Manufacturing went through a massive push to move everything to the cloud. Then it ran into the physical realities of the factory floor. Production workloads need millisecond-level responsiveness to function properly. You cannot tolerate a round trip to a distant data center for critical processes. Legacy operational technology sits at the edge of your network instead of in a centralized availability zone.
This reality forces a reset for IT leaders across the sector. The conversation is shifting from cloud-first to cloud-smart, as you must place workloads where they perform best based on latency and cost realities rather than architectural convention. Most infrastructure has simply not caught up to that shift. This guide will help you understand why smart manufacturing initiatives stall and how optimizing your hybrid infrastructure provides the exact foundation you need to scale successfully.
The pilot trap
The numbers tell a clearstory about the state of modernization. A 2025 Rockwell Automation report reveals 95% of manufacturers are using or evaluating smart manufacturing technology1. That sounds like incredible momentum. But only 20% have scaled beyond the pilot stage¹. Another 56% remain stuck in piloting.
The main barriers are governance gaps, data readiness issues and infrastructure inconsistency. Infrastructure inconsistency is the hardest problem to solve retroactively. You can fix a governance model or clean up a data architecture, but if your underlying infrastructure was built for a different operating model, every workload you try to scale runs into structural friction.
Leading manufacturers with integrated data and technology foundations implement new artificial intelligence use cases significantly faster. A 2023 McKinsey and World Economic Forum report shows some of these leaders skip the pilot stage entirely2. If you cannot achieve that integration, you remain stuck and unable to realize your expected cost and performance outcomes.
The real problem with multi-cloud
Most global manufacturers run several cloud environments simultaneously. You might have private infrastructure at some sites, public cloud from multiple providers and edge computing close to production lines. Each environment has its own management console and visibility gaps.
This fragmentation means you make workload placement decisions based on convention or cost assumptions instead of actual performance data. No one has a unified view of how your full estate is behaving. Operations see one slice. IT sees another. Finance sees a third.
This is where the cloud conversation goes wrong. People frame it as a choice between private and public environments. It is actually an infrastructure design question. You must ask where each workload performs best and who has visibility across all of it.
Moving from cloud-first to cloud-smart
GTT approaches this challenge by placing workload performance first. Workload placement should follow performance requirements. You need hybrid infrastructure designed around what each workload actually needs; this includes latency, data sovereignty, availability and cost.
Cloud Connect, as part of Managed Hybrid Cloud, delivers private, high-bandwidth interconnects to AWS, Azure, Google Cloud and Oracle Cloud. This eliminates the performance variability that comes with routing production-critical workloads over the public internet. Managed SD-WAN provides cloud-optimized routing at every site.
The GTT Envision platform provides a unified view of performance, cost and utilization across edge, private and public cloud. Every stakeholder works from the same picture. GTT Envision is a platform that gives you true control over your network. GTT manages the full lifecycle so your internal IT teams avoid absorbing the orchestration headache of running a multi-cloud estate on their own.
Infrastructure is your differentiator
There is a reason 80% of manufacturers have not scaled smart manufacturing beyond pilots. It is an infrastructure problem dressed up as a technology issue. The manufacturers pulling ahead solved the infrastructure question first and then built on top of it. That sequence matters more than which cloud platform or AI tool you choose.
Take the next step
We put together a solution brief detailing how manufacturers are making this shift. It includes verified customer outcomes and practical steps for network optimization. If you are stuck in the pilot to production gap, read our Cloud-Smart Factory Innovation solution brief to learn how to scale your operations effectively.
1 Rockwell Automation, “State of Smart Manufacturing Report 2025.”
2 McKinsey & Company and World Economic Forum, “The Next Frontier of Operational Excellence: AI at Scale in Manufacturing,” 2023.
