Matrix in 2025: The peak is here, but 2026 will be tested by Magic and cash quality

Getting To Know The Company

Matrix is not a classic software company, and it is not a simple staffing-heavy IT house either. It is a broad IT platform with a very Israeli core and a growing international layer, spanning projects, outsourcing, cloud infrastructure, cyber, data, core systems, software distribution, consulting, and engineering. That matters because anyone who looks only at the revenue line misses the actual economics. This is a group that sells a lot of human expertise, but also meaningful layers of software, licenses, and solutions that carry a very different margin profile.

What is working right now? Quite a lot. In 2025 the company posted records in revenue, operating profit, net profit, and cash flow. Growth was not acquisition-only. Excluding the increase in transactions presented on a net basis, revenue growth was 15.4%, and adjusted organic growth was 10.7%. The fourth quarter was also strong, with revenue up 16.4%, while the net-presentation effect in the quarter was already negligible. That gives a cleaner view of the operating engine than the full-year headline alone.

This is also where the story gets less clean. The operating margin looks better on the surface, rising from 8.1% to 8.4%, but on an adjusted revenue base it stayed at 8.1%. In other words, part of the visible improvement is not a true step-up in earnings quality. It is a function of a change in the mix of transactions presented net, mainly in cloud activity. Cash flow, strong as it was, also got a boost from a fourth-quarter receivables assignment transaction. So Matrix's peak year is real, but it is not as clean as a first read suggests.

Why does this matter now? Because the 2026 discussion is no longer about Matrix on a standalone basis. After the Magic deal closed on February 24, 2026, the company immediately started presenting itself at a different scale: more than 17,000 employees worldwide, more than 6,000 customers, activity in over 50 countries, and 2025 pro forma revenue of NIS 8.39 billion with EBITDA of NIS 1.02 billion. That means 2026 is not just another continuation year. It is a test year for integration, financing, accounting translation, and capital-allocation discipline.

There is also a market layer worth putting on the table early. In early April 2026, Matrix's market cap stood at about NIS 8.1 billion, with daily trading value around NIS 7.9 million. This is not an illiquid stock. At the same time, short interest as a percentage of float rose to 2.24% by the end of March 2026, versus a sector average of 0.72%, and SIR of 3.24 was also above the sector average. That is not an extreme no-confidence vote, but it is the market's way of saying the next year will be judged with less patience.

The bottom line is that Matrix exits 2025 stronger, larger, and more liquid. What is still missing for a fully clean thesis is a double proof point: that adjusted growth turns into genuine margin improvement, and that strong cash generation does not depend on year-end moves that are hard to repeat.

A quick economic map of 2025:

Events And Triggers

Magic closed, and the 2026 story changed

Matrix's single biggest event does not sit inside 2025. It comes right after it. On February 24, 2026, the merger with Magic was completed, and Matrix issued 28,861,564 shares to Magic shareholders, at an exchange ratio of 0.5878202 Matrix shares for each Magic share. From that point onward, Magic became a wholly owned private subsidiary.

This is not a standard leveraged acquisition story. The accounting treatment is a combination under common control using an as-pooling approach. In practice, that means this is not the kind of transaction that immediately loads the group with purchase-price accounting noise and acquisition amortization. Management even published 2025 pro forma numbers as if the two companies had always been together. That is an important signal. Management wants the market to start reading 2026 through a new group structure now, not through old Matrix alone.

Series 2 looks like optimization, not fire-fighting

On February 4, 2026, Matrix completed the issuance of its Series 2 convertible bond, with about NIS 297 million nominal value, five-year maturity, 0.5% annual interest, and a single principal repayment in February 2031. The shelf-offering documents and the rating report make clear that the proceeds are meant for debt refinancing and ongoing activity, and the investor presentation sharpened that message further by saying the main use should be repayment of more expensive debt at Magic.

That distinction matters. If the issuance had been required to plug a liquidity hole, it would need to be read very differently. But Matrix came into it with NIS 902.9 million of cash and equivalents, NIS 1.235 billion of undrawn credit lines, and leverage ratios sitting far away from covenant pressure. This looks more like optimization of the combined structure than a rescue move.

At the same time, the other side must stay in view. Every convertible is also a potential dilution layer, even if here the full-conversion impact is only around 2.62% of issued and paid-up capital. So the financing is cheap and comfortable, but it still makes capital allocation and value distribution slightly less straightforward.

