Malam Team 2025: Bigger, More Global, but Earnings Still Aren't Clean

Getting To Know The Company

Malam Team is not an AI company, and it is not a narrow software house either. It is a broad IT-services platform spanning infrastructure and cloud, integration, software and projects, payroll and HR, plus a small technology-investment layer on top. That matters because a superficial read can miss the real economics. Anyone who reads only the growth headline and the AI language can come away thinking this is already a clean software story. In practice, more than half of revenue still comes from infrastructure and cloud, while a larger share of earnings quality increasingly comes from software and payroll.

What is clearly working now is scale. Revenue rose 9.4% in 2025 to NIS 4.138 billion, operating profit rose 8.3% to NIS 219.2 million, and the group became more international, with about 13% of revenue now generated abroad, mainly through Automat-IT and Synerion. The local demand base also remains strong. In management's presentation, about 50% of Israeli infrastructure revenue, 44% of Israeli software revenue, and 41% of Israeli payroll revenue are tied to the defense and public sectors. That is an important buffer in a noisy environment.

But this is still not a clean story. That is the key point. Infrastructure and cloud grew revenue by 12.6%, yet segment operating profit barely moved and operating margin fell to 3.9% from 4.3%. At the bottom line, reported net profit fell to NIS 77.7 million from NIS 92.6 million, partly because of NIS 30.3 million of other expenses and NIS 75.9 million of net finance expense. In other words, the company already knows how to grow, but it still does not translate that growth into bottom-line earnings with the same quality.

The active bottleneck is earnings quality and cash quality. Consolidated operating cash flow was NIS 174.7 million, but under an all-in cash flexibility view, most of it was absorbed by leases, fixed-asset spend, development spend, and acquisitions. After NIS 45.7 million of fixed-asset investment, NIS 45.1 million of additions to intangible assets, and NIS 63.9 million of lease cash, only about NIS 20 million remains. After NIS 18.5 million of acquisitions and investments, almost nothing is left. That is not a balance-sheet crisis, but it is a reminder that the gap between profit and genuinely leftover cash is still open.

That is also why 2026 reads like a proof year. The group enters it with larger scale, wider backlog, and more international exposure, but also with three open tests: whether infrastructure margins recover, whether the pension-processing reset actually closes the hole, and whether the AI, product, and special-contract layer starts looking like real economics rather than optionality. On top of that sits a structural overlay: after the balance-sheet date, the controlling shareholder published a full exchange tender offer, so in the near term the market reads Malam Team both as an operating company and as a corporate-structure event.

This is the short economic map:

Events And Triggers

Software Now Generates More Operating Profit Than Infrastructure

This is the deepest change in the report. Infrastructure and cloud still account for 56.1% of revenue, but software, projects, and business solutions already generated NIS 94.5 million of operating profit in 2025 versus NIS 89.4 million in infrastructure. That is not a reporting-segment revolution, but it is a real shift in how the group should be read: infrastructure still carries the volume, while software is increasingly carrying more of the earnings improvement.

The reason matters. In software, the company benefited from growth in outsourcing, managed services, Data and AI projects, and capability expansion through acquisitions such as Rosslare, Qumulex, and Data Cube. At the same time, segment margin improved to 6.2% from 5.9%. So this is not just another revenue line. It is a business that is moving in the right direction in quality as well.

Infrastructure Kept Growing, But Did Not Become Cleaner Economically

The infrastructure and cloud segment rose 12.6% in revenue to NIS 2.321 billion, but operating profit was almost unchanged. The fourth quarter made that even clearer: revenue rose 4.1% to NIS 638.6 million, while operating profit fell 18.4% to NIS 22.6 million and EBITDA fell 14.1% to NIS 29.5 million.

Management points to three reasons: lower-margin defense and public-sector mix, an average dollar decline of about 12% in the fourth quarter versus the prior-year period, and supply delays in storage systems and endpoint equipment because major cloud providers were prioritized for AI-related allocations. That is the difference between clean growth and expensive growth. Demand is clearly there, but not every extra shekel of revenue turned into extra profit.

