DSIT 2025: The Backlog and the IPO Bought Time. Now It Has to Show Cash and Margin

Introduction To The Company

DSIT is not just another small defense company selling generic "technology." It is a project-heavy business with two very different economic engines. The first is underwater defense systems: sonar, underwater area-protection systems, acoustic signal-processing solutions, and simulators. The second is RT assemblies and sub-assemblies for defense and civil customers, including anti-jam GPS systems, rugged computers, encryption, data links, and assemblies for semiconductor-testing systems.

What is working right now is clear. Demand is real, revenue rose 77% to $37.4 million, backlog reached $100.4 million, and by the report date the company had already added another $14.6 million of new orders. The March 2026 IPO also changed the liquidity picture materially: DSIT raised about NIS 48.6 million net, and by the report date it was already talking about roughly $21.4 million of liquidity and no expected need to raise additional funding over the coming year.

But anyone reading the story only through backlog or through the IPO misses the active bottleneck. The issue is not demand. The issue is conversion. The underwater segment, which now carries most of the revenue and most of the backlog, still has not proven that it can produce clean margin and clean collections at the same time. In 2025, contract assets jumped to $17.5 million from $4.7 million, and operating cash flow fell to negative $6.9 million despite net income of $5.6 million.

This is also where the superficial read goes wrong. The company looks "clean" because it has no bank loans outstanding, but that does not mean there is no balance-sheet pressure. The model depends on guarantees, fixed-price contracts, milestones, acceptance tests, and state customers. In that kind of business, quality is determined not only by whether backlog exists, but by whether backlog funds itself or forces the company to fund it.

The screen-first layer is straightforward. In the latest market snapshot, market cap stood at about NIS 406.9 million, the share price reached NIS 16.34 on April 3, 2026 versus an IPO price of NIS 10, and the market is no longer pricing immediate stress. It is pricing a growth story it is willing to believe. What is still missing for a cleaner thesis is proof that the jump in backlog and revenue can move through the operating line and through cash flow without getting stuck again in work performed but not yet billed.

This quick economic map captures the starting point:

Events And Triggers

The IPO Changed The Financing Question

The first trigger: the March 2026 IPO bought the company time. DSIT issued 5,203,100 ordinary shares for gross proceeds of NIS 52.0 million, with net proceeds after fees and expenses of about NIS 48.6 million. That is not a cosmetic detail. At the end of 2025, cash stood at only $2.6 million. Without that raise, 2026 would have started from a much tighter place.

More importantly, the company says that by the report date liquidity had already reached about $21.4 million and that, given its current plans, it does not expect to need additional funding in the coming year. In other words, the immediate funding risk eased materially. That is relief, not resolution. An IPO can finance a transition period, but it cannot by itself fix collection quality or the execution profile of fixed-price contracts.

The second trigger: on March 15, 2026 the exchange announced that the company would join the TA Tech-Elite index through the fast-track route in the August rebalance. That improves visibility, but it does not change the economics of the business. In the near term it mostly means the stock now sits under a wider market spotlight, which also increases sensitivity to any miss versus the story being sold.

European Expansion Adds Validation, But Also Pushes Cash Further Out

The third trigger: in February 2026 DSIT signed a framework agreement with a European private company to supply underwater area-protection systems to a NATO-member state end customer. Commercially, that matters a lot: a new European customer, an initial contract of about $9.7 million with a 30% advance, and the option for additional volume of up to about $4 million. By March 2026, the customer had already completed the actual orders, bringing the total to $13.5 million.

But this also has a second side. The final payment under that order is only expected in 2028. That sharpens the familiar pattern: a strong contract can open a new market door and support backlog, but it does not necessarily create fast cash. So the disclosure is very positive on commercial validation, but only partly positive on cash conversion.

The fourth trigger: even before that European framework agreement, the company had already signed an additional contract with customer B in September 2025 for underwater-system supply and installation through a G2G structure. Here too, pricing is fixed and recognition and payment are spread across two to three years. This is exactly the point: the market sees the order headline, but the real economics are only measured later, through milestones, testing, and billing.

2026 Looks Like A Proof Year, Not A Breakout Year

The fifth trigger: management is targeting roughly $50 million of 2026 revenue at an operating margin of 8% to 10%. On the surface that looks strong. In practice it is a more cautious message than it first appears. Revenue is expected to grow by about 33.7% versus 2025, but the guided operating margin is not materially above the roughly 9.4% delivered in 2025.

