Ceres Power H1 2025: Doosan kicks off mass production, setting the stage for high-margin royalties amid a revenue dip but steady cash position.
This article covers information on Ceres Power Holdings plc.
LON:CWRLast updated:
Ceres Power has posted its interim numbers for the six months to 30 June 2025. The headline story is strategic rather than financial: Doosan has started mass production of fuel cell stacks based on Ceres technology, setting up the first royalty stream under its licensing model. Financially, revenue and profit are down year-on-year, but cash is steady and working capital discipline is clearly tightening.
| Metric | H1 2025 | H1 2024 |
|---|---|---|
| Revenue | £21.1 million | £28.5 million |
| Gross profit | £16.6 million | £22.9 million |
| Gross margin | 79% | 80% |
| Adjusted EBITDA loss | £11.3 million | £9.0 million |
| Operating loss (before exceptionals) | £17.2 million | £13.8 million |
| Cash and short-term investments | £104.1 million | £126.1 million |
| Net cash from operating activities | £1.0 million | Outflow of £13.2 million |
| FY2025 revenue guidance | Around £32 million | Not disclosed |
In July, Doosan became the first partner to begin mass production of products using Ceres’ solid oxide fuel cell (SOFC) technology in South Korea. That is a clear inflection point. Under Ceres’ asset-light, licensing model, royalties should follow Doosan’s first commercial sales. Early target markets include AI data centres, grid stabilisation, buildings and marine auxiliary power.
Why it matters: royalties are high margin and recurring. They are the key bridge from R&D-heavy development to scalable, profitable growth. Timing of those sales is not disclosed, but this validates the model and the intellectual property strategy.
Ceres also reported the first hydrogen production at its megawatt-scale solid oxide electrolyser cell (SOEC) demonstrator with Shell in Bangalore. The module efficiency is cited at 37 kWh/kg of hydrogen, with potential output up to 600 kg per day at full capacity during a five-year operational phase.
Why it matters: SOEC is Ceres’ hydrogen bet. It is later-cycle, but efficiency advantages over PEM and alkaline can be compelling when scaled. With final investment decisions in hydrogen still slow in parts of the world, credible demonstrations like this help build the case for future adoption.
Delta Electronics has committed approximately NT$6.95 billion (about £170 million) to land and factory facilities in Taiwan for large-scale manufacturing, including on Ceres’ solid oxide technology. Delta is already applying Ceres’ SOFC in microgrid pilots and is deeply embedded in data centre power and cooling systems.
Why it matters: it is another external validation and a pathway to another long-term manufacturing and royalty channel. The cadence of Delta’s ramp is not disclosed, but capital is being put to work.
Revenue fell 26% to £21.1 million, as expected after a one-off licence revenue uplift in H1 2024 from the Delta deal. Hardware revenue actually grew to £6.6 million (H1 2024: £3.2 million), but engineering services and licences fell to £14.5 million (H1 2024: £25.3 million). Asia dominated, contributing £18.2 million of revenue, with Europe at £2.8 million and North America negligible at £39,000.
Gross profit decreased to £16.6 million, but margins stayed very healthy at 79%, highlighting the capital-light licensing model. Operating costs before exceptionals fell 6% to £35.6 million, reflecting 2024’s cost rationalisation. Even so, the adjusted EBITDA loss widened to £11.3 million, and the reported loss after tax was £19.6 million.
Cash is a bright spot. Cash and short-term investments edged up to £104.1 million, and net cash from operations was £1.0 million, helped by working capital management and the receipt of an R&D Expenditure Credit (RDEC) due at year end. Contract liabilities rose to £13.5 million, which represents deferred revenue – a constructive indicator of future delivery.
Ceres is rolling out a business transformation programme to align the organisation to a commercial phase. The plan includes a “single product” approach via a dual-purpose stack platform for both power and hydrogen, targeted for launch next year, and a sharper commercial focus across priority geographies.
Phase one – the operational restructuring and resource realignment – is slated to complete by the end of 2025. Going into 2026, the company expects operating expenses to be about 20% lower than for the year ending 31 December 2025. That, coupled with manufacturing partners scaling up, is the route toward breakeven.
The Board now expects full-year 2025 revenue of around £32 million. Management is in later stage negotiations for a new manufacturing licence agreement, but completion and revenue timing are uncertain. If it lands and any revenue is recognised this year, that would be incremental to the £32 million guidance.
Read that as conservative near-term revenue, with optionality if licence activity crystallises. The real needle-mover, however, will be royalties from partners’ commercial sales, particularly Doosan.
Management is leaning into the fastest-growing power market: AI-driven data centres. With grid connection delays stretching years in some regions, 24/7, high-efficiency on-site power is in demand. Solid oxide fuel cells can offer rapid time-to-power, high efficiency and modular deployment. Ceres also flags attractive pockets in microgrids, combined heat, power and cooling for buildings, and marine auxiliary power – with policy tailwinds like the 30% Investment Tax Credit under Section 48E in the US, and supportive regimes in South Korea and Japan.
In short, power now, hydrogen later. That sequencing makes sense given current market conditions.
Ceres is shifting from promise to product. The numbers show the growing pains of that transition – lower revenue, wider losses – but the operational milestones are the most important part of this RNS. With Doosan live, Delta investing, Shell producing hydrogen at class-leading efficiency and a cost base reset under way, the ingredients for the next phase are falling into place. The near-term test is turning those ingredients into recurring royalties and steady licence income.
Related
Polar Capital Technology Trust sees 102% NAV growth in FY2026, beating its benchmark by 47 points thanks to AI and semiconductor exposure.
JoshuaJuly 10, 2026
Impax Q3 AUM rises to £23.3bn despite £1.7bn net outflows, driven by market gains and strong investment performance.
JoshuaJuly 10, 2026
MJ Gleeson FY2026 trading update: steady profits, mixed home sales with operational restructuring improving outlook.
JoshuaJuly 10, 2026
Last updated
Category
InvestingViews
48 viewsLikes
No ratings yet
No comments yet - start the conversation.