The Pension-Processing Layer Moved From Disappointment To Strategic Reset

In the fourth quarter, the company recorded about NIS 30 million of expense around the provident and pension-processing activity after prolonged negotiations with an Israeli bank failed to produce an agreement. In the impairment note, it wrote down NIS 27.2 million of software-development assets, and the recoverable amount now rests on two weighted scenarios: sale of most of the activity, or continued operation through system consolidation and efficiency improvements.

That means the company has effectively moved away from a reading in which an external deal solves the issue. It is now on an internal repair track. The unified Mango system is supposed to consolidate three IT systems in this activity, customer migration started in March 2026, and completion is expected to move toward the end of 2026. This is a material test point, because the weakness here no longer looks like accounting noise. It looks like a layer that has to prove it can return to operating balance.

After The Balance-Sheet Date, Two Events Added Real Upside And Structural Friction

The first trigger: on February 3, 2026, Comtec signed a non-binding memorandum of understanding with an Israeli insurer to sell an unlimited self-use source-code license for an insurance core system for about NIS 45 million plus VAT. The company estimates pre-tax profit close to the consideration and says the maintenance fees that would stop are not material. If completed, this would be strong evidence that the group does own software assets that can generate nearly clean high-margin economics.

The second trigger: on February 26, 2026, the controlling shareholder published a full exchange tender offer for all public shares. It does not change the operating economics, but it does change the frame through which the stock is read: is this still a separately listed operating company, or is it already a structural simplification story in which shareholders need to think first about value access.

Efficiency, Profitability, And Competition

The central picture in 2025 is not just growth. It is a change in internal earnings quality. At first glance, gross margin barely changed, 10.5% versus 10.6% in 2024, and operating margin slipped only slightly to 5.3% from 5.4%. But below the surface, there is a clear split: infrastructure and cloud preserved scale, while software and payroll did the heavier lifting in earnings quality.

In infrastructure, the volume effect was positive and strong, but the mix effect and FX effect worked against it. This is a segment in which a large share of sales still comes from hardware, storage, servers, communications, and cloud products and services. In 2025 those products and services alone represented NIS 1.868 billion, about 45% of total group revenue. So the real competitive question here is not whether demand exists. It is whether the company can preserve pricing and margin in an environment where availability, allocation, and currency matter more than the revenue headline.

In software the story is different. Here volume rose, profit rose faster, and the business direction looks cleaner. The company describes an ongoing market shift from perpetual licenses to subscription models, which can hurt near-term recognition in some pockets but widen recurring revenue over time. At the same time, wins in tenders, growth in managed services, and the addition of Data Cube are creating a segment that grows slower than infrastructure, but earns better.

The payroll and HR activity looks even cleaner economically. Even on the reported segment basis, operating margin rose to 20.0%, and management's presentation shows that excluding long-term savings, revenue reached NIS 266.3 million and operating profit NIS 68.7 million, implying a 25.8% margin. That tells us the problem here is not in the core payroll, attendance, and HR engine. It is in the pension and long-term savings layer.

All of this leads to an important competitive conclusion: Malam Team is strongest when it sells knowledge, managed services, and software products, and less clean when it depends on hardware, integration, and supply-chain execution. The company benefits from scale, long-standing relationships with the public and defense sectors, and a very broad solution stack, but exactly for that reason it is also exposed to mix, suppliers, and FX.

There is another yellow flag here. In the infrastructure segment, three key suppliers represented 28.1%, 26.9%, and 14.1% of the group's purchases. The report does not disclose names, but the concentration itself is material. It does not erase the power of the platform, but it does mean that part of the business strength sits in execution and distribution capability, not only in proprietary IP.

