Skylock: Can the 2026 Backlog Really Turn Into a Comeback
Skylock ended 2025 with an almost 58% revenue collapse, but it entered 2026 with $8.8 million of backlog that had already risen to $9.689 million near the filing date. That is a real base for a recovery, but almost two-thirds of the backlog sits in the fourth quarter and the product layer is still not fully finished, so 2026 remains an execution-proof year.
Why Skylock Needs To Be Isolated
The main article already set the group-level frame: Bizness carried 2025, while Skylock still had to prove it could turn demand back into revenue and profit. This follow-up isolates only that question. Not whether there is interest in Skylock's products, but whether the backlog built for 2026 is actually strong enough to turn a collapse year into a comeback year.
The first number to start with is neither the Japan order nor the new-product roadmap. It is the depth of the 2025 hole. Revenue in the segment fell to $5.226 million, versus $12.423 million in 2024 and $11.790 million in 2023. That is a decline of almost 58% in a single year. Anyone arguing for a 2026 comeback has to explain how the business climbs out of a drop like that, not just point to a few better headlines.
At the same time, it would be too simplistic to read 2025 as a story of vanished demand. Skylock ended 2025 with $8.8 million of backlog mapped into 2026 revenue, versus $6.078 million that had been mapped into 2025 one year earlier. Close to the filing date, that backlog had already risen to $9.689 million. So the 2026 debate is not about whether work exists on paper. It is about the quality of that work, the timing of conversion, and how much execution risk still sits between backlog and recognized revenue.
That chart intentionally mixes a realized number with a number that still has to be delivered. That is exactly the point. 2025 is already closed as a revenue collapse. 2026 is still only a recovery setup.
What The Backlog Does Prove
There is a real constructive read here. Skylock explicitly defines a return to revenue and profit growth as one of its business targets. That is not vague language about expanding into markets. It is an admission that 2025 fell short of what management expects from the business. Beyond that, the company's goals for the coming year include completing a new command-and-control system, integrating a kinetic interception layer into the multi-layer solution, continuing product improvements, and hiring across departments. In other words, the company is not just sitting on old backlog and hoping for better timing. It is trying to widen both the product envelope and the commercial engine at the same time.
The customer base also supports the thesis to a point. Skylock says many of its customers are governments and government-related organizations that operate numerous strategic sites, which means repeat business is part of the model. That matters because a real comeback in this kind of market is not necessarily built on one new customer, but on the ability to deepen relationships with customers that already installed the system and now need additional layers, upgrades, and spare parts.
But the most important datapoint in the revenue disclosure almost cuts the other way. Skylock sells almost entirely products. Services sold each year were negligible, below 1% of the segment's consolidated revenue. That means there is no recurring service layer to cushion another weak quarter. If systems are not delivered, revenue simply does not arrive. This is a key point for backlog quality: backlog matters, but it still has to turn into physical deliveries.
| Layer | What supports the positive read | What is still missing |
|---|---|---|
| Demand | 2026 backlog rose to $8.8 million at year-end and to $9.689 million near the filing date | 2025 itself still ended in a sharp revenue collapse |
| Customers | The company emphasizes strategic customers and expanding activity with existing customers | Repeat business still has to become recognized revenue, not only order headlines |
| Business model | A product-heavy model can produce a sharp rebound when projects are delivered | The same model also creates high volatility when delivery moves to the right |
Why 2026 Is Still Not Locked In
This is where the real problem sits. Even after the backlog recovery, the 2026 recognition profile remains highly back-end loaded. Out of the $9.689 million backlog shown close to the filing date, only $0.3 million is mapped into the first quarter, $2.489 million into the second, and $0.6 million into the third. The fourth quarter alone carries $6.3 million, or about 65% of the updated total.
That is not a footnote. It is the core reason this is still a proof year rather than a clean comeback call. When nearly two-thirds of backlog sits in the fourth quarter, one timing slip can once again push a meaningful part of revenue to the right. In a model where services are negligible and revenue is recognized mainly through product delivery, that concentration leaves very little room for error.
That chart sharpens another point. The improvement between year-end and the filing date did not come from an even stronger fourth quarter, which was already heavy enough. It came from a moderate addition to the second and third quarters. That is helpful, but not enough to make the year look balanced and comfortable. It is still an execution year that leans mainly on the back half.
