Sunflower 2025: Asset Sales Bought Time, but Growth Still Needs Proof

Company Introduction

By the end of 2025, Sunflower is no longer a US renewable-development story, and it is not a clean organic-growth story either. It is a relatively small listed power platform with two real engines: the wind and solar portfolio in Poland, and the rooftop and energy-services layer in Israel. What is working now is not necessarily growth, but balance-sheet cleanup. The US activity is gone, Israeli assets keep getting monetized, debt has fallen sharply, and by the time the annual report was published, parent-level debt had already dropped from about NIS 21.7 million to about NIS 5 million after another early repayment in January 2026.

That is also why a superficial read can be misleading. A reader who sees NIS 26.5 million of net profit in 2025, and NIS 55.5 million of net profit in the fourth quarter, could easily conclude that Sunflower is already back on track. That would be the wrong takeaway. The bottom line was inflated mainly by roughly NIS 60 million of gain on the Prime sale, while the recurring engine weakened: revenue fell to NIS 140.2 million from NIS 174.6 million, gross profit fell to NIS 57.5 million from NIS 94.6 million, EBITDA fell to NIS 100.7 million from NIS 131.1 million, and FFO fell to NIS 60.9 million from NIS 83.7 million.

What still supports the story is the combination of cash-generating assets in Poland and a much less stretched balance sheet than a year ago. What blocks a cleaner thesis is the question of whether the company is rebuilding an operating base, or simply converting existing MW into cash. During 2025 it sold 13.1 MW in Israel, after year-end it signed another 4.13 MW disposal, it shut down the entire US activity, and the attempt to acquire Afcon Renewable Energies did not reach binding terms. That is why 2026 currently looks less like a breakout year and more like a bridge year: cash comes in, debt goes down, but proof of the next growth leg is still missing.

There is also a practical actionability filter. The current market cap stands at about NIS 393.2 million, but the latest trading-day turnover was only about NIS 116 thousand. This is not a forgotten stock, but it is not a highly liquid one either. Add a project-finance structure that limits upstream cash movement from subsidiaries, and the result is a thesis in which value created is not automatically the same as value immediately accessible to common shareholders.

The Economic Map

Events and Triggers

Trigger one: the Prime transaction is the main accounting and balance-sheet event of the year. In December 2025 the company completed the sale of Helios Systems K.S., which held operating Israeli solar systems with total capacity of 13.1 MW, for about NIS 76.8 million. About NIS 47.8 million was received at closing, and in February 2026 the remaining NIS 29 million was paid together with roughly NIS 473 thousand of accrued interest plus VAT. Sunflower booked about NIS 60 million of gain, and the buyer also stepped into guarantees of roughly NIS 128 million. This is a balance-sheet-cleaning move, but it also reduces the operating asset base.

Trigger two: another Israeli disposal was signed after year-end, this time for operating systems totaling 4.13 MW, for about NIS 36.4 million. The company expects roughly NIS 20 million of net free cash flow from the deal, although about NIS 2.1 million of the consideration is deferred until two systems totaling 0.711 MW are connected and required approvals are received. This further improves financial flexibility, but again shifts the story from operating to monetizing.

Trigger three: the US chapter is effectively over. In the second quarter of 2025 the board decided to stop US development activity because of elongated development timelines, delayed grid connections, policy changes, and the lack of binding offers. During 2025 the company recorded NIS 21.8 million of other expenses, including NIS 19.1 million of US investment impairment, and also wrote off about NIS 7 million of deferred tax asset. On the other hand, the company released guarantees and deposits, so the US exit hurt earnings but helped liquidity.

Trigger four: in the first quarter of 2025 Sunflower acquired two Polish subsidiaries holding 19.5 MW of operating solar assets for about NIS 65 million including transaction costs. That move is meant to replace part of what was sold in Israel, but at the report date those assets were still not project-financed. The company says it is in advanced negotiations with a Polish bank for about PLN 30 million of financing, but with no certainty of completion. So the acquisition improves the portfolio profile, yet for now sits more heavily on equity.