Cash Flow, Debt, And Capital Structure

This is where framing discipline matters. Under an all-in cash flexibility view, Malam Team does not look weak, but it also does not look like a comfortable cash machine. Operating cash flow was NIS 174.7 million. Against that stood NIS 45.7 million of fixed-asset investment, NIS 45.1 million of additions to intangible assets, and NIS 63.9 million of lease cash. After those three items, only about NIS 20 million remains. After NIS 18.5 million of acquisitions and investments, almost no surplus cash is left.

That matters because anyone looking only at the NIS 489.6 million cash balance can easily conclude that financial flexibility is very wide. The balance sheet is indeed more comfortable than a superficial read may suggest, but most of that cash sits on top of a business that still requires working capital, lease cash, development spend, and small bolt-on acquisitions. That is very different from a company generating wide, clean free cash flow.

The gap between net profit and cash is also built through working capital. During the year, higher receivables, higher inventory, and larger long-term balances weakened working-capital support for cash flow, and only part of that was offset by higher supplier balances. That does not mean the earnings are unreal. But it does mean the platform still needs a lot of operational plumbing in order to convert earnings into cash.

At the same time, the debt structure itself is not under pressure. The company ended the year with NIS 684.6 million of gross bank debt against NIS 489.6 million of cash, so net financial debt stood at NIS 195.0 million. Net debt to EBITDA was only 0.60, almost unchanged from 0.61 a year earlier, and the equity-to-assets ratio improved to 29% from 28.5%. Covenant room is also wide: minimum equity of NIS 320 million, minimum equity-to-assets ratio of 20%, and maximum net debt to EBITDA of 3. Malam Team sits comfortably above and below those thresholds, respectively.

That is why the right question is not whether the balance sheet is dangerous. It is whether the company knows how to turn a strong operating base into genuinely leftover cash. At this stage, the answer is only partially.

There is also an FX layer that matters. The company says it does not use derivatives for hedging, and its sensitivity analysis shows about NIS 26.0 million of impact on net financial assets and liabilities in the event of a 10% move in the dollar. Because the group had no dollar-linked loans at the end of 2025, this is not a story of direct dollar debt. It is a story of pricing, suppliers, and balance-sheet exposure.

Outlook And Forward View

Before getting into the details, these are the five non-obvious findings the report leaves us with:

1. Software has already become the largest operating-profit engine in the group, even if not the largest revenue engine.
2. The problem in payroll is not payroll. It is long-term savings and whether Mango actually fixes the operating model.
3. The AI story is real, but still spread across services, integration, and product layers rather than showing up as one standalone economic engine.
4. The balance sheet is stronger than the headline suggests, but cash quality remains much weaker than revenue quality.
5. In 2026, the market will look less at whether Malam Team can keep growing and more at where the next clean shekel of earnings actually sits.

The coming year looks like a proof year, not a breakout year. For the read to improve meaningfully, the group has to show three things. First, infrastructure and cloud need to generate more profit per shekel of sales even if the supply and FX environment stays messy. Second, the pension-processing activity needs to move toward operating balance as the Mango migration progresses. Third, the software, Data, and special-contract layer needs to account for more of future profit.

Backlog supports cautious optimism, but not complacency. Infrastructure and cloud backlog rose to NIS 858.2 million from NIS 795.3 million, and software and business solutions backlog rose to NIS 1.087 billion from NIS 1.043 billion. That is not an explosive jump, but it does show that future work volume is intact. The test is less about whether there is enough work and more about what kind of work sits inside that backlog and at what margin.

The fourth quarter already shows what the market will be testing. Malam Team reported NIS 1.1 billion of revenue, NIS 52.8 million of operating profit, NIS 79.7 million of EBITDA, and NIS 25.4 million of adjusted net profit. Those are good numbers, but they also exposed the unevenness of the story: infrastructure margin was under pressure, the long-term savings layer took a reset charge, and software and payroll looked better. So 2026 does not need one more good quarter. It needs a sequence proving that weight is gradually moving toward cleaner engines.