That is why 2026 for Skylock is not a demand year. It is a conversion year. If the next few quarters do not start showing real revenue before the fourth-quarter bulge arrives, the market will struggle to give full credit to backlog even if the headline number stays large.
What The Product Roadmap And The Japan Order Really Add
It is easy to read the development section as if it solves the problem. In practice it mostly defines what still has to happen before the problem is solved. The new command-and-control software remains under development and is expected, according to Skylock, to be completed by the end of the third quarter of 2026. That detail matters even more because the previous command-and-control layer relied on technology developed by MyDefence, and that IP was sold back. So part of the 2026 effort is not only about adding a new product. It is also about bringing the core command layer back into full group ownership.
The hardware layer is also still moving in parallel. The company describes ongoing cooperation with a leading 4D radar supplier, with capabilities that include operation while moving and detection of hovering drones. In addition, the development and assembly line in Poland has already been established and received all regulatory approvals in 2025. Each of those steps improves the overall product envelope. But together they also say that 2026 is not a year in which Skylock is only harvesting. It is still building.
This is where the Japan order matters. On March 13, 2026, Skylock received an order for spare parts from a Japanese customer totaling about $0.6 million, with 30% paid upfront and 70% on delivery, and delivery scheduled for September 2026. The more important detail is that these are spare parts for systems previously ordered in an aggregate amount of about $4.4 million, most of which had already been supplied. That means two things. First, there is an installed base that is generating follow-on demand. Second, the quality of this order is better than a first sale because it comes from a customer that has already gone through implementation.
But the order also needs to be kept in proportion. $0.6 million equals about 11.5% of Skylock's 2025 revenue and about 6.2% of the updated 2026 backlog. It is an important proof point for repeat customers and for the persistence of the installed base. It is still not an order that can on its own turn the story from potential recovery into a proven comeback.
| Move | What it does prove | What it still does not prove |
|---|---|---|
| New command-and-control software | Skylock is trying to bring the core layer back into full ownership and widen the solution | Commercial completion is still expected only toward the end of Q3 2026 |
| 4D radar layer | The technology envelope is expanding toward more advanced threats | The filing does not yet show a proven revenue engine attached to it |
| Assembly line in Poland | There is operating infrastructure that can add flexibility and redundancy | Infrastructure alone is not revenue if delivery timing slips again |
| Japan spare-parts order | Repeat customer, installed base, 30% advance payment, delivery planned for September 2026 | This is not a new-system order large enough to change the year on its own |
Bottom Line
The 2026 backlog can absolutely turn Skylock from a revenue-disappointment story into a recovery story. There is a real base here: backlog is higher than the one that fed 2025, management explicitly targets a return to growth, the product layer is being rebuilt, and the Japan order proves that at least part of the installed customer base is still ordering.
But as of the filing date, this is still not a proven comeback. It is a conditional one. Almost two-thirds of the updated backlog sits in the fourth quarter, services barely exist as a stabilizing layer, and the new command-and-control software is only expected to be completed toward the end of the third quarter. So the coming year will be judged less by headline order flow and more by two simple proofs: whether Skylock starts showing revenue already in the middle of the year, and whether the new product layer actually arrives on time.
The Japan spare-parts order improves the thesis, but mostly because it strengthens the argument that there is a live customer base, not because it closes the execution gap by itself. The right way to read Skylock today is not "the problem is solved," and not "demand disappeared." The right way is to understand that 2026 has become a proof year: demand is back on the screen, and now it has to come back into the income statement.
Disclosure: Deep TASE analyses are general informational, research, and commentary content only. They do not constitute investment advice, investment marketing, a recommendation, or an offer to buy, sell, or hold any security, and are not tailored to any reader's personal circumstances.
The author, site owner, or related parties may hold, buy, sell, or otherwise trade securities or financial instruments related to the companies discussed, before or after publication, without prior notice and without any obligation to update the analysis. Publication of an analysis should not be read as a statement that any position does or does not exist.
The analysis may contain errors, omissions, or information that changes after publication. Readers should review official filings and primary sources before making decisions.