In other words, management is not promising an immediate step-up in earnings quality. It is effectively saying: allow the company to keep growing, invest in marketing and R&D, and build the public-company platform, and only then ask for a cleaner margin profile. That makes 2026 a proof year. If DSIT delivers the revenue without another deterioration in cash flow and margin quality, the thesis improves. If not, market patience will probably shorten quickly.

Efficiency, Profitability, And Competition

The key number here is not growth by itself. It is who grew and how.

Who Brings The Revenue And Who Brings The Profit

The underwater segment generated $26.5 million of 2025 revenue, up sharply from $13.7 million in 2024. That is the revenue engine. But its segment profit barely moved, reaching $1.43 million versus $1.34 million a year earlier. By contrast, RT rose from $6.49 million to $9.94 million, while segment profit climbed from $1.26 million to $1.90 million.

The implication is sharp: 71% of revenue came from underwater, but only about 41% of segment profit. RT, which accounts for only about 27% of revenue, already generates about 54% of segment profit. Anyone focusing only on the top-line acceleration misses that the earnings base currently depends much more on the assemblies business than on the underwater business that attracts most of the attention.

Why Margin Compressed Inside The Core Growth Story

Total gross profit rose to $10.2 million, but gross margin fell to 27.2% from 35.2%. Operating profit rose to $3.5 million, but operating margin declined to 9.4% from 13.5%. Inside that, the more important story is that underwater cost of sales rose to $20.0 million, and the company explicitly points to a $2.5 million increase in the estimated cost base of project D, reducing that project's expected gross margin by 11%.

The reason is not some abstract accounting footnote. The company says the estimate changed because of progress in site installation and a better understanding that the complexity of the work and the actual environmental conditions were different from the original plan and from the information previously available. This is the core risk in the model. A fixed-price contract looks attractive when backlog is signed, but it can start consuming margin precisely when the project reaches installation and acceptance.

The embedded read of the second half also sharpens the issue. In the second half of 2025, revenue rose to $22.9 million from $14.5 million in the first half, an increase of about 58.5%. But gross profit rose only about 13.1%, operating profit rose about 19.5%, and net profit actually fell to $2.67 million from $2.94 million. So even when revenue accelerated, profit did not scale with it. That is not a red flag. But it is definitely a yellow one.

There is another point that is easy to miss. Pre-tax profit stood at $6.68 million, of which net finance income contributed $3.17 million. Nearly half of pre-tax profit did not come from core operations, but from finance effects, mainly FX movements, deposit interest, and interest on loans. That does not make the profit "fake," but it does mean the market should be careful about reading 2025 as a clean operating breakout.

The Competitive Position Exists, But It Does Not Remove Execution Risk

The company does have a real moat. In both segments it serves customers that require specialization, regulation, acceptance testing, integration, technical resilience, and long-term relationships. In underwater systems it can at times operate as a sole-source supplier. In RT it sits inside complex systems delivered by prime contractors, which reduces the probability of quick supplier replacement once a product has been embedded.

But that moat does not cancel the rigidity of the model. Some customers can demand changes, delay schedules, or stop work. A large portion of the contracts are fixed-price. Most cash is tied to milestones and sometimes also to acceptance testing at the port, at sea, or at the customer site. So the moat here is a moat of access and stickiness, not a moat of guaranteed margin.

Cash Flow, Debt, And Capital Structure

The cash picture here has to be read through an all-in cash-flexibility lens. In other words, not just how much profit was recorded, but how much cash remained after real working-capital movements, leases, investments, and dividends.

From that perspective, 2025 was much weaker than the net-income line suggests. Operating cash flow came in at negative $6.9 million. Investing cash flow was slightly positive because of shareholder-loan repayment, but financing cash outflow was another $2.98 million, mainly a $2.08 million dividend and $895 thousand of lease-principal repayment. The result was a $9.4 million decline in cash before FX, leaving the company with only $2.6 million of cash at year-end.

So if the post-balance-sheet IPO is temporarily ignored, 2025 did not end with strong cash flexibility. It ended with a respectable accounting profit and a cash balance that had eroded quickly.

Working Capital Flipped, And That Is The Real 2025 Story

The source of that erosion is clear. Trade receivables rose to $7.8 million from $6.0 million, but the real jump came from contract assets, which rose to $17.5 million from $4.7 million. Contract liabilities, by contrast, rose only slightly, to $12.6 million from $12.0 million.