There is also real optionality, but it should not be overloaded. Automat-IT added more than 300 new customers in 2025, crossed 800 customers, signed a new multi-year Strategic Collaboration Agreement with AWS replacing the RESTACK agreement, and is expanding in the US and Europe. At the same time, 4Cast signed a global cooperation agreement with SAP in 2025 and also signed a first pilot with a NATO-army customer. That is interesting, but the numbers are still small: 4Cast generated NIS 3.5 million of revenue and an NIS 11.0 million operating loss in 2025. This is optional upside, not yet a load-bearing pillar of the thesis.

Risks

The first risk is earnings-quality risk in infrastructure. As long as a large share of revenue still comes from hardware, integration, and supply-sensitive projects, the company can show strong volume without equivalent profit improvement. Supply delays, a weaker dollar, or lower-margin defense mix can all drag on conversion even when demand stays high.

The second risk is that the pension and provident-fund reset takes longer than planned. The accounting reset has already been taken, but the business reset is not finished. If Mango migration takes longer than expected or fails to deliver the hoped-for efficiency, 2026 can become another bridge year rather than a repair year.

The third risk is institutional and public-sector concentration. The Ministry of Defense alone accounted for 11.5% of group revenue in 2025, and the presentation shows broader dependence on defense and public-sector demand across all three core engines. That is a source of strength while demand is high, but also a source of risk if budgets, priorities, or procurement dynamics shift.

The fourth risk is unhedged FX exposure. The group does not use derivatives, and in a business that is becoming more global, currency moves can affect both operating results through suppliers and reported finance expense through balance-sheet effects. Net finance expense already included NIS 18.7 million of FX and other components in 2025.

The fifth risk is that the optional layer remains only optional. 4Cast, applied AI, global expansion, and special software contracts can create upside, but until they scale into visible revenue and earnings, the market may continue to discount them.

Short Interest View

Short positioning in Malam Team does not look like an attack, but it is not negligible either. As of March 27, 2026, short interest stood at 2.53% of float and SIR at 2.7. That is far from an extreme setup, but it is still meaningfully above the sector averages of 0.72% of float and 1.339 days to cover.

The more interesting point is direction. At the end of January, short interest peaked at 3.50% of float and nearly 9.9 days to cover, and it has since compressed. That is not full confidence. It is a move from sharper skepticism to milder skepticism. In other words, the market is not pricing collapse, but it is still asking for proof that growth can turn into cleaner earnings and cleaner cash.

Conclusions

Malam Team exits 2025 as a bigger, broader, and more international company. The operating base itself looks strong, and in software and payroll it arguably looks stronger than the net-profit headline suggests. But the core bottleneck has not gone away: earnings are still not clean enough, and cash is still not left over freely enough. In the near term, the market will focus on three things: infrastructure margins, the pension-processing reset, and whether the software and special-contract layer starts carrying more of group profit.

Current thesis: Malam Team has already proved scale, but 2026 will decide whether it can also turn that scale into cleaner earnings and cleaner cash.

What changed versus the older read of the company? Software became a more important profit engine, international activity is now visible in the numbers, and the weakness in pension processing has now been formally recognized as a reset issue rather than a temporary annoyance.

Counter-thesis: one can argue that this reading is too cautious, because the group sits on a strong balance sheet, wide backlog, resilient demand from the public and defense sectors, and several technology options that could improve earnings faster than the current numbers imply. On that reading, 2025 is simply a transition year before another step up.

What could change the market reading in the short to medium term? A binding completion of the Comtec deal, steady improvement in infrastructure margins, first signs of operating balance in Mango, or on the negative side another quarter in which revenue grows but earnings quality does not.

Why this matters: Malam Team is no longer being tested on whether it can grow. It is being tested on growth quality, cash quality, and how much of the value created inside the platform actually remains with shareholders.

What has to happen over the next 2 to 4 quarters? Infrastructure needs to stop diluting margin, the pension and provident-fund migration needs to move toward real balance, and the software and special-contract layer needs to account for a larger share of profit. What would weaken the thesis? More write-downs, more quarters of growth without cleaner net earnings, or proof that the technology options remain too small to matter at group level.