Danya Cebus: How Much of the Backlog Really Sits Inside the Africa and Lapidot Ecosystem, and What That Means for Demand Quality

What This Follow-up Is Isolating

The main article argued that Danya Cebus should not be judged only by how much signed work it has, but by how much of that work turns into margin and cash. This follow-up isolates another layer of the same question: demand quality. Not every backlog shekel carries the same meaning. There is a difference between demand won repeatedly from unrelated developers and demand sourced from within the same control ecosystem.

In Danya's case, the notes provide three clear layers of exposure. The first is the backlog line: out of NIS 21.965 billion of order backlog at year-end 2025, NIS 1.643 billion was for Africa Residences and or Africa Residences partnerships. The second is the revenue line explicitly named in the customer-concentration note: in residential construction, companies and partnerships from the Lapidot group generated NIS 460.6 million of 2025 revenue. The third is the wider line: works revenue from interested parties and related parties totaled NIS 766.9 million in 2025.

This is not a fake-demand story. The projects are real, the revenue is real, and the company states that all related-party transactions were conducted on ordinary commercial terms. But it is a different way to read the same numbers. When part of backlog and revenue comes from within the Africa and Lapidot orbit, those figures no longer measure pure outside-market demand in a clean way.

The First Number: NIS 1.643 Billion of Backlog Already Sits Inside Africa

The Africa backlog line is too large to be treated as a footnote. NIS 1.643 billion out of NIS 21.965 billion of total contractor backlog is about 7.5% of Danya's signed work. For a company of this scale, that is not most of the backlog. But it is far too large to ignore when judging what the backlog really proves.

More importantly, the exposure sits in the center of the residential story rather than at the edge of it. The investor presentation markets residential construction through branded flagship projects, and that list includes DUO and Semel Tzafon. In the annual report, both projects are explicitly tied to developer groups that include Africa Residences. In DUO, Danya's share is expected to equal half of total consideration of about NIS 888 million, and in Semel Tzafon total consideration to Danya is expected to be about NIS 430 million. That is an important clue: Africa exposure is not parked in one side project. It also runs through some of the jobs the company itself chooses to showcase as proof of residential depth.

This still needs proportion. About 92.5% of consolidated backlog is not labeled here as Africa Residences. So this is not a single-customer dependency story in the simplistic sense. But it does mean that when backlog is used as a test of demand quality, part of that signal is ecosystem demand rather than purely external demand. That is already a different reading.

The Second Number: Almost One-Fifth of Residential Revenue Comes From the Lapidot Group

The customer-concentration table is even sharper than the backlog note. In residential construction, revenue from Lapidot group companies and partnerships rose from NIS 334.3 million in 2023 to NIS 417.4 million in 2024 and NIS 460.6 million in 2025. This is not only an absolute increase. It is also a rising share of the segment itself, from 15.0% in 2023 to 17.8% in 2024 and 19.5% in 2025.

In plain English, almost one-fifth of residential-construction revenue in 2025 came from within the Lapidot ecosystem. At the full-company level that is about 7.1% of revenue. Inside the residential execution engine, it is a much more meaningful concentration layer.

There is a subtler point underneath that. The wider line for works revenue from interested parties and related parties actually fell in 2025 to NIS 766.9 million from NIS 836.9 million in 2024, yet the Africa-specific line kept rising. That means Africa itself became more dominant inside the wider ecosystem exposure. By 2025 it represented about 60% of total related-party works revenue.

There is also a small spillover beyond residential construction. In the infrastructure customer table, Lapidot group companies and partnerships accounted for NIS 29.6 million of 2025 revenue. That is still only about 2.2% of infrastructure revenue, so residential remains the core issue. But it does reinforce the point that the ecosystem does not stop at one customer table.

This Is Not Only a Revenue Line, It Is a Structural Relationship Layer

If the exposure stopped at one revenue line, it could be read mainly as repeat business. But the related-party note shows something broader. As of December 31, 2025, balances with interested parties and related parties included NIS 184.6 million of advances received versus NIS 77.4 million a year earlier, NIS 251.8 million of customers and contract assets versus NIS 225.3 million, and NIS 157.7 million of loans and investments versus NIS 50.5 million. At the same time, the company has ongoing service agreements across the ecosystem: Lapidot provides management and business-support services, Africa Residences receives payroll and HR services, and Geodanya receives consulting, repair, and equipment services from Lapidot.

