For more than a decade, industrial groups have proclaimed their ambition to become software companies. Billions were invested, platforms were launched, and digital strategies announced with great conviction. Yet progress has been uneven. Many initiatives stalled, fractured, or quietly rebranded. The usual explanations point to technology: immature platforms, legacy IT, customers that were not ready, or AI that has yet to deliver. These explanations are comforting — and largely wrong.
The problem was never software. It was governance.
The first phase of industrial digitization was built around the idea of horizontal IIoT platforms. The vision was compelling: one universal layer to connect machines, data, applications, and customers across the enterprise. Large incumbents sought to replicate the scale economics of enterprise software in an industrial setting. What they underestimated was how deeply industrial organisations are structured around autonomous business units, each with its own architectures, incentives, pricing logic, and customer relationships. Integration was costly, adoption voluntary, and value creation diffuse. Platform teams optimised for architectural elegance while budgets and accountability remained resolutely vertical.
As economic and operational pressure mounted, this tension became impossible to ignore. Business units gravitated back to what they controlled and what customers were willing to pay for. They prioritised solutions tied to specific workflows, assets, and outcomes. The universal platform did not disappear, but it was re-scoped. In many cases it became an integration backbone or architectural framework — necessary infrastructure, but no longer the growth engine it was once imagined to be.
Value creation moved upward, closer to business ownership. Vertical software — manufacturing execution, asset performance management, simulation, digital twins, and optimisation — began to attract real budgets and executive attention. Not because it was more sophisticated, but because it aligned with how industrial customers buy, govern, and measure value. The IIoT platform vision did not fail. It collided with organisational realities it was never designed to overcome.
The industrial groups that made the most progress recognised this early and responded not with better technology, but with better structure. Instead of forcing software to mature inside hardware-dominated organisations, they separated governance while preserving domain depth. Software assets were consolidated, given independent operating models, and allowed to scale with their own P&Ls, talent strategies, and market narratives. Only once these capabilities had matured were they reintegrated — not as experimental add-ons, but as credible businesses. The transformation was not organic. It was deliberate.
Nowhere is the governance gap more visible than in sales. Industrial companies excel at selling hardware and embedded software through capex-driven, project-based models built on engineering credibility. SaaS inverts this logic. Budgets shift to opex, sales cycles shorten, and success is measured after the contract is signed through adoption, retention, and expansion. Value must be articulated in business outcomes, not specifications. The economic buyer often moves from the plant to central IT or digital leadership. Overlaying SaaS onto legacy sales organizations without changing incentives, ownership, and accountability creates friction that no amount of product quality can resolve. What looks like a technology problem is almost always a commercial one.
These tensions ultimately surface as an organisational question that no industrial leader can avoid: should software amplify the core hardware business, or should it become a core business in its own right? Some companies embed software tightly within divisions, preserving proximity to customers but limiting scale and speed. Others establish corporate software units with independent governance, enabling focus and valuation clarity at the risk of cultural distance. Still others keep software fully embedded, optimising execution while constraining growth. None of these choices is inherently right or wrong. But avoiding the choice altogether is costly.
This is why industrial leaders, despite similar rhetoric, occupy very different positions on the software transformation journey. The differentiator is not technological sophistication, but the degree to which software has been granted real autonomy: decision rights, capital allocation authority, and a distinct go-to-market model. Where these conditions exist, software scales. Where they do not, it remains constrained by hardware logic, regardless of ambition.
The current enthusiasm around artificial intelligence risks repeating the same mistake. AI will not rescue industrial digital strategies that lack coherent data models, clear ownership, and aligned incentives. AI amplifies what already exists. Without governance, it accelerates fragmentation rather than resolving it. Industrial companies do not have an AI problem. They have a data and decision-rights problem.
The next phase of industrial software will therefore not be defined by platforms, clouds, or algorithms. It will be defined by governance. Who owns software P&Ls. Who controls roadmaps and capital allocation. Who carries accountability beyond the initial sale. Companies that address these questions explicitly will build durable software businesses. Those that do not will continue to diagnose technological symptoms while leaving organizational causes untouched.
The future of industrial technology will not be built by connecting more machines. It will be built by connecting software strategy to organizational power.
No comments:
Post a Comment