Amanet 2025: backlog expanded, but the cash-conversion test is still open

Getting To Know The Company

At first glance Amanet looks like a scattered group of consulting, project-management, software, and logistics activities. That reading is incomplete. In practice this is one listed company sitting on two very different engines. One is a broad consulting, project-management, and logistics arm built more around hours and projects. The other is a computing, software, and hardware arm that now carries most of the scale, most of the backlog, and most of the profit. Anyone reading Amanet as a balanced services group is missing the main point. In 2025 it looks much more like an IT and testing platform with a consulting arm that is still being rebuilt.

What is working is working well enough. Revenue rose to NIS 338.9 million from NIS 323.6 million in 2024, gross profit rose to NIS 25.7 million from NIS 23.2 million, and operating profit rose to NIS 11.0 million from NIS 10.2 million. If the other income and expense line is stripped out, the comparison looks better still, because 2024 included a NIS 4.9 million gain from a real-estate sale, about NIS 1 million of goodwill impairment, and a NIS 2.8 million write-down on Amgon. That means the 2025 improvement was less about a headline and more about real operating progress, mainly in computing, software, and hardware, helped by projects tied to the defense industries.

That is also where the friction starts. Operating cash flow fell to only NIS 3.7 million from NIS 14.3 million in 2024. This is not a leverage story. Amanet has no bank debt, no credit facilities, and equity of NIS 99.7 million, about 62% of the balance sheet. It is a cash-conversion story. The company paid NIS 8 million of dividends in 2025, paid NIS 5.0 million of lease principal, bought the Ran Shtok activity, and ended the year with more cash mainly because it sold Amgon and realized securities. The active bottleneck today is therefore not debt. It is cash quality, and the question of how backlog does, or does not, turn into accessible cash.

The backlog itself also needs a more careful read than the headline suggests. At year-end 2025 it stood at about NIS 545 million versus about NIS 439 million at year-end 2024, but almost all of that jump came from the technology segment. Consulting backlog fell to NIS 151 million from NIS 160 million, while computing backlog rose to NIS 394 million from NIS 279 million. On top of that, the disclosed material contracts leave customers with meaningful flexibility. The Ministry of Health can end one contract with notice, the software-testing contract can be terminated on 90 days' notice, and the hardware-testing contract does not commit the customer to any minimum order volume at all. The backlog is real, but it is not the same thing as locked cash.

There is also a practical screen that belongs early in the article. Amanet joined the exchange's market-making program in November 2025, but the stock remained very illiquid even after that step, and the latest trading snapshot still showed negligible turnover. That is not cosmetic. Even if 2026 is a better operating year, a company like this usually needs longer to turn operating improvement into a cleaner market story.

What Matters Right Away

• Finding one: The group is growing mainly through the computing, software, and hardware arm, not through broad-based strength across every engine.
• Finding two: A NIS 545 million backlog sounds like a step change, but a meaningful part of it sits on service frameworks that still leave the customer substantial flexibility.
• Finding three: The balance sheet is strong, yet the 2025 all-in cash flexibility test was weak, and the company relied on asset realization and the Amgon sale to end the year with more cash.
• Finding four: The consulting arm still has not returned to a position of strength. It cleaned up the base by selling Amgon, added a small add-on acquisition through Ran Shtok, and is now searching for a bigger move through a non-binding engineering MOU.

Amanet's Economic Map

Events And Triggers

The technology engine gained another layer of visibility

Two 2025 developments strengthened the reading of Amanet as an IT and testing company much more than as a broad consulting group. The first was Tasanet's new agreement, signed in January 2025, to continue providing software-testing, consulting, and technology services to an existing client through August 2029. Revenue from that contract was NIS 22.2 million in 2025, and the company estimated the remaining expected revenue at about NIS 77 million, while the current monthly retainer implies an annual run rate of about NIS 21 million.

The second, and louder, event was Tasanet's June 2025 win in the follow-on hardware-testing tender. This already carries a framework estimated at about NIS 100 million for the first five-year term, and about NIS 200 million if the full option period is exercised. That matters because it deepens Amanet's exposure to hardware testing and broader technology systems. But the small print matters just as much. The customer can increase, reduce, or not order at all, and is not committed to any minimum. That means the market should measure real revenue and real service-call flow here, not just the size of the framework.