The dividend tells a story of its own

Matrix continues to operate with a distribution policy of up to 75% of profit attributable to shareholders, and for the fourth quarter of 2025 the company declared a dividend of NIS 73.1 million, or NIS 0.79 per share. What matters here is not only the dividend itself, but the fact that Matrix framed it as 75% of the combined fourth-quarter profit of Matrix and Magic together.

That is a very explicit market message. Even before the full accounting consolidation enters the financial statements, management is already telling the story of the group on a combined basis. It strengthens the scale thesis and the sense that the deal has already matured strategically. It also sharpens the discipline test. If management wants to keep distributing at a similar pace while funding integration and refinancing, cash quality becomes much more important.

The outside signal is supportive, but not a substitute for analysis

Midroog affirmed an Aa3.il issuer rating with a stable outlook in January 2026, and gave the same rating to the new bond series. That is a meaningful external signal, especially because it came right before the new issuance and in the same window as the Magic transaction.

It supports the view that Matrix's balance-sheet story remains very strong. It does not answer the questions around cash-flow quality, adjusted profitability, or post-merger execution, but it does draw a clear line: the market may debate the quality of growth, but basic credit stability is not the main argument here.

Efficiency, Profitability And Competition

The margin improved in the statements, but not in the underlying economics

This may be the single most important finding in the whole report. On first glance, Matrix looks like a story of both growth and stronger profitability: revenue rose 11.8%, operating profit rose 16%, and the operating margin moved from 8.1% to 8.4%. But once the company itself strips out the increase in transactions recognized on a net basis, the picture changes. Adjusted 2025 revenue reaches NIS 6.44 billion, and the operating margin goes back to 8.1%, exactly where it was in 2024.

This is not a footnote-level accounting nuance. It is the difference between a productivity re-rating story and a story of strong growth with stable profitability. Anyone reading 2025 as a year of true margin re-rating is missing that a meaningful part of the reported improvement is tied to revenue presentation, especially in cloud.

That is not the same as weakness. Adjusted revenue growth of 15.4% alongside a stable 8.1% margin is still a strong outcome for a company of this size. It just needs to be read correctly: Matrix accelerated in 2025, but it did not yet move to a higher profitability tier.

Israel still carries the group

The Israeli IT solutions, consulting, and management segment grew 14.5% to NIS 3.82 billion, while operating profit rose 16.5% to NIS 291.5 million. This is still the heavy engine of the group, both in revenue and in earnings contribution, driven by data, analytics, digital, core systems, defense, and financial customers.

The interesting point is not growth by itself, but the quality of it. The company says explicitly that the main driver was organic, while Segev Systems added first-time consolidation support. At the same time, the fourth quarter already absorbed one-off costs from the Magic deal, which pushed the quarterly margin down to 7.4% from 8.0% despite revenue growth of 21.1%. That means the Israeli core is still working well, but it has already started paying for the strategic story of 2026.

The US is small in revenue, but more important in profit

In shekel terms, the US segment barely grew in 2025 and even edged down slightly to NIS 458.6 million. Anyone stopping there will read this as weakness. That would be a mistake. In dollar terms, revenue rose 6.6% to $132.9 million, and operating profit rose 11.2% to $20.1 million, with the operating margin improving to 15.1%.

That has two implications. First, the US activity remains more efficient than the group average. Second, the apparent weakness in reported shekels reflects mostly dollar translation, not business deterioration. That matters even more ahead of the Magic combination, which adds a much broader international layer. If the dollar stays weak, reported revenue can still look less impressive even while the underlying business keeps expanding.

There is also a real counterweight. Matrix-IFS depends materially on software systems from external vendors and on large institutional customers, especially in GRC and financial verticals. The company itself acknowledges that if those software vendors deepen direct service to customers, or if the relationship with them deteriorates, US profitability could be hit meaningfully. So the US is a quality lever, but it also carries higher outside dependence.

The smallest segment is also one of the most revealing

Software products marketing and support fell 3.6% in revenue to NIS 440.4 million, but operating profit jumped 21.5% to NIS 55.1 million, and the segment margin rose from 9.9% to 12.5%. In the fourth quarter alone, the margin reached 29.3%.

That is not a trivial data point. When revenue falls and profitability jumps, it usually means the transaction mix shifted materially. Here the company says so directly: customers bought more data and AI solutions, communication equipment for AI activity, cyber defense, digital acceleration, and cloud solutions. In other words, the group did not simply sell more product. It sold more selectively, into categories with a better earnings profile.