Put simply, the business moved from a position where customers were funding more of the execution through advances and early billings to a position where the company itself was funding more of the work until the billing point.

This is also visible through operational working capital. Based on receivables, inventory, contract assets, suppliers, and contract liabilities, operational working capital swung from about negative $4.4 million at the end of 2024 to about positive $7.6 million at the end of 2025. That is a move of more than $12 million in the wrong direction for cash flow. This was not a marginal fluctuation. It was a change in how growth was being funded.

There Is No Financial Debt, But This Is Still A Guarantee-Heavy Business

This is where another easy market mistake appears. The company has no bank loans outstanding and no utilized cash borrowing lines. That is positive. But it does operate material bank-guarantee lines, and they are already meaningful. At December 31, 2025, utilized facilities stood at NIS 75.5 million out of NIS 91 million of committed lines, while guarantees outstanding stood at NIS 118.3 million.

That means the real balance-sheet question here is less "how much debt exists" and more "how much banking capacity the operating model consumes in order to run." In a project-based company that receives customer advances and has to issue performance and advance guarantees, that is a practical constraint. It does not show up as classic financial debt, but it directly affects growth capacity and bank dependence.

On the other hand, covenant room looks comfortable. Tangible-equity ratio stood at 28.5% versus a 16% floor, tangible equity stood at NIS 27.6 million versus a NIS 10 million minimum, and the ratio of financial debt plus guarantees to tangible equity stood at 1.85 versus a ceiling of 4. So the 2025 issue was not covenant pressure. It was the conversion quality of the business itself.

What Really Changed After The Balance Sheet Date

This is where two wrong readings have to be avoided. First, year-end 2025 should not be read as if the company is still sitting today with only $2.6 million of cash. That is no longer true after the IPO. Second, the IPO should not be read as if it eliminated the working-capital problem. That is also not true. The IPO bought time. It did not resolve whether the large projects can translate into invoices, collections, and cash.

Outlook

Before getting into the forward read, here are four non-obvious findings that determine how 2026 should be interpreted:

First finding: 2025 net income looks stronger than the operating core. Nearly half of pre-tax profit came from net finance income.

Second finding: backlog grew exactly where profitability is still unproven. End-2025 backlog is 84% underwater, in the segment where segment profit barely moved despite the revenue jump.

Third finding: the IPO changed immediate liquidity risk, but not conversion quality. That is an important easing, not a change in the underlying economics.

Fourth finding: 2026 guidance is not guidance for a margin breakout. It is guidance for growth with roughly stable operating margin, which means the market will mostly be measuring execution and collections.

Backlog Is Large Enough To Support A Strong Year

It is hard to argue with the amount of revenue coverage. Backlog stood at $100.4 million at the end of 2025 versus $64.2 million a year earlier. Of that, $49.8 million is expected to be recognized in 2026 and another $50.6 million in 2027 and beyond. The underwater segment accounts for 84% of the backlog, with RT making up the remaining 16%.

There is another nuance here. On one side, the company says timing slippage has so far mostly shown up as quarterly shifts and movements between adjacent years. On the other side, it also says recognition depends on milestone achievement, component availability, acceptance tests, and customer performance under the contracts. So the large backlog is real, but it is not an automatic assembly line.

Management Is Moderating Growth Expectations While Asking For Time To Build The Platform

The 2026 target is about $50 million of revenue and 8% to 10% operating margin. The four-to-five-year target is already about $80 million to $100 million of annual sales at operating margin of about 15%.

That contains an interesting message. From 2024 to 2025, the average growth pace was about 58% per year, but the five-year plan already assumes only 15% to 20% average annual growth. Management is effectively telling the market that the exceptional 2025 pace is not the long-term base case. That is more responsible, but it also sharpens the question of what happens if 2026 fails to deliver even the more moderate growth path.

What Has To Happen Over The Next 2 To 4 Quarters

The company has to pass four clear checkpoints:

1. Revenue: move toward $50 million without relying mainly on accounting timing shifts between periods.
2. Margin: defend at least the 8% to 10% operating-margin range, and preferably show that underwater stops diluting quality.
3. Collections: stop the surge in contract assets and show that backlog is beginning to move through billing and into the cash balance.
4. Execution: get through 2026 without another project D type of surprise, meaning no major cost-estimate revisions in fixed-price work.