The consulting arm first cleaned the base and then turned to acquisitions

Consulting, project management, and logistics went through a two-part move in 2025. On one side, Amanet sold all of its Amgon holdings in March 2025 for NIS 7 million, after already recording a NIS 2.8 million impairment at the end of 2024. That was effectively an admission that activities which do not create value or only complicate the picture do not have to stay inside the group. On the other side, in November 2025 Aman bought the Ran Shtok management and business advisory activity for a total fair value of NIS 1.682 million. It is a small acquisition, but it shows the company is not only shrinking weaker pockets. It is also trying to add a layer of managerial and economic consulting.

The bigger move has not closed yet. In December 2025 the company signed a non-binding memorandum of understanding to acquire 85% of a civil-engineering company specializing in planning, control, and quality assurance in infrastructure, at a valuation derived from a multiple on the average operating profit of 2024 and 2025 and in any case not less than NIS 14.3 million for 100% of the equity. The exclusivity period was extended to March 31, 2026 for due diligence and binding agreements. This is not a side note. It suggests management itself understands that the consulting arm needs a new professional layer in infrastructure and engineering, not just a continuation of the old base.

Liquidity and dividends have become part of the read

Management did not spend 2025 only on operations. In January the board approved a new dividend policy for 2025 and 2026 under which the company plans to distribute dividends twice a year at no less than 60% of profit attributable to shareholders, subject to a separate board decision and the legal distribution tests. In practice Amanet paid NIS 4 million in April 2025 and another NIS 4 million in September 2025, and then approved an additional NIS 3 million dividend in March 2026. This happened alongside the market-maker appointment that became effective in November 2025.

The double message is clear. On one hand the company is signaling confidence in the balance sheet and in earnings. On the other hand it is pulling cash out at a time when operating cash flow has weakened, while also trying to improve liquidity because the stock still does not trade easily. Those two things together make 2026 a year in which the market will judge not only activity levels, but also capital discipline.

Efficiency, Profitability And Competition

This is not a balanced two-engine business

Anyone looking for two equal engines inside Amanet is going to get the wrong picture. According to the presentation, in 2025 the computing, software, and hardware segment generated about 68% of revenue but about 81% of segment operating profit. Consulting, project management, and logistics accounted for about 32% of revenue and only about 19% of segment operating profit. That is the core point. Even if both segments grew, only one of them is really pushing the operating line forward.

This matters because it blocks a flat reading of "diversification." Yes, Amanet operates in two segments. In practice, the technology engine is funding a large part of the group, while consulting is still searching for a cleaner profit structure. Any weakness in computing therefore becomes a group issue, not just a segment issue.

Profit improved, but only after 2024 is normalized

The consolidated numbers look good on first read: revenue rose 4.7%, gross profit rose 10.8%, and operating profit rose 8.7%. But the more important point is that the true operating improvement was sharper, because 2024 included unusual items that distort the comparison. That year included a NIS 4.9 million gain on real-estate sales, a NIS 2.8 million write-down on Amgon, and about NIS 1 million of goodwill impairment. In the presentation the company already shows 2024 on a cleaner basis, and on that basis 2025 operating profit looks stronger.

The second half also supports the idea that momentum improved. In the second half of 2025 revenue rose to NIS 170.9 million from NIS 164.7 million a year earlier, while operating profit rose to NIS 6.0 million from NIS 1.8 million. That comparison also needs caution because the second half of 2024 sat on a weaker base. Still, it is clear the company did not finish 2025 with operating fatigue.

The edge is tenure and relationships, not unlimited pricing power

Amanet operates in competitive markets in both segments. Consulting faces a relatively large number of competitors, some smaller, some similarly sized, and some willing to offer services at lower prices. The company itself says there are cases where it struggles to compete while maintaining the required quality level. Computing faces a different kind of pressure: service pricing, a tight labor market, customer demands for efficiency, and competition coming both from offshore models and from automation and AI tools.

What protects Amanet is therefore not a hard moat. It is the combination of long-standing customer relationships, customer depth, and the ability to provide several services under one group. In consulting, customers with relationships longer than five years generated NIS 88.5 million of revenue in 2025, about 26% of group revenue. In computing, that figure was NIS 201.3 million, about 60% of group revenue. In addition, the Ministry of Health represented about 11% of group revenue in 2025. That identity matters. It means at least part of the business is built on deep relationships with large public-sector customers rather than on occasional sales.

But there is a second side here too. Long-standing relationships with public bodies and large corporate customers create stability, but they do not remove the question of at what price and with how much customer flexibility. That is especially true in the technology segment, where some contracts are built around adjusted retainer structures or service frameworks based on actual usage.

