Ein Shlishit: Growth Bought Through Thin Margin, Was 2025 a One-Off Trough or a New Cost Structure
Ein Shlishit's revenue did rise to NIS 20.9 million, but cost of sales rose much faster, gross margin fell to 11.4%, and part of the manufacturing effort was pushed into work in progress instead of flowing through the income statement. The real post-2025 question is not only how much the company can sell, but at what economic price.
The main article already set the broader frame: Ein Shlishit has real demand, real orders, and a real move into serial production, but cash, inventory, and credit still sit heavily on the story. This continuation isolates only one question: what kind of growth quality 2025 actually had. Not whether revenue existed, but whether that revenue was built on a sound economic base or through lower-margin selling and cost loading.
The short answer is that 2025 was both. On one side, this was clearly a heavy transition year: management explicitly points to tender-related product development, production-line setup, and preparation for serial manufacturing. On the other side, it also says plainly that the company strategically accepted some orders at lower profitability in order to deepen market penetration and develop new products. That is no longer just a transition accident. It is a commercial choice.
So anyone describing 2025 as a purely one-off trough is missing the element of managerial choice. And anyone rushing to call this a fully reset cost base is missing the fact that part of the effort has not yet been tested through 2026 serial deliveries. The real issue is where the line runs between transitional setup cost and a structurally lower-margin, broader-cost profile.
What Actually Broke
The headline is simple: revenue rose 19% to NIS 20.933 million, but cost of sales rose 56% to NIS 18.553 million. The result was gross profit of only NIS 2.38 million, down 58% from 2024, and a gross margin that fell from 32.2% to 11.4%.
What matters is not only the depth of the erosion, but also its shape. In the first half of 2025 gross margin was about 10.1%, and in the second half it improved only modestly to roughly 12.7%. That is a partial stabilization signal, but nowhere near a return to 2024 profitability. If 2025 were only a temporary distortion, the second half should probably have shown a sharper rebound already.
Management itself gives three explanations, and all three matter. The first is high development expense for a product tailored to a tender the company won. The second is the setup of the production line and preparation for serial manufacturing. The third, and the most important one for growth-quality analysis, is the explicit statement that the company strategically accepts some orders at lower margin in order to deepen market penetration and develop new products.
That is the center of the issue. Once management says directly that it accepted lower-profitability orders to open doors, revenue growth can no longer be separated from its economic cost. This is not only external pressure on margin. It is an active choice to prioritize penetration and commercialization over clean short-term profitability.
Revenue mix changed, and 2025 was no longer the same product basket sold on the same terms
Even without product-level margin disclosure, the mix tells a clear story. MEDUZA remained the largest product family, but its share fell from 74% of revenue in 2024 to 56% in 2025. At the same time ARGOS climbed from 10% to 30%, while CHIMERA fell from 17% to 5%.
The important limitation is worth stating clearly: the filing does not disclose product margin by family, so the erosion cannot be pinned on one product line as a hard fact. But it does prove that 2025 was not another year of the same basket under the same conditions. Sales mix moved, new products were pushed forward, and management simultaneously acknowledged a willingness to take lower-margin orders. Together, that is already enough to mark the year as lower-quality growth even without product-level profitability disclosure.
Part of the Manufacturing Effort Never Hit 2025 Gross Profit
One of the most important clues in the filing does not sit in the gross-margin line itself, but in the cost-of-sales note. Before deducting the increase in work in progress, total 2025 production effort reached NIS 24.121 million. After deducting a NIS 5.568 million increase in work in progress, reported cost of sales landed at NIS 18.553 million.
This does not mean the company "hid" cost. That would be the wrong reading. It means that close to one quarter of the year's production effort, roughly 23%, was pushed forward through work in progress. That detail matters because it means 2025's weak gross margin already benefited from some accounting cushioning. Part of the effort will only be tested once those units convert into recognized revenue.
That also sharpens the 2026 test. If that inventory converts into serial deliveries at better pricing and terms, 2025 can indeed look like a temporary trough. If it converts under similarly thin pricing, or if another cost wave is needed to complete delivery, then 2025 will look less like an outlier and more like a lower steady-state margin level.
The 2024 comparison reinforces the point. In 2024 the increase in work in progress was only NIS 1.335 million. In 2025 it jumped to NIS 5.568 million. So the company not only produced more, it relied much more heavily on carrying part of that effort into future periods.
This Is No Longer Only a Margin Issue, but a Broader Cost Base
To judge whether 2025 was a one-off trough or the start of a new cost structure, the layer underneath cost of sales matters just as much. Here the picture is more complex. Cost did not rise because of one isolated item. It rose across most of the system.
| Item | 2024 | 2025 | Change | Why it matters |
|---|---|---|---|---|
| Salaries and related within cost of sales | 4.450 | 9.416 | +112% | The manufacturing labor base was already resized for a broader phase |
| Raw materials | 7.783 | 12.000 | +54% | Production is consuming more material, not only more engineering time |
| Subcontractors | 0.086 | 0.784 | +812% | A sharp jump that points to execution load and a wider support layer |
| Depreciation within cost of sales | 0.231 | 0.558 | +142% | Equipment and production-line assets are already sitting in the P&L |
| R&D expense | 4.443 | 6.356 | +43% | Product investment did not disappear outside cost of sales |
| Sales and marketing expense | 1.496 | 2.459 | +64% | The company is building the market while also building production capacity |
| Net finance expense | 1.827 | 1.346 | -26% | The core 2025 pressure came from operations and productization, not mainly from financing |
This table matters because it helps separate two different layers. On one side, it confirms a temporary transition burden: the company expanded production capacity, bought testing systems, hired people, and built a support array for larger manufacturing volumes. On the other side, once those costs are already visible in payroll, depreciation, raw materials, and commercial expense, it is no longer possible to pretend they have not changed the company's cost base.
Put simply, 2025 was not hurt only by the act of setting up the line. It also built an organization for the next phase. If revenue does not outrun that base, the company can remain heavier even without another extraordinary event.
The Product-Investment Table Also Says the Pressure Is Not Fully Behind Us
The product-investment table reinforces exactly that reading. In 2025 the company invested about NIS 15.29 million in product development, versus about NIS 9.07 million in 2024. Against that, it shows expected investment of about NIS 5.85 million for the next 12 months.
This point cuts both ways. On one hand, there is logic in the view that 2025 was an unusually heavy year. Total product investment rose sharply, and the number shown for the next year is much lower. On the other hand, it is not dropping to zero. CHIMERA alone still carries NIS 3.6 million of expected investment over the next 12 months, alongside NIS 0.8 million for MEDUZA, NIS 0.4 million for ARGOS, and NIS 0.8 million for fEYE/vEYE.
So the company may be approaching the end of one phase, but it is not exiting a world of heavy investment. Anyone expecting an automatic snap-back to 2024 margin needs to assume not only stronger revenue, but also that the next wave of product and commercialization spending will be moderate enough to absorb less of the gross and operating lines.
So Was 2025 A One-Off Trough Or A New Cost Structure
The more precise answer is that 2025 was a trough year with a structural scar.
It was a trough year because it included clearly temporary or unusually intense elements: tender-specific product development, production-capacity buildout, workforce expansion, and preparation for serial manufacturing. In addition, part of the effort still sits in work in progress and can turn into revenue later.
But it also left a structural scar because the company deliberately accepted some lower-margin work, widened its cost base, and shifted its sales mix. Even if 2025 marked the low point, there is no evidence in the filing that margin simply snaps back by itself. That recovery needs three things together: a clear improvement in serial-order margin, conversion of work in progress without another cost wave, and some moderation in the ratio of development spending to revenue.
This is therefore not really a volume question. It is a conversion question. If 2026 brings more revenue without better economics, the market may find that Ein Shlishit has built a larger business but not necessarily a better one. If, instead, new revenue arrives after 2025 already absorbed most of the setup cost, then 2025 can genuinely be remembered as a one-off trough.
Disclosure: Deep TASE analyses are general informational, research, and commentary content only. They do not constitute investment advice, investment marketing, a recommendation, or an offer to buy, sell, or hold any security, and are not tailored to any reader's personal circumstances.
The author, site owner, or related parties may hold, buy, sell, or otherwise trade securities or financial instruments related to the companies discussed, before or after publication, without prior notice and without any obligation to update the analysis. Publication of an analysis should not be read as a statement that any position does or does not exist.
The analysis may contain errors, omissions, or information that changes after publication. Readers should review official filings and primary sources before making decisions.