1. Investment is growing. Actual deployment remains selective.

According to the Hydrogen Council’s Global Hydrogen Compass 2026, committed clean hydrogen investment reached USD 130 billion worldwide, including approximately USD 30 billion in Europe.

Yet Hydrogen Europe’s Clean Hydrogen Monitor 2026 reports only 677 MW of operational water electrolysis capacity in Europe by July 2026, with another 3.7 GW under construction.

Progress is real. But the gap between announced ambitions and operating installations remains substantial.

As we observed at the World Hydrogen Summit 2026, projects increasingly depend on credible offtake, financing, electricity supply and implementation plans.

The market is not simply growing. It is becoming more selective about what gets built.

2. Industrial demand must shape hydrogen production.

Brussels highlighted the importance of connecting hydrogen production, infrastructure and end users.

Integrated hydrogen hubs and valleys offer promising opportunities, particularly around ports and industrial clusters. But proximity alone does not guarantee commercial viability.

Existing industrial applications offer another perspective. Where hydrogen is already an essential process input, on-site production may provide advantages in logistics, supply continuity and process integration.

The choice between local production and centralised supply depends on actual demand, electricity costs, operating requirements and available infrastructure.

Hydrogen production must fit the industrial application, not the other way around.

3. Electrolyzer prices are falling under pressure. But what is the real cost?

Hamburg reflected an increasingly competitive electrolyzer market, with global manufacturing overcapacity and strong Chinese price competition.

Lower prices can improve project economics. But comparisons based solely on equipment cost per kilowatt are often misleading.

Does the quoted price include the complete system? What about power conversion, purification, cooling, controls and commissioning?

More importantly, what are the implications for electricity consumption, availability, maintenance and component lifetime?

An inexpensive electrolyzer may become an expensive production asset if operating costs or unplanned downtime erase the initial savings.

The meaningful comparison is the cost of delivering usable hydrogen reliably over the equipment’s operating life.

4. Supplier continuity is becoming an industrial risk.

Several electrolyzer manufacturers have faced restructuring, financial difficulties or market exits as order volumes failed to meet earlier expectations.

For industrial customers, this raises an uncomfortable question.

What happens when an electrolyzer outlives its manufacturer?

An installation may require specialised components, software, replacement stacks and technical expertise for many years.

Without access to these resources, even technically sound equipment may become difficult or expensive to maintain.

Supplier selection should therefore consider serviceability, spare-parts availability, technical documentation and credible long-term support arrangements.

Industrial reliability extends beyond equipment design. It also depends on keeping that equipment operational throughout its lifetime.

5. Performance needs proof, not just specifications.

A particularly relevant technical development in Hamburg was the attention given to comparable electrolyzer performance testing.

Bureau Veritas addressed standardised performance verification, while a GroenvermogenNL / TNO workshop explored harmonised test protocols, with Exion Hydrogen contributing its field experience.

The challenge is straightforward: manufacturers may report efficiency, durability and hydrogen output under different conditions and system boundaries.

But industrial installations operate beyond nominal conditions.

Start-up, shutdown, varying loads, water quality, thermal management and gas handling all influence actual system performance.

For Exion Hydrogen, this reinforces the importance of complete system engineering, validation and long-term operating data, rather than relying on individual stack performance figures.