This is also a forward signal. If Magic really expands the group's layer of proprietary solutions and customer stickiness, this is the direction management seems to want: less dependence on linear headcount growth, more weight in products, software, and integration layers with stronger earnings quality.

Cloud grew, but part of the distortion sits there too

Cloud and computing infrastructure revenue grew 8.5% to NIS 1.65 billion, while operating profit rose 18.2% to NIS 125.8 million, with the margin improving to 7.6% from 7.0%. That looks like a solid business improvement, and part of it is.

But this is also the segment where most of the increase in net-presented transactions was recorded. So it contributes both to real economics and to a distorted first read of the numbers. The company says explicitly that cloud EDP transactions presented net became more material and are supportive of the forward view. That is positive from a demand perspective, but it requires analytical discipline. Not every gain in cloud margin is necessarily operational improvement. Part of it is a presentation effect.

Cash Flow, Debt And Capital Structure

Cash flow is strong, but only if the framing is explicit

Matrix ended 2025 with NIS 835.3 million of operating cash flow and NIS 902.9 million of cash and cash equivalents. That is a very strong picture. The company also moved from net financial debt of NIS 116.6 million at the end of 2024 to net cash of NIS 149.9 million at the end of 2025. At the headline level, it is hard to ask for much more.

But the framing matters. On a normalized cash generation basis, meaning recurring cash production before strategic capital uses, Matrix looks strong. Net profit was NIS 330.3 million, non-cash P&L adjustments added NIS 386.7 million, and working-capital changes added another NIS 244.6 million. Cash interest payments were still relatively modest at NIS 43.6 million for the year, while interest income was NIS 20.3 million. This is not a company choking on funding costs.

On an all-in cash flexibility basis, the picture is more demanding. In the same year the company spent NIS 34.5 million on reported CAPEX, NIS 65.4 million on acquisitions and activities, NIS 269.1 million on dividends, NIS 128.7 million on lease liability repayments, NIS 125.1 million on net credit repayment, NIS 67.9 million on bond principal repayment, and smaller additional amounts related to minority shareholders, put options, and acquisition liabilities. So Matrix generates a lot of cash, but it also uses that cash very actively.

2025 cash flow got meaningful help from the year end

This is the main forensic point in the report. Receivables and accrued income fell by NIS 127.3 million, and in the cash-flow statement the decline in receivables contributed NIS 190.9 million. In the directors' report, Matrix explains that the decline was partly due to a receivables discounting transaction executed in the fourth quarter. In the accounting note, the company states that it entered into a receivables assignment transaction for about NIS 261.4 million, structured as a true sale, and therefore derecognized the financial asset from its books.

That does not make the cash flow weak. It does mean the 2025 jump is not entirely a pure function of organic collection. Part of the cash improvement came from a real sale of receivables to a financial institution. Anyone treating NIS 835 million as a clean baseline that can be reproduced every year should be careful. The first question is how much working capital was genuinely released, and how much was simply moved outside.

The balance sheet is strong, and covenants are barely relevant right now

From a liquidity and debt-structure perspective, Matrix is in a comfortable place. The company has NIS 1.405 billion of credit lines, of which NIS 1.235 billion were unused. Liquid resources, including cash plus available lines, amount to NIS 2.138 billion. It still carries NIS 300 million of commercial paper, Series B bonds of roughly NIS 271.9 million nominal around the reporting date, and the new convertible Series 2 at NIS 297.0 million nominal.

The more important point is the distance from covenants. Consolidated net financial debt to total balance sheet stands at negative 3.3% versus a ceiling of 45%. Net financial debt to adjusted EBITDA stands at negative 0.21 versus a ceiling of 5. Equity of NIS 1.215 billion sits far above the NIS 400 million floor. Even the tighter distribution restrictions in Series 2 currently look remote: leverage ceilings of 37% and 4 times net debt to EBITDA, plus a minimum equity threshold of NIS 450 million.

That means Matrix's 2026 test is not a balance-sheet survival test. It is a capital-allocation test. Will the company use its balance-sheet comfort to improve group quality, or simply to keep growing while leaving less cash genuinely available to shareholders than the headline suggests?

