Follow-up to Bubbles: How Much the Omni-Channel Layer Really Adds Beyond Core Retail
The 2025 numbers show that Bubbles' Omni-Channel layer is still not a standalone economic engine. Most revenue and profit still come from classic retail, while AI and partnerships mainly improve operating support and execution.
What This Follow-up Is Testing
The main article argued that Bubbles still stands on a working retail engine while its funding pressure keeps building. This follow-up isolates the open question left behind by that thesis: is the Omni-Channel, data, and AI layer already becoming a separate economic engine, or is it still mainly an operating wrapper around a much more conventional retail business.
The 2025 numbers give a fairly direct answer. For now, it is still a wrapper, not an engine. The company clearly uses AI tools, it clearly operates a partnership model that can lighten inventory exposure and strengthen digital capabilities, and management clearly wants the market to read this as a more innovative structure. But when the analysis moves from language to weight, most of the revenue, most of the profit, and most of the operating stability still sit inside the retail and wholesale business.
Where The Economic Weight Actually Sits
The channel view for 2025 is the right place to start because it strips the story down to simple numbers. The two heaviest channels were still independent wholesale at NIS 16.756 million and physical retail under operating franchise arrangements at NIS 16.027 million. Together that is about NIS 32.8 million, roughly 70% of group revenue. Online retail contributed only NIS 5.677 million, classic franchise contributed another NIS 2.766 million, and the entire partnership layer contributed NIS 5.506 million.
That already says something important. Even if full credit is given to both the online retail channel and the partnership layer, the combined digital-partnership stack still remains below one quarter of group revenue. By contrast, two far more traditional routes to market, wholesale and physical stores, still carry the bulk of the business on their own.
Inside the core retail business, the mix also does not support the idea that online has already become the center of gravity. In 2025, physical retail rose 4.8% to NIS 16.027 million, while online retail fell 14.4% to NIS 5.677 million. Wholesale fell 5.1%, and classic franchise fell 3.7%.
In other words, the Omni-Channel layer has not replaced the old business. At most, it still sits on top of it and tries to improve it.
The Profit Test Is Even Sharper Than The Revenue Test
This is where the thesis becomes hard to avoid. In the segment view, the retail and wholesale segment ended 2025 with NIS 41.226 million of revenue and NIS 4.941 million of segment profit. The partnerships segment ended the same year with NIS 5.506 million of revenue and just NIS 109 thousand of segment profit.
| Segment | 2024 revenue | 2025 revenue | 2024 segment profit | 2025 segment profit | 2025 segment margin |
|---|---|---|---|---|---|
| Retail and wholesale | NIS 40.504 million | NIS 41.226 million | NIS 6.711 million | NIS 4.941 million | 12.0% |
| Partnerships | NIS 6.394 million | NIS 5.506 million | NIS 1.404 million | NIS 109 thousand | 2.0% |
The implication is twofold. First, the partnerships segment generated only about 11.8% of group revenue in 2025. Second, it generated only about 2.2% of total segment profit. That is no longer just a size gap. It is an economic-quality gap. The segment margin in partnerships collapsed from about 22% in 2024 to about 2% in 2025, while the core retail segment fell from 16.6% to 12.0% but still remained the obvious profit engine.
This is the core finding. If the Omni-Channel layer were already a separate economic engine, it should have shown at least a reasonable profit contribution relative to its revenue weight. Instead, it almost disappeared in the profit line.
Why This Is Not Just A “Small Segment” Problem
One easy defense would be to say the partnerships business is simply too small to judge yet. That explanation only goes so far. The problem is not just that the segment is small. The model that should have been lighter on working capital and structurally leaner still did not produce clear value in 2025.
Under the partnership model, brand owners are responsible for supplying inventory and retain ownership until final sale to the end customer. BNC manages the online commerce layer, charges customers, passes the agreed share of revenue back to brand owners, and can return unsold products. In plain terms, this should be a lighter layer than ordinary retail, with less hard capital tied up on the shelf and less end-customer credit risk.
That is exactly why NIS 109 thousand of segment profit in 2025 looks especially weak. If even a relatively light model on paper cannot show separate earnings power, it is hard to argue that this layer is already reshaping group economics.
The company itself leaves an important clue. In the board commentary, it says the year included a transition to Omni-Channel cooperation structures and a decline in partnership lines tied only to the online channel. In the same section, it also says the contribution of partnerships to gross profit declined because some collaborations with low operating profit but 100% gross margin were reduced. That line matters because it shows how this layer can look attractive at the gross-profit level while being much less convincing once the analysis runs all the way down to operating profit.
There is one more point that should be treated explicitly as an inference, not as a directly reported split. The company says Miyo Scent added about NIS 4 million of revenue in 2025, while the entire partnerships segment ended the year with NIS 5.506 million of revenue. That strongly suggests that a large part of the segment in 2025 no longer came from the legacy fashion-partnership layer alone. The company does not provide a formal split between Miyo Scent and the older partnership activity, so the exact breakdown cannot be stated as fact. But the economic direction is still clear: even after adding a new growth leg, the segment as a whole left almost no profit behind.
AI Probably Improves Process, But It Still Does Not Create A Separate Earnings Layer
It is important not to fall into either lazy extreme here. On one side, the AI language is not just cosmetic. The company does describe the use of AI tools in design, photography, customer service, analysis, and marketing. It also says that, as part of this shift, it updated roles and reduced employees in line with new capabilities. In the board commentary, general and administrative expenses also declined versus 2024, partly because salary and headcount costs fell by about NIS 1 million and technology costs were cut by NIS 450 thousand.
On the other side, this still does not look like a distinct revenue engine or profit engine. At group level, direct headcount still rose from 28 at the end of 2024 to 34 at the end of 2025. Marketing and sales stayed flat at 15 employees, logistics and operations rose from 5 to 9, and development rose from 3 to 4. So AI probably helps internal efficiency and lowers part of the process cost, but it has not yet produced the kind of labor or margin step-change that would alter the company’s economics on its own.
The right reading of the AI layer in 2025 is therefore more modest. It is an operating technology that strengthens the business system, not yet an earnings layer the market can value on its own. It may improve decision-making, marketing, inventory handling, and employee training. It still has not proven that it deserves platform-style economics.
What Has To Change For This Story To Look Different
If the Omni-Channel layer is going to deserve engine status rather than wrapper status, three things need to happen together:
- The partnerships segment needs to return to meaningful segment profit, not just revenue, and certainly not NIS 109 thousand of profit on NIS 5.506 million of revenue.
- Online retail needs to stop shrinking against physical retail and wholesale. Without that, it is hard to argue that the data and AI stack already changes demand or pricing power in a visible way.
- The company needs to show that the partnerships layer, including the new activity added through Miyo Scent, creates value beyond simply adding another operating layer. Right now it adds complexity faster than it adds profit proof.
Conclusion
The main article argued that Bubbles still rests on a retail engine that works while financing remains stretched. This follow-up sharpens the second half of that equation: the Omni-Channel layer still does not offer a true economic alternative to the core retail business. It mostly acts as a support system around it.
There is a real platform here, there is real technology inside the operating model, and there is a real logic to the partnership structure. But 2025 shows that profit still comes from the trade business, not from the narrative. Until partnerships and online regain separate profitability and more visible weight, the value of this layer should be judged mainly by what it does for the core business, not by what it promises on top of it.
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.