That is the key distinction. The ecosystem appears here across several floors at once: backlog, revenue, working capital, loans and investments, and service agreements. The analytical question is therefore not only whether Danya has a related customer, but how deeply that customer relationship is already woven into the company's operating economics.

Precision still matters. The operating service agreements are not large relative to contracting revenue. Management fees to Lapidot were NIS 1.763 million in 2025, and the Africa Residences services agreement sets quarterly consideration of NIS 113.5 thousand, CPI linked. So the story is not that these service agreements move earnings by themselves. The story is that they mark an ongoing, multi-layer relationship that does not begin and end with one construction contract.

Why This Changes Demand Quality

In a construction business, repeat work from a familiar group is not automatically bad news. It can reduce business-development costs, improve manpower planning, support utilization, and create execution continuity. In that sense, the Africa and Lapidot ecosystem is also a moat.

But the same moat carries an analytical cost. When part of the work comes from a corporate cousin, backlog says a little less about Danya's standalone drawing power in the open market. Even if pricing is commercial, execution is strong, and both sides benefit, this is still a demand layer that reflects corporate proximity and not only open-market competition.

That is why the question here is not whether the work is real. It is real. The question is what exactly it proves. A large third-party backlog proves competitiveness, pricing discipline, and market reputation against unrelated developers. An ecosystem backlog proves something slightly different: Danya is a preferred execution arm within the broader group. That is a real business asset, but it is a different quality of demand.

The company tries to put a guardrail around this, rightly, by stating that all transactions with interested parties and related parties were carried out on ordinary commercial terms. That matters. It lowers the risk of unusual pricing or clearly non-market terms. But it does not solve the source question. A market-priced deal can still be an ecosystem deal, and therefore a less pure measure of outside-market demand.

Why This Looks Structural Rather Than Accidental

That read becomes even stronger when looking at the activity-demarcation arrangement between Danya and Africa Residences. Under that arrangement, Danya undertook not to engage by itself in residential development activity in Israel except under defined exceptions and conditions. In several cases, including projects with an external developer or urban-renewal projects above 30 units, Danya is required to offer Africa Residences participation in development rights, sometimes at 100%.

This goes deeper than customer concentration. The arrangement does not automatically guarantee that every opportunity becomes Danya construction backlog, but it does mean that part of the residential opportunity funnel already passes through Africa Residences. More than that, the arrangement is set to expire automatically only if Africa Residences and or Danya cease to be subsidiaries of Lapidot and Africa Investments. In other words, the relationship is not only commercial. It is also structural and corporate.

That leads to the most important conclusion of this follow-up: in Danya, residential demand quality is not tested only in the open market. Part of it is routed through a corporate architecture that connects the company to Africa Residences and Lapidot. That creates stability, but it also reduces the purity of the demand signal.

What Needs To Be Tracked From Here

Three checks are enough to see whether the concern is easing or intensifying.

The first is the Lapidot share of residential-construction revenue. If that share stabilizes or falls while the segment keeps growing, the ecosystem can still be read mainly as support rather than as a heavier crutch. If it moves above the current near-20% level, the warning gets stronger.

The second is the Africa backlog line itself. NIS 1.643 billion is large enough to monitor directly in the next cycle. If it shrinks as a share of backlog because new external work comes in faster, the read improves. If it stays sticky or grows, backlog quality remains more mixed than the headline suggests.

The third is the balance-sheet layer. If advances, contract assets, and loans to related parties continue to rise faster than the business as a whole, the ecosystem is not only a work source. It is also consuming working capital and balance-sheet flexibility.

Bottom Line

Danya Cebus does not suffer from a lack of work, and the Africa-Lapidot relationship is not an accounting red flag. But it does change the way backlog should be read. About 7.5% of consolidated backlog is already labeled as Africa Residences and or its partnerships, almost one-fifth of residential-construction revenue comes from Lapidot group companies and partnerships, and the wider related-party exposure also touches the balance sheet and the governance layer.

So the conclusion is not that demand is weak. The conclusion is that part of the demand is less externally pure than the NIS 22 billion backlog headline suggests. In Danya, the Africa and Lapidot ecosystem is both an operating advantage and a quality filter. Anyone trying to judge real demand quality needs to look not only at how much work exists, but also at who is bringing it in.