Finance expense rose, but not because cash interest suddenly exploded

Net finance expense rose 26.6% to NIS 84.7 million. Again, the surface reading is misleading. The investor presentation breaks the increase into two main pieces: FX expense, which rose to NIS 23.1 million, and accounting finance expense, which rose to NIS 36.7 million. By contrast, net interest, fees, and other financing costs were almost unchanged at NIS 24.9 million versus NIS 25.4 million.

That matters a lot for 2026. The company even states that, following completion of the Magic merger, it intends to change its accounting policy for minority put options and move to a partial-recognition approach, so value updates would no longer run through the income statement but through equity. If that change happens, the 2026 finance line may look cleaner even without a dramatic business change.

Outlook

Four findings need to stay in view before anything else

First: 2025 was a peak year, but not a margin-re-rating year. Adjusted growth was very strong, while the adjusted margin stayed flat.

Second: the strong cash flow does not eliminate the cash-quality question. It only shows that Matrix can generate cash and also manage year-end balance-sheet timing very actively.

Third: Magic changes Matrix's scale much more than it changes the underlying 2025 economics. That is why 2026 will be judged less on strategic storytelling and more on execution.

Fourth: Series 2 and the stable rating reduce financing pressure, so if something weighs on the stock in 2026 it is more likely to be integration quality, margin quality, and cash conversion than debt stress.

2026 looks like a bridge year, not an automatic breakout

There is no formal numerical guidance here that locks in a hard target, but there is enough to characterize the year. After a record report, the Magic deal, and cheap debt financing, 2026 looks like a bridge year in which Matrix needs to prove three things at once: that adjusted growth continues beyond 2025, that profitability does not slip under the new mix, and that Magic improves the group beyond the headline scale story.

That is the heart of it. The pro forma numbers show a company roughly 34% larger on revenue and roughly 42% larger on EBITDA than reported Matrix alone. But pro forma is not cash, not integration, and not proof that a very human-capital-heavy services platform can be combined cleanly with a broader proprietary software layer and a more internationally concentrated customer base.

The backlog is large, but it should not be confused with certainty

Matrix's backlog rose to NIS 7.49 billion, of which NIS 5.48 billion is intended for the current year. That is a very large number. US backlog rose 17.4%, Israel rose 6.8%, and software products rose 8.6%. Separately, Magic disclosed backlog of NIS 1.63 billion, of which NIS 1.51 billion is for the current year.

But Matrix also says explicitly that most customer engagements can be canceled or reduced with prior notice, and that a significant part of backlog relies on framework agreements. That does not erase the value of the number, but it changes how the number should be interpreted. This backlog gives good direction, not hard contractual certainty.

So the real 2026 trigger is not another big backlog announcement. The trigger is conversion: how quickly that backlog turns into revenue and profit once the group is larger and more complex.

Magic brings what Matrix was missing, and also what Matrix is less used to

The upside case is clear. Magic adds low-code platforms, vertical software, integration solutions, and a broader customer footprint. Matrix itself emphasizes that the deal expands activity in ERP, cyber, cloud, data, digital, and AI, while also increasing the group's footprint in defense.

The less comfortable part sits in the details. According to the transaction disclosure, Magic's two largest customers account for 13.3% of revenue, and its top five customers account for 21%. That is not an extreme concentration risk, but it is not Matrix-like diversification either. Magic also brings 4,070 employees and activity across about 80 active corporations. So 2026 requires not only growth, but the ability to manage real complexity.

The 2026 finance line may look better, but that does not automatically improve cash to shareholders

Matrix plans to change the accounting treatment of put options starting in 2026, and Series 2 is intended to replace more expensive debt at Magic. Both moves can improve the accounting appearance of 2026. They do not necessarily improve shareholder-accessible cash to the same degree.

This is exactly where created value and accessible value have to be separated. If accounting finance expense falls and the group looks cleaner, that can help the multiple. But if dividend payouts remain aggressive, if receivables assignment is needed again, and if integration consumes cash, then the translation to common shareholders will be less generous than the pro forma story suggests.

What must happen in the next 2 to 4 quarters for the thesis to strengthen?

• The combined group needs to show revenue growth without leaning mainly on net-presentation effects or acquisitions.
• The operating margin needs to improve on an adjusted basis, not just in reported presentation.
• Operating cash flow needs to stay strong without another large year-end receivables move.
• Magic integration needs to progress without a meaningful leak into selling costs, financing costs, or customer friction.

What would weaken the thesis?

