Alltrade Recycling: ARPE, The Backlog, The Advances And Whether This Revenue Can Really Turn Into Profit
ARPE already shows backlog of NIS 26.4 million at year-end 2025 and NIS 33.1 million near the report date, but part of that backlog is internal, two customers dominate the exposure, and deferred revenue of NIS 22.2 million still has not produced positive EBITDA. The question is no longer whether demand exists, but whether this model can leave profit after installation.
Where The Conversion From Backlog To Profit Gets Stuck
The main article identified the right problem: ARPE already creates a growth headline, but it still has not proven profit economics. This follow-up isolates where that gap actually sits, the backlog, the advances, deferred revenue and customer concentration.
That is the key point. ARPE already has demand, customer cash and a fairly near-term timetable for revenue recognition. What it still does not have is proof that the revenue survives installation, logistics, overhead and interim financing friction on its way to operating profit.
ARPE is positioned as a German execution arm for global markets, with three lines of activity: engineering consulting, plant construction and machine development and sales. The full activity description makes the same point in more detail. This is a project business, not a shelf-product business. Anyone reading ARPE’s numbers as if they describe smooth recurring revenue is missing the core issue.
ARPE’s Model, Cash Arrives Early, Risk Moves Into Execution
The contract structure is fairly explicit. Large projects can run from several months to roughly a year and a half, and they can be delivered either as turnkey jobs or on a back-to-back service-contractor basis with a defined spread. Customer advances are used by ARPE to pay suppliers. In other words, the advance is not a free cash bonus. It is part of the project funding mechanism.
The payment profile says the same thing. In most cases, up to 60% of the order is paid near signature, around 80% is collected once the machine is ready for shipment, another 10% to 15% is paid when the machine reaches the destination country, and only 5% to 10% remains after commissioning and on-site training. In most cases, letters of credit also secure the non-advance payments. That structure reduces part of the credit risk, but it does not remove execution risk. If installation drags, revenue recognition drags with it, and cash that already came in still tells you little about margin.
And even that was not enough on its own. Despite the advance-heavy model, ARPE still relied on shareholder funding. In November 2024 it received a EUR 500 thousand loan, with about EUR 480 thousand still outstanding at year-end 2025, and in January 2026 it received another EUR 150 thousand project-financing loan. That matters. Advances reduce the working-capital hole, but they do not necessarily close it. They prove customer commitment, not profit.
| Layer in the model | What the filing says | Why it matters |
|---|---|---|
| Project duration | Several months to around a year and a half | Revenue does not close on signature day. It stretches across execution |
| Payment terms | Up to 60% on signing, around 80% on readiness for shipment, 10% to 15% on arrival, 5% to 10% after commissioning | The customer pre-funds a large part of execution, but the last step still depends on successful delivery |
| Security | In many cases, letters of credit | Credit risk is lower, but execution risk remains |
| Extra funding | EUR 500 thousand shareholder loan and another EUR 150 thousand in January 2026 | Even an advance-funded model can still need bridge financing |
The Backlog Is Real, But The Headline Is Flattered
The raw number looks very good. ARPE’s backlog stood at NIS 26.4 million at the end of 2025 and rose to NIS 33.1 million near the report date, an increase of about 25.5%. At first glance, that clearly supports the case that demand is real and that 2026 already starts with signed work.
But this is where the more careful read matters. The backlog table includes, in a footnote, an EUR 800 thousand order for the tin-extraction prototype for the company itself. In plain terms, part of the backlog is not external demand at all. It is an internal order inside the group. At year-end 2025, the entire second-half 2026 backlog component, about NIS 3.0 million, was effectively that internal order. Near the report date, the same order was still sitting inside backlog at roughly the same scale.
So the more conservative read looks like this:
| Metric | Reported headline | After removing the internal prototype order | What it means |
|---|---|---|---|
| Backlog as of December 31, 2025 | NIS 26.4 million | NIS 23.4 million | Roughly 11.4% of backlog is not external demand |
| Backlog near the report date | NIS 33.1 million | NIS 30.1 million | Even after the increase, about 9.1% is still internal |
The timing profile matters too. At the end of 2025, about 88.6% of backlog was scheduled for recognition in the first half of 2026. Near the report date, about 73.5% was still scheduled for the first half of 2026 and the remainder for the second half of the same year. There is almost no distant 2027 optionality here. This is a near-term proof year.
