Beyond the Pilot: Three Ways for Benelux Manufacturers To Scale AI for Growth
27 April 2026
10 minute read
To take artificial intelligence beyond isolated experiments and create enterprise-wide impact, manufacturers need stronger governance, clearer objectives and a more iterative approach, says Raf Ganseman, Partner for Uvance Wayfinders in Benelux.
Manufacturers across Belgium, the Netherlands and Luxembourg (Benelux) face factory closures, declining investment, rising energy costs, fragile supply chains and sustained pressure from global competitors. With the odds stacked against them, can they reach their growth ambitions?
Many manufacturers are turning to AI for solutions, and for good reason. Implemented well, AI can increase production, optimize complex processes (such as supply chain management) and even help to improve the quality of products and services.
But most manufacturers are stuck in pilot mode: isolated AI experiments within a single function that fail to scale across the enterprise. Teams start implementing use cases that are too complicated, with no structured approach or central vision. This leads to delayed, paused or even abandoned projects that remain within a single business unit rather than embedded across the organization.
Here are three ways to move out of pilot mode and scale up AI effectively.
“AI can extend across limits and help companies get to the next level. But, on average, the manufacturing industry in Benelux is struggling to move from pilots to production-ready use cases, which suggests it needs a more structured approach.”
1. Use a hub-and-spoke approach
“It’s crucial to work on AI from a hub-and-spoke model, where you have centralized governance but allow business units to run at their own pace.”
Most manufacturers are employing an approach to AI that enables business units to experiment but is disjointed, with no clear goals or guidelines. To implement and scale AI successfully across the organization they need a more centralized approach. Hub and spoke, for instance, is where there’s a central team that upholds the organization’s AI vision, educates the workforce on different AI tools and use cases and decides what the organization should aim to achieve. Then each business unit takes ownership of a use case that fits its purposes.
Each department will have its own set of goals and challenges: the reasons for HR to use AI will be different from the finance function’s reasons, for instance. And AI skills in IT, for example, will be far more advanced than would be in, say, sales. The hub-and-spoke approach enables these different business units to implement AI at their own pace while staying connected to the rest of the organization through shared resources and a roadmap that’s clearly defined from the top.
It’s also a way to mitigate regulation risk when different regions are operating under different rules. The EU AI Act, for example, has created a risk-based compliance framework that has no direct equivalent in the US or China. A hub-and-spoke approach can help overcome this challenge by allowing manufacturers to treat regional entities as separate business units — maintaining central AI governance while enabling each region to advance initiatives within its own regulatory context. Non-European manufacturers, for instance, tend to have Europe-specific AI initiatives or even separate IT systems for their European entities.
2. Build momentum and measurable impact
Manufacturers in Benelux must resist the temptation to start with overly ambitious, complex AI projects that attempt to solve everything at once. Often, these initiatives fail to generate visible success in a short amount of time, and therefore struggle to take off. Instead, manufacturers should start with simple AI initiatives that are easy to implement and produce value quickly. Early “quick wins” create momentum, build confidence and help the wider organization to see tangible impact.
We support this approach through a structured “AI Factory” model. It identifies and prioritizes use cases based on impact, risk and complexity, defines metrics upfront, and delivers a compliant minimal viable product within weeks – enabling organizations to demonstrate measurable value before scaling further.
The value of AI initiatives can be hard for manufacturers to quantify. It’s not just about how much they will cost to launch and operate, and how much time is saved – it’s also about how much they will increase the quality of the product and help to mitigate risks. By starting simply, manufacturers can set realistic KPIs, which help to demonstrate AI’s impact and show how to scale it successfully across the business.
“ROI isn’t just financial. You need to look at quality, productivity, risk and process metrics as well, and weigh them together when selecting your AI use cases.”
3. Prepare for change
Starting simple doesn’t mean thinking small — it means sequencing ambition. Early wins create the capability and confidence needed to take on more advanced AI applications. This process is essential given that among all the challenges manufacturer’s face, the underlying issue of AI is the pace of the technology’s evolution.
Organizations have progressed from classical AI and Generative AI to more advanced applications in quick succession. There’s agentic AI, for instance, which unlocks opportunities for manufacturers to automate entire processes, giving operators on the factory floor the superpowers they need to bolster production. Physical AI is the next frontier: it embeds AI systems in machines, such as robots, vehicles and industrial equipment, that enables them to perceive, decide and act in the physical world with increasing autonomy. For manufacturing, this marks a shift from rigid automation to adaptive systems where machines can handle variability and optimise operations in real-time.
A pharmaceutical manufacturing company operating in Belgium, for example, is using agentic AI to reduce downtime in its production process. AI agents flag anomalies at different stages of the production line, and these can then be dealt with by operators. The agents are connected to each other, so operators and agents can learn from downtime events and continuously improve.
This last stage is important, because it embeds flexibility in the manufacturer’s approach. With AI constantly evolving, a manufacturer could launch an AI initiative today that becomes obsolete within two months. So in addition to a hub-and-spoke model, it’s important for manufacturers’ approach to be iterative: start simple, test, measure, learn and improve. This cycle embeds flexibility because manufacturers can relatively easily incorporate new AI technologies at the “improve” stage.
“Organizations have progressed rapidly from classical AI and Generative AI to more advanced applications such as agentic AI. My advice? Adopt an iterative approach to give yourself flexibility.”
Rethink your AI approach
To use AI for growth and scale it successfully, manufacturers in Benelux need to:
-
Centralize their approach
A decentralized approach takes innovations only so far, and often leaves them stuck in pilot mode within a single business unit. With a hub-and-spoke approach, you can make sure that AI is helping every unit to work toward a shared goal while still operating at each one’s pace.
-
Take one step at a time
It’s easy to get excited by everything AI can do. But remember that without initial quick wins, your teams could lose confidence in the technology and stop your organization from getting to the next level of AI.
-
Become more flexible
Technology is constantly developing, and fast. So adopt an iterative approach to make sure you can implement new AI tools quickly and in a way that works for you.
AI can offer manufacturing companies in Benelux much-needed growth opportunities. To make the most of its potential, manufacturers will need to rethink their approach to AI governance — starting with a vision from the top and a simple idea to build a strong foundation from which to implement more advanced solutions.
Raf Ganseman
Partner, Uvance Wayfinders in Benelux
Raf has extensive consulting experience from EY, Atos, and CGI, having held senior leadership roles driving advisory growth and technology-enabled transformation. With expertise in AI, data, and digital strategies, particularly in manufacturing, the public sector, and financial services, Raf is passionate about leveraging innovation for sustainable business models. In his current role, Raf focuses on the Benelux team, building local advisory practices, shaping the consulting portfolio, and driving strategic initiatives for clients. He holds a Master's Degree in Economics from Katholieke Universiteit Leuven alongside a Post Graduate Diploma in Big Data from Ghent University.
Redesigning security through white-hat hacking
Driving Japan's digital future: Fujitsu's full-stack development for sovereign platform
Contact us