Trigger five: the inorganic growth leg remains unproven. On December 25, 2025 the company extended the memorandum of understanding with Afcon to complete due diligence and move toward a detailed agreement, but by the report date it already stated that the process did not mature into binding terms. That matters because management still talks about a strategic goal of roughly 1,000 MW of operating and ready-to-build assets by 2030, but for now that goal is backed more by aspiration and early pipeline than by signed deals.

That chart makes the point clearly: the fourth quarter does not show an operating breakout. It shows how a disposal gain can overwhelm a softer recurring revenue and gross-profit picture.

Efficiency, Profitability and Competition

What Actually Weakened in 2025

The operating heart of the 2025 story is Poland. Segment revenue fell to NIS 65.7 million from NIS 103.3 million, and segment result fell to NIS 28.5 million from NIS 70.2 million. That is not only a wind-resource issue, even though annual wind output was down about 8%. The deeper point is price quality. Management explicitly says Polish revenue was hurt by updated electricity-sale agreements for 2025-2026 and by weaker green-certificate prices, even though the average Polish market power price actually rose from PLN 423 to PLN 447 per MWh. In other words, knowing what happened to the market price is not enough. You have to know what economics the company locked into its contracts.

That matters because Sunflower’s Polish economics are shaped by the exact contract profile, indexation, hedging, and revenue mix, not by the broad direction of headline market prices alone. The average green-certificate price fell from PLN 46 to PLN 28 per MWh, and the agreement covering about 70% of green certificates expired at the end of 2025. So even if black-power pricing remains decent, the certificate layer is thinner.

Israel Looks Strong, but Mostly Because of Monetization

In Israel the picture is the opposite. Revenue rose slightly to NIS 74.5 million from NIS 71.4 million, but segment result jumped to NIS 81.9 million from NIS 41.1 million. That looks excellent until it is decomposed. The main driver was not exceptional organic improvement, but the gain on sale of the Prime transaction. Anyone looking at segment result alone could easily overread the Israeli business.

That does not mean there was no operating progress. There was. Sunflower Energy Services completed the connection of about 7 MW during 2024-2025, with about NIS 22 million of investment, and the company moved in May 2025 to in-house operation of the Israeli systems rather than outsourcing the full management layer. But relative to the disposal gain, those are still smaller drivers. At this stage Israel contributes to the thesis more through capital flexibility than through clean operating acceleration.

Earnings Quality and Competitive Position

EBITDA fell to NIS 100.7 million from NIS 131.1 million, Adjusted EBITDA fell to NIS 103.5 million from NIS 136.0 million, and FFO fell to NIS 60.9 million, or NIS 75.4 million under management’s presentation. That is the more useful measure than net profit because it strips out the disposal effect. It says, quite simply, that the recurring business weakened.

There is also concentration that is easy to miss. One major customer linked to the Polish segment accounted for about NIS 62.7 million of revenue in 2025. That is almost the entire segment. The company does not disclose the name, so it is hard to judge whether this is broker concentration, a single PPA buyer, or another type of counterparty. In Israel, the segment is also structurally concentrated around the relevant electricity counterparties. That can be normal in power generation, but when names are undisclosed, it becomes harder to assess contract quality and bargaining power.

Cash Flow, Debt and Capital Structure

The Cash Lens: all-in Cash Flexibility

This is a case where the right frame is all-in cash flexibility, because the thesis is about financing capacity rather than pure earning power. Cash flow from operations fell to NIS 62.5 million from NIS 81.4 million. At the same time, investing cash flow moved from positive NIS 9.0 million to negative NIS 26.4 million, mainly because of the Polish solar acquisition and continued buildout and upgrades in Israel. Financing cash flow was negative NIS 49.1 million because of debt repayments, even if partly offset by new financing in Poland and in the energy-services platform.