If those four things happen, DSIT can begin to look like a small defense-systems platform entering a healthier growth cycle. If not, the market may revert to reading it as a project company growing revenue faster than it generates cash.

Risks

Customer Concentration Is Still Very High

The underwater business depended in 2025 on just two major customers: customer A at 31.3% of company revenue and customer B at 32.6%. In RT, Israel Aerospace Industries accounted for 13.6% and customer D for 10.0%. That means at least 87.5% of annual revenue came from four major customers.

The company tries to soften this by arguing that in underwater these are project customers rather than long-term economic dependencies. There is some truth to that. But economically, concentration is still real. If one of those customers delays, freezes, reduces, or changes requirements, the effect will show up quickly in revenue, collections, or margin.

Fixed-Price Contracts Are A Source Of Value And A Source Of Damage

Project D already showed what can happen when site conditions, installation complexity, or the real environment turn out more difficult than expected. Once the contract is fixed-price, any mistake in cost estimation, including procurement, subcontracting, installation, logistics, testing, or environment, can flow directly into margin.

And that is not the full list. Some customers can demand changes and modifications, some projects include delay penalties, and some contracts depend on acceptance, export approval, and regulatory conditions. In this business, strong backlog is an asset. But strong backlog with weak execution can also become a problem very quickly.

Shekel Exposure, Supply Chain, And Regulation Remain Embedded In The Model

The company's functional and reporting currency is the dollar. Most revenue and future backlog are dollar-linked, but a meaningful portion of costs is in shekels. Shekel depreciation supported reported profitability in 2025, and the company states that clearly. The implication runs the other way just as easily: if the shekel strengthens, part of that tailwind disappears.

The company uses financial hedges and natural hedges from time to time, but it does not claim full elimination of the exposure. In addition, it remains exposed to export delays, component shortages, supply-chain disruption, and budget shifts among defense customers. These are not theoretical risks. They sit directly inside the backlog-conversion cycle.

The Public-Market Governance Layer Is Still Young

The company only listed in March 2026. It cancelled the old shareholder agreement, became a public company, adopted a new compensation framework, and granted options and restricted shares to management. At the same time, as of the report date it still had not appointed an internal auditor and had no independent directors in office. That is not necessarily a material problem by itself, but it is a reminder that DSIT is still building its public-company governance layer while also trying to scale operations quickly.

---

Conclusions

DSIT enters 2026 from a much stronger place than it appeared to be at the end of 2025, but still not from a clean place. The IPO reduced immediate liquidity risk, backlog provides real visibility, and demand in underwater systems looks genuine. The core bottleneck is still the same one: whether the main growth engine can generate both cash and margin, not just more work in progress.

The current thesis in one line: DSIT has already proven demand, but it still has not proven that the underwater business can convert a large backlog into the quality of earnings and cash flow expected from a growing public company.

What changed versus the year-end 2025 read is fairly clear. Then the focus was the cash balance getting squeezed. Today the immediate funding risk is lower because the IPO bought time, and the European contract added another layer of commercial validation. But that shift does not remove the core test. It simply moves the question from "is there enough cash to get through the period" to "is the economics of the growth actually healthy."

The strongest counter-thesis is that the caution here may be too conservative: the company now has good liquidity, no financial debt, a very large backlog, new post-balance-sheet orders, and several products expected to enter serial production during 2026, alongside a medium-term operating-margin target of 15%. If the contract-asset build is mostly timing, cash may simply arrive later and the market may already be correct in looking through the noise of 2025.

What could change the market's interpretation over the short to medium term? Mainly three things: better cash flow and slower growth in contract assets, stable underwater margin after project D, and credible progress toward the $50 million target without another estimate surprise. If the company shows all three, the story can move from noisy growth to higher-quality growth. If not, the market will go back to asking who is really funding the revenue.

Why this matters: in a small defense-systems company, the move from a project business that finances work from its own balance sheet into a business that converts backlog into cash and profit is the move that determines whether a better business is being built, or whether only another layer of volume is being added.

What needs to happen over the next 2 to 4 quarters for the thesis to strengthen? The company needs to show that backlog is beginning to translate into revenue on time, that the underwater business is returning to a more stable margin profile, and that contract assets stop growing faster than revenue. What would weaken the thesis? Another major cost-estimate revision, another working-capital spike, or growth that remains mostly accounting rather than cash-generative.