Cash Flow, Debt And Capital Structure

The balance sheet is clean, but that is not the same as spare cash

On a balance-sheet view, Amanet looks orderly. At year-end 2025 it had NIS 25.7 million of cash and cash equivalents, NIS 21.1 million of fair-value financial assets, a NIS 85.0 million working-capital surplus, and NIS 99.7 million of equity. There are no bank loans, no credit facilities, and management says the company has not needed them. The rate backdrop also matters less here than it does for many other companies, because there is no financial debt and most of the group's agreements are not CPI-linked.

That is a good starting point. It also explains why management feels comfortable thinking about acquisitions, further expansion, and continued distributions. But anyone stopping there is missing the real pressure on cash. Part of the liquidity pool is invested in securities and deposits meant to support operations and acquisitions, while another part is funding real and recurring cash uses such as leases, dividends, small add-on deals, and working capital.

On an all-in cash flexibility basis, 2025 was weak

This is the main yellow flag in the report. Operating cash flow fell to only NIS 3.7 million. In the same year Amanet paid NIS 5.0 million of lease principal, NIS 8.0 million of dividends, NIS 0.2 million of minority dividends, NIS 1.0 million of cash consideration for the Ran Shtok acquisition, and NIS 1.2 million of capital expenditures. In other words, once real cash uses are included, the business did not comfortably fund itself.

What closed the gap? The Amgon sale contributed NIS 7.0 million, net realization of financial assets contributed NIS 8.5 million, and the repayment of a long-term loan added NIS 1.2 million. That is how the cash balance still rose by NIS 5.0 million. There is nothing improper about that. But it does need to be read correctly. In 2025 Amanet increased cash not because operations were producing cash at a high rate, but because the company also realized assets and sold a holding.

Working capital reinforces the caution. Average customer credit days rose to 99 from 87, while supplier credit days fell to 69 from 84. That is not what a year looks like when earnings simply drop into the bank. The company itself explains that the decline in trade payables weighed on cash flow, and that is exactly the kind of moving part the market should monitor in 2026.

The dividend policy has become a capital-allocation test

There is no legal issue around distributions. Retained distributable profits stood at about NIS 48 million at year-end 2025, and after the March 2026 dividend approval the remaining distributable profit was still about NIS 45 million. But the more interesting question is not whether the company may distribute. It is whether it should keep distributing at the same pace while also trying to buy new activities and preserve strategic flexibility.

That matters especially because management explicitly describes about NIS 48 million of invested liquid balances as a support pool for subsidiaries and for M&A opportunities. If that is the case, every shekel paid out in dividends is also a shekel that is no longer standing behind the expansion plan.

Outlook

Four points will decide 2026

• Finding one: 2026 is a proof year for backlog quality in the technology segment, not a discovery year. The backlog is already there. Now it has to translate into revenue, profit, and cash.
• Finding two: The new hardware-testing framework can materially strengthen the story, but as long as there is no minimum commitment, the market will need to see real execution, not only an impressive framework value.
• Finding three: Consulting has still not shown a broad organic step-up. For now it rests on the existing base, on the Ran Shtok add-on acquisition, and on the hope that the engineering MOU becomes a real deal.
• Finding four: The dividend policy makes 2026 a year that will also be judged through cash flow, not just through operating profit.

What has to happen in technology

The technology side enters 2026 with a clear advantage: NIS 394 million of backlog, software and hardware testing activity with established customers, growth in defense-related projects, and a continuing strategic push around ERP, MES, and AI-based tools. This is a real engine. But it still sits in a market defined by wage competition, pricing pressure, and customers looking for more efficiency at lower cost.

So the 2026 question is not whether Amanet can fill hours. It is whether it can preserve the quality of those hours. If software testing stays around the current retainer pace, if hardware testing turns into real service-call volume approaching the economic framework, and if MES projects start becoming more than a strategic option, that segment can keep carrying the group. If not, Amanet can still end up with another year of impressive backlog and slower conversion into net profit and cash.

What has to happen in consulting

Consulting, project management, and logistics is not broken, but it is also not yet orderly enough to carry the thesis on its own. Its backlog declined, and the company itself says it is trying to widen activity in military projects, infrastructure, rehabilitation projects in the south and north, and the private, insurance, and banking markets. That is a good work list. It is still a work list.

The Ran Shtok acquisition adds managerial, economic, and business advisory services, but the company itself defines that deal as non-material. The implication is that consulting needs either a larger win or a larger acquisition to change its profile. The civil-engineering MOU could be that kind of move because it opens exposure to planning, control, and quality assurance in infrastructure. But as long as there is no binding agreement, it cannot be treated as part of the current thesis. It remains a strategic option, not a closed fact pattern.

