Demand-First Hydrogen: Avoiding Infrastructure-Driven Policy Capture
When policy becomes the product: why building the pipe before the customers is risky architecture
We applaud big infrastructure announcements because they feel decisive. But there’s a recurring trap-public or private-that’s familiar to enterprise architects: build large, capital‑intensive platforms first, then try to force supply and demand to meet via subsidies, mandates or procurement. That pattern creates fragile ecosystems and institutional lock‑in long before the market proves itself.
What triggered this note: I recently read an analysis of a European hydrogen strategy where repurposed gas pipelines, broad subsidy rules and generous station/vehicle grants create the appearance of momentum-but not the commercial fundamentals beneath it. The policy sequence there-infrastructure first, demand later-turns empty capacity into the rationale for more subsidies, and subsidies into the test of market interest.
Why enterprise architects and CTOs should care
This is not just an energy policy problem. The same dynamics appear in technology programs: oversized data lakes with no ingestion pipelines; cloud regions stood up without workloads; national digital stacks designed without verified user journeys. The architectural lesson is that infrastructure without credible, contracted demand quickly becomes a calcified asset base that benefits institutions, vendors and regulators more than end users.
Key parallels and trade‑offs
- Sunk cost and regulatory returns mimic technical-debt incentives. When an asset is large and sanctioned (or regulated), the stakeholder incentives shift from delivering value to preserving asset value-think “regulated asset base” in utilities vs. “billable platform” in IT.
- Vanity metrics replace operational metrics. Kilometres of pipe, stations opened or nodes deployed are seductive KPIs; utilisation, throughput, delivered cost per unit and repeat purchases reveal true viability.
- Platform‑first accelerates lock‑in. Once capital and governance structures attach to an asset, reversing course is politically and economically painful-similar to switching away from a monolithic on‑prem system after years of integration and vendor dependencies.
- Subsidies and procurement distort product‑market fit. Grants can prove that organisations respond to incentives, not that consumers or downstream businesses will sustain a service post‑support.
Actionable architectural guidance
- Start with demand signals: require contracted off‑take, firm pilot agreements, or measurable user adoption before committing large, irreversible capacity.
- Design modular, staged infrastructure: enable incremental capacity upgrades (horizontal scaling, not a single large “diameter” decision), and prefer interoperable interfaces so unused components can be repurposed.
- Make economic KPIs first‑class: track utilisation, delivered cost per unit, total cost of ownership and willingness‑to‑pay during pilots-and condition further funding on them.
- Build sunset clauses and repricing triggers into public tech projects and procurement to avoid perpetual subsidy cycles.
- Avoid institutional monocultures: diversify governance, include independent auditors and private‑sector offtakers to reduce “institutioneller filz” dynamics.
A note for Indian practitioners and policymakers
This pattern has resonance in large digital and infrastructure programmes I’ve encountered in India. The antidote is the same: run tight, measurable pilots that validate demand (not just procurement), design for frugal scale from the start, and align incentives so vendors and institutions succeed only when users do. In Northeast India-where constrained budgets and unique last‑mile challenges demand efficiency-this discipline matters even more.
Takeaways
- Measure customers, not capacity. Prioritise utilisation and delivered cost over deployment counts.
- Build incrementally and keep interfaces open to avoid stranded assets.
- Use conditional funding and clear KPIs to align institutional incentives with real market demand.
- Treat large public platforms as product‑development exercises: pilot, measure, iterate, scale.
Closing thought
Infrastructure is essential, but architecture is about serving needs-first demand, then capability. If we invert that order, we build monuments to good intentions rather than systems that sustain themselves.
About the Author: Sanjeev Sarma is the Founder Director and Chief Software Architect at Webx Technologies. With a core focus on Generative AI integration, Cloud-Native Scalability, and Enterprise Software Architecture, he has spent over two decades driving digital transformation across Northeast India and beyond. Beyond his corporate leadership, Sanjeev is deeply invested in shaping the future of the IT industry. He serves as an Industry Expert on the Board of Studies for Assam Don Bosco University’s School of Technology, advises state technology committees, and actively mentors emerging tech startups at STPI. He brings a unique, dual perspective of high-level enterprise execution and future-ready academic curriculum development.