Dimri: The Real Economics of Bavli and Sde Dov

What This Follow-up Is Isolating

The main article already argued that Dimri's reported margin still looks strong, but the real cash test has moved to Bavli and Sde Dov. This follow-up isolates only those two projects, because that is where nearly NIS 943 million of expected gross profit sits in Dimri's share, but also where most of the friction sits between tabular profit and profit that still has to survive financing, execution, and project structure.

Three points stand out immediately. At Bavli, the same roughly NIS 415 million of gross profit appears once on expected revenue of NIS 1.815 billion and a 23% gross margin, and again in the presentation on a net revenue base of roughly NIS 1.343 billion and a 31% gross margin. At Sde Dov, the roughly NIS 528 million of expected gross profit refers only to the residential component, while the 70 hotel rooms and commercial area are left outside the quantified economics. And in both cases, the projects are still very early: 3% completion at Bavli, 1% at Sde Dov, and more than NIS 2.08 billion of cost still left to complete.

So the question here is not whether there is value. There is. The question is what kind of value it is. At Bavli, the main friction is project structure and the way part of the economics leaks to a third party. At Sde Dov, the main friction is that financing has already started to eat into the project long before deep execution and delivery.

The Numbers Before the Story

That chart does more work than a long paragraph. At Bavli, almost all of the profit is still future tense: no signed sales at year-end, only NIS 308.5 million of cost invested so far, and more than NIS 1 billion still left to complete. At Sde Dov, the picture is only partly further along: 32 units have already been sold and project financing has already been fully drawn, but engineering completion is still just 1% and the remaining cost is still close to NIS 1 billion. In other words, both projects are already large enough to explain the thesis, but still too early to resolve it.

Bavli: Margin Looks Different Once the Revenue Base Changes

The first misleading number at Bavli is the gross margin. In the annual project table, the company shows expected revenue of NIS 1.815 billion, expected cost of NIS 1.400 billion, and expected gross profit of NIS 415.4 million, which means a 23% gross margin. In the presentation, the same roughly NIS 415 million of gross profit is shown against revenue of roughly NIS 1.343 billion, so the gross margin jumps to 31%.

That is not an accounting contradiction. It is a reminder that Bavli has to be read through the deal structure before it is read through the margin. In the project note, the company says it is obligated to pay a third party 26% of apartment sale proceeds for 162 units, and 45% of the profits on the remaining units. In return, that third party bears betterment levies, legal fees, and Sale Law guarantee commission on part of the units. The presentation chooses to show the economics on a net basis, so the revenue base gets smaller and the margin looks sharper.

That chart is the core of the Bavli story. If you read only the 31% in the presentation, Bavli can look like an unusually rich project. If you read only the 23% in the annual table, you can miss that some of the dilution comes from structure rather than from weak pricing. The right read is different: Bavli is a project with good economics, but not one where every shekel of revenue stays with Dimri.

The second problem is timing. As of 31.12.2025, no contracts had been signed, completion stood at 3%, and NIS 1.091 billion of cost was still left to complete. The company has already begun self-performed execution, which is usually a source of advantage for Dimri, but here self-build also means the project has not yet passed the market test. The big profit already sits in the forecast table, but the cash has not started to come back.

Even the disclosed day-one financing is still fairly small relative to the size of the move. At the end of 2025, the company shows specific project financing of NIS 152 million from Discount Bank, at prime plus a margin of up to 0.35%, with no financial covenants, and not on a non-recourse basis. That does not mean Bavli cannot be funded. It does mean the gap between NIS 415 million of expected profit and the actual path to harvest it is still very large, and it will have to run through sales pace, construction finance, execution, and eventual release of excess cash.

Sde Dov: The NIS 528 Million Does Not Sit on Top of Free Funding

At first glance Sde Dov looks like the richer project. The company shows expected revenue of NIS 2.654 billion, expected cost of NIS 2.126 billion, and expected gross profit of NIS 527.8 million, a 20% gross margin. But here too the small print matters. The company explicitly says those numbers relate only to the residential component, excluding the commercial area and the 70 hotel rooms that are part of the project. So the NIS 528 million is important, but it is still not the full economic picture of the site.

The maturity profile is also still very early. By the end of 2025, 32 units had been sold, the marketing rate was 7%, engineering and cost completion stood at 1%, and a construction contract had still not been signed. In the presentation published two days after the annual report, another 5 contracts and one purchase request worth about NIS 40 million including VAT were added. That means there is early traction. It is still far away from a project that has started to release cash.

This is where Sde Dov differs from Bavli. At Sde Dov, financing is no longer just a future question. It is already in the numbers. To finance the land purchase, the company took an NIS 825 million loan. Of that, NIS 275 million is for 24 months at prime plus 0.5%, and NIS 550 million is for 24 months with no stated coupon, both due on December 2, 2026. The company itself says part of the consideration was effectively attributed to the interest value of that loan, and a separate asset was recorded for prepaid interest and amortized over the life of the loan.

The economic meaning of that disclosure shows up in two more places. In the finance-expense note, the company records NIS 34.3 million of imputed interest expense in Sde Dov in 2025. And in the financial-instruments note, it says the NIS 550 million land loan carries an accounting effective rate of 6.5%, while its fair value at the end of 2025 is only NIS 516.993 million, versus a carrying value of NIS 550 million.

That is exactly where Sde Dov departs from the simple "expensive land, big gross profit" story. The loan that looks free is not free. It simply does not arrive as ordinary running cash interest on the full NIS 550 million. It arrives through effective interest, through amortization of the prepaid-