The critical point is not only that operating cash flow remained positive, but that it did not fully cover both the investment layer and the financing layer. That is why the cash balance decline was manageable rather than dramatic, but it did not disappear. This was not a year in which a self-funding operating machine comfortably financed growth. It was a year of capital recycling: disposals, a US exit, new asset purchases, and debt repayment.

Add the NIS 12.2 million of lease cash paid during 2025, and there is another reminder that even after the debt burden eased, real cash commitments did not vanish.

Cleaner Balance Sheet, but Not All Cash Is Free

The good news is obvious. Net financial debt fell to NIS 131.4 million from NIS 277.1 million. Financial liabilities fell to NIS 280.0 million from NIS 467.1 million. Current assets stood at NIS 194.2 million against NIS 65.2 million of current liabilities, implying roughly NIS 129.0 million of working capital. At the parent level, the picture is also much calmer: standalone bank debt fell to NIS 21.7 million at December 31, 2025, and after another early repayment in January 2026 it dropped to about NIS 5 million.

But that is only half the picture. The group relies heavily on non-recourse project finance, and the report explicitly says that distributions from subsidiaries are subject to coverage ratios, financial covenants, and restrictions on dividends, shareholder-loan repayment, and excess-cash transfers. So an improvement in consolidated net debt does not mean every shekel sitting in project entities is immediately available to common shareholders.

Covenant pressure, for now, does not look acute. In Polish wind project companies, management says all entities meet debt-service coverage and all financial covenants. At the parent level, the ratio between distributions received and current maturities stood at 4.53 against a minimum of 1.2, while equity stood at about NIS 276 million against a NIS 140 million floor. That is no longer a close-call balance sheet.

What the Disposals Improved, and What They Also Weakened

The more interesting point is that the balance-sheet repair was achieved by shrinking the asset base. Photovoltaic generation assets fell to NIS 222.6 million from NIS 326.6 million, right-of-use assets fell to NIS 50.3 million from NIS 67.6 million, and other non-current assets fell by NIS 49.0 million. In other words, the company became lighter in debt terms, but also smaller in asset and cash-earning terms. That is good if the proceeds are being recycled into a better portfolio. It is less good if monetization is running ahead of replacement.

Outlook

Finding one: 2026 currently looks like a bridge year, not a breakout year. The existing operating base is expected to generate revenue of NIS 86.5 million to NIS 95.6 million in 2026, with EBITDA of NIS 67.1 million to NIS 74.2 million. That forecast is for the operating portfolio already in view, and mainly shows how much the company still needs both to preserve what it has and to add a new layer.

Finding two: Poland remains the core of the story. The wind portfolio is expected to generate NIS 51.4 million to NIS 56.9 million of revenue in 2026 and NIS 36.4 million to NIS 40.3 million of EBITDA, while the acquired Polish solar portfolio is expected to add NIS 4.5 million to NIS 5.0 million of revenue and NIS 2.7 million to NIS 3.0 million of EBITDA. That means most of the visible recurring earnings power is still there.

Finding three: Israel will still contribute, but for now more through project connections and capital management than through a sharp step-up in results. The rooftop platform is expected to deliver NIS 23.8 million to NIS 26.3 million of revenue and NIS 22.3 million to NIS 24.7 million of EBITDA, while the remaining operating layer in Israel is expected to add NIS 6.8 million to NIS 7.6 million of revenue and NIS 5.6 million to NIS 6.2 million of EBITDA. That is a decent base, but not one that can carry the whole growth case on its own.

Finding four: the issue is not only development speed, but path cleanliness. One Polish project with 75 MW of solar and 300 MWh of storage potential is affected by a land-owner breach, and the company says it cannot yet estimate the consequences for further development. At the same time, the possible acquisition of roughly 48 MW of solar assets is still only under negotiation, not signed.