• Another year in which the adjusted margin stays stuck despite revenue growth.
• A return to aggressive working-capital transactions to support the cash picture.
• A slow Magic integration that adds more complexity than quality.
• A weak dollar that keeps eroding US translation without enough operating offset.

Risks

The backlog is not bulletproof, because much of it sits on framework agreements

Matrix itself says that most of its customer engagements can be canceled or reduced with prior notice. That is especially relevant in expert-services activity and framework agreements. So a NIS 7.49 billion backlog is important, but it is not the same thing as a fully locked contractual moat.

Cash quality is still an open question

The fourth-quarter receivables assignment is not a red flag by itself. It was structured as a true sale. But when a company uses exactly that kind of move in a peak cash-flow year, it is fair to ask how much of the working-capital improvement is genuinely repeatable. The risk here is not immediate liquidity. The risk is that the market starts reading future cash-flow jumps more skeptically.

Currency exposure already hurt reported 2025 numbers

The US segment posted good growth in dollars, but barely grew in shekels. At the same time, FX differences contributed to higher finance expense, and management itself described a meaningful translation drag. If the dollar stays weak, 2026 can look less strong in reported numbers even if US operations keep progressing.

Magic adds scale, but also concentration and complexity

Matrix does not appear to rely on a single customer in its main Israeli engine. Magic, by contrast, gets 13.3% of revenue from its two largest customers and 21% from its top five. Add around 4,070 employees and roughly 80 active corporations, and the risk here is less financial than operational.

An aggressive distribution policy can become friction

A policy of distributing up to 75% of profit is very comfortable for shareholders when the business is simple and stable. It is less simple in a year when the company is absorbing a large target, refinancing debt, and still has to prove that its cash generation is genuinely clean. Covenant room is wide, but the market can become more conservative long before covenants become relevant.

Short Interest View

Short interest in Matrix is not extreme, but it is clearly no longer negligible. As of March 27, 2026, short interest as a percentage of float stood at 2.24% versus a sector average of 0.72%, while SIR stood at 3.24 versus a sector average of 1.339. That is already more than indifference. It is a demand for proof.

The more interesting point is the direction. In November 2025, short float was still below 1%. From January through March 2026 it climbed gradually. The market is not attacking Matrix's credit quality, because the rating, cash, and covenant profile are too strong for that. It is testing whether the combination of Magic, apparently strong cash generation, and improved reported margin is being read too quickly as a perfectly clean story.

Put differently, short sellers are not arguing that Matrix is weak. They are asking whether 2026 will look as clean as the presentation deck, or closer to the complexity already visible in the notes.

Conclusion

Matrix ends 2025 with an excellent report, but not with a simple one. Growth is strong, the balance sheet is strong, liquidity is wide, and the Magic deal can materially change the scale of the group. On the other hand, the adjusted margin did not really improve, part of the cash-flow strength leaned on a one-off receivables move, and 2026 will test execution quality much more harshly.

Current thesis in one line: Matrix enters 2026 from a position of strength, but the next year will stand or fall on whether Magic and adjusted growth improve the quality of margin and cash, not just the size of the group.

What has changed versus the earlier understanding of Matrix? The company is no longer just a large Israeli services house with some complementary international exposure. After Magic, it is moving toward a much broader IT and software platform. At the same time, 2025 proved that the core business could grow even without the deal. The change is not growth itself. It is the level of complexity investors now need to digest.

The strongest counter-thesis is that the market is being too demanding on quality details. One could argue that Matrix has already delivered what matters: strong organic growth, stable operating margin, a strong balance sheet, cheap financing, and a Magic transaction that widens the mix toward proprietary solutions without meaningfully stressing credit quality.

What could change the market's interpretation in the short to medium term? The first combined reports need to show that the group can hold an adjusted operating margin without relying on year-end customer-balance moves and without unpleasant surprises in selling costs or finance costs. On the other side, any sign that the accounting improvement is cleaner than the economic improvement could weigh on the stock quickly.

Why does this matter? Because Matrix is no longer judged only as a very good execution company. It is now judged as a company that has to convert scale, software, cheap debt, and a major integration into more accessible cash and higher earnings quality for shareholders.

What must happen in the next 2 to 4 quarters for the thesis to strengthen, and what would undermine it? To strengthen the case, the company needs to show that the combined engine converts growth into better adjusted margin and into cash that depends less on working-capital management. What would weaken it is a year in which the report still looks good, but cash flow, working capital, and integration all require more explanation than comfort.