What kind of year is ahead

Management's 2026 work plan is built on existing backlog, new incoming work, and an ambition for higher activity and better profitability. That framework makes sense. The right interpretation of it is a proof year, not a breakout year. Amanet is not trying to discover a new engine from scratch. It is trying to prove that what was already put on the table in 2025, above all inside Tasanet and the broader technology arm, can now produce better economics at the group level.

What could improve the read over the next 2 to 4 quarters is a combination of three things: a clear recovery in operating cash flow, proof that the hardware-testing framework is filling with real activity rather than remaining only a headline number, and visible progress in the structure of consulting, either through new backlog or through a binding version of the engineering deal. What would weaken the read is another year in which backlog rises but cash still depends on realizations while capital allocation remains too generous relative to operating cash flow.

Risks

Risk one is that the backlog may look stronger than the economics behind it. That is especially true in contracts where the customer can change volumes, cancel on short notice, or avoid committing to minimum orders. The hardware-testing framework says this explicitly. The software-testing contract has a retainer, but it also allows termination on 90 days' notice.

Risk two is that weak cash flow continues just as the company keeps paying dividends and looking for acquisitions. Amanet has a clean balance sheet, but that should not be confused with spare cash available to shareholders. If 2026 does not show a real cash-flow recovery, the tension between dividends, M&A, and flexibility will grow.

Risk three is ongoing pressure in the IT services market. The company itself writes about competition, customer demands for efficiency, a shortage of skilled labor, and the possible effect of AI and automation tools on the demand structure. This is not a theoretical risk. It is exactly the kind of pressure that can erode price or raise wage cost in the group’s main engine.

Risk four is that the consulting rebuild remains unfinished. Ran Shtok is too small to change the picture by itself, and the engineering MOU is still non-binding. If that path does not mature, Amanet will become even more dependent on the technology segment.

Risk five is legal and operational. In 2025 an NIS 18 million claim was filed against Tediad and Sisnet, and in September the group filed a counterclaim of about NIS 9.7 million. The company also mentions another urban-renewal matter where it cannot yet estimate exposure. These are not the risks defining the thesis right now, but they do remind the reader that this is a project-based services group rather than a platform with completely locked contractual cash flow.

Risk six is liquidity. Short positioning is almost nonexistent, only about 0.07% of float, so the stock does not point to a deep bearish disagreement against the fundamentals. But extremely weak trading turnover can still leave real operating improvement trapped inside a stock that struggles to price efficiently.

Conclusions

Amanet ends 2025 in a better operating position, but not in a cleaner cash position. What supports the thesis right now is a stronger technology engine, larger backlog, a balance sheet with no bank debt, and the ability to keep looking for acquisitions. What blocks a cleaner thesis is that cash generation from operations weakened, some of the material contracts are more customer-flexible than the headline implies, and the consulting arm still has not built a new earnings tier for itself.

Over the short to medium term the market is likely to focus less on the fact that backlog exists and more on three issues: whether Tasanet converts the testing frameworks into revenue and cash at a pace that justifies the story, whether operating cash flow returns to a reasonable level while distributions continue, and whether consulting gets a real growth engine beyond a small add-on deal and a memorandum of understanding.

Current thesis: Amanet now looks more like an IT and testing platform with a consulting arm under reconstruction, and the 2026 story will run through the quality of backlog conversion into cash.

What changed: backlog and profitability moved higher, but the center of gravity shifted even further toward technology just as operating cash flow weakened and consulting moved into cleanup mode plus a search for a new engine.

Counter-thesis: It can be argued that the clean balance sheet, large backlog, and Tasanet's new contracts are enough to carry Amanet through another period of relatively weak cash flow without materially hurting the thesis, because the group can keep funding itself and complete acquisitions from the existing pool.

What could change the market's interpretation over the short to medium term: a clear operating cash flow recovery, proof that the hardware-testing framework is filling in real activity, and tangible consulting progress through wins or a binding acquisition.

Why this matters: Amanet has already proved that it has a platform, customers, and a balance sheet. What it still has to prove is that the combination produces high-quality cash, not only backlog and operating profit.

What must happen over the next 2 to 4 quarters: the technology arm has to preserve profitability while turning the testing contracts into real work, consulting has to show either fresh backlog or a binding deal that explains the next step, and operating cash flow has to recover to a level that supports both dividends and strategic flexibility. What would weaken the thesis is continued cash-flow erosion alongside aggressive distributions and a consulting arm that still lacks a clear growth engine.