What the Company Already Knows About 2026

That table does not promise rapid growth. It mostly shows what is already in sight. Anyone looking for upside beyond that has to look at the still-unproven layer: 16.4 MW under construction and ready to build in Israel, 208 MW of solar and 600 MWh of storage in Poland, the Cable Pooling expansion, and possibly further acquisitions.

What Must Happen Over the Next 2 to 4 Quarters

First, the Greenlight sale has to close and convert the expected roughly NIS 20 million of free cash flow into actual cash. Second, project financing has to be closed for the Polish PV portfolio so the newly acquired asset base does not keep sitting largely on equity. Third, Israeli projects need real grid connections, not just pipeline language. Fourth, the market needs a cleaner update on Poland, both around the troubled 75 MW land situation and around the possible 48 MW acquisition.

If those things happen, 2026 can end up looking like a successful bridge year on the way to renewed growth. If they do not, the year may look in hindsight like a sequence of asset sales that bought financial breathing room without rebuilding a large enough operating engine.

Risks

The first risk is price and revenue quality in Poland. The green-certificate sale agreement expired at the end of 2025, and the company is still evaluating a replacement agreement. For 2026 it assumes an average certificate price of PLN 25 and an average black-power price of PLN 421 per MWh. If pricing comes in lower, or FX moves the wrong way, Poland will feel it quickly.

The second risk is cash accessibility. Even after debt came down, group cash is not the same as freely available parent cash. Distributions from subsidiaries remain subject to DSCR, dividend restrictions, and financing agreements. Any thesis that says “there is plenty of cash” without asking where that cash sits will miss the point.

The third risk is replacing sold MW with operating MW. Recent disposals fixed the balance sheet, but if the Israeli build layer slips, if Polish financing does not close, or if the 75 MW land issue stays unresolved, the company could end up with a leaner portfolio and delayed growth.

The fourth risk is concentration. In Poland, one major customer accounts for about NIS 62.7 million of revenue and the company does not name it. In Israel, revenue is also structurally concentrated around the relevant power counterparties. That is not necessarily unusual in the sector, but it does mean the contract profile matters more than the broad narrative of demand for electricity.

The fifth risk sits in the market itself. Updated short-interest data show a jump from 9,105 shares on March 20, 2026 to 104,084 shares on March 27, 2026, with SIR of 7.51 and a reported 1.32% short float. That is still not an extreme short setup, but against weak day-to-day liquidity it is enough to raise volatility around new disclosures.

Conclusions

Sunflower at the end of 2025 is a company with a much better organized balance sheet than it had a year earlier, but also with a slimmer operating base and more open questions around the next growth leg. What supports the thesis today is financial cleanup, already-closed monetizations, and a Polish core that still generates cash. What blocks a cleaner read is that this repair came through asset shrinkage, while the replacement layer is still not proven.

Current thesis in one line: Sunflower bought itself time and flexibility, but it has not yet proved that it can turn that flexibility back into clean cash-generating growth.

What changed versus the earlier understanding of the company: the center of gravity has shifted from a multi-geography development story to a capital-recycling, balance-sheet-cleanup, and asset-rebuild story around Poland and Israel.

The strongest counter-thesis: it is possible that the company has already done the hard part, and that lower debt, the 19.5 MW Polish acquisition, Israeli project connections, and continued monetization will let it enter 2026-2027 with a smaller but better and more profitable portfolio.

What could change the market’s interpretation in the short to medium term: a full Greenlight closing, financing for the Polish solar assets, and a credible update on the 75 MW project and the possible 48 MW acquisition.

Why this matters: the real test is no longer whether Sunflower can create value through disposals, but whether it can turn that value into new assets and cash flow that actually reaches shareholders.

What must happen over the next 2 to 4 quarters: monetization proceeds need to become lasting debt reduction and live projects, Poland needs to show revenue stability even without the old certificate agreement, and Israel needs to connect enough new assets to replace what was sold. What would weaken the thesis is continued pressure on recurring earnings without proof of replacement, or further delays in the development layer.