts on what all of it reveals.

Thinking through and planning for these shaping factors is a step back from the adoption-as-a-cost exercise and toward treating the adoption as a once-in-a-decade opportunity to shape the future of your business. Organizations that treat cloud adoption as a transformation, not a transaction, are the ones that look back on it as a turning point.

Cloud