Now add the customer note. ARPE has two main customers, Norway’s Revac AS and its Swedish subsidiary Revac Sverige AB. Purchase orders from them, from launch through December 31, 2025, totaled about NIS 29 million, and advances received on those orders totaled about NIS 15.6 million, around 53.8% of the order value. That has two implications at once: these customers are clearly willing to pre-fund execution, but a very large share of the proof still rests on two entities.
Deferred Revenue Is Already In The Balance Sheet, Profit Still Is Not
The balance sheet tells the same story from another angle. Deferred revenue jumped from NIS 35 thousand at the end of 2024 to NIS 22.179 million at the end of 2025. The balance comes from the subsidiary operating in the overseas projects segment, with revenue recognized over the installation period. Most of the increase in payables and other current liabilities, about NIS 20.4 million, also came from deferred revenue received from customers in international projects.
This is already too large to ignore. The deferred-revenue balance was 2.16 times the overseas-projects segment’s 2025 revenue, and it represented about 56.2% of total segment liabilities at year-end. In other words, a large part of the segment’s year-end liability stack already consisted of cash received before the revenue hit the income statement.
But this is exactly where the easy conclusion becomes dangerous. A high deferred-revenue balance can be healthy in a project business because it reflects signed work and customer cash. It does not mean the segment is already economically proven. To know that, one has to move to profitability.
Revenue Has Arrived, Profit Still Has Not
This is the real test. The overseas-projects segment ended 2025 with revenue of NIS 10.272 million, gross profit of NIS 1.372 million and negative EBITDA of NIS 1.103 million. In 2024, the same segment generated NIS 1.205 million of revenue, negative gross profit of NIS 164 thousand and negative EBITDA of NIS 1.246 million.
So yes, there is real progress. Revenue jumped, gross profit turned positive, and negative EBITDA narrowed somewhat. But the bottom line cannot be skipped: even after a year of sharp growth, ARPE still did not produce positive EBITDA. That is the difference between proving demand and proving a profit engine.
It also helps to remember how management itself frames profitability in this business. Gross-margin rates can range between 5% and 30%, depending on the service type, project complexity, customer risk, geography and sometimes supplier receipts. In practice, the segment ended 2025 with a gross margin of about 13.4% on a 100% activity basis, which is within the range management describes, but still not enough to generate operating profitability. So the question is no longer whether ARPE can create gross profit at all. The question is whether enough margin remains after labor, overhead, logistics and installation.
One more caution is worth keeping in view. Segment results are reviewed on a 100% basis, even though ARPE is only 80%-owned. So even if the segment turns profitable, not all of that economic improvement ultimately belongs to Alltrade’s shareholders. That does not break the thesis, but it does lower the amount of clean upside one can assign to the raw segment numbers.
What Has To Happen Now
For ARPE to stop being a story of backlog and advances and become a profit engine, four things need to happen in fairly quick sequence:
- Most of the external first-half 2026 backlog has to convert into reported revenue on time rather than being pushed forward again.
- Deferred revenue has to start unwinding into profit, not just into reported sales.
- The segment has to show a real operating-margin step-up, not just another increase in turnover.
- The need for additional shareholder funding has to disappear, or at least shrink materially, if the model truly funds itself through customers.
That is why ARPE still needs to be read with discipline. The company has already shown that it has an engineering product real customers want, that some customers are willing to prepay for it, and that it has short-dated workload for 2026. It still has not shown that this combination creates a stable profit engine. Until that happens, backlog is proof of demand, not proof of earnings quality.
Conclusion
ARPE has already passed the "is there a market?" test. It has not yet passed the "does money remain?" test. The backlog, the advances and the deferred revenue all say customers are ordering, paying and waiting for delivery. The negative EBITDA, the shareholder loans and the fact that part of backlog is actually internal say the economics are still mid-journey.
That is the difference between a future growth engine and a proven earnings engine. In 2025, Alltrade got demand validation from ARPE. In 2026, it still needs conversion validation, from external backlog, through installation, into operating profit. Until then, this revenue may turn into profit, but it still has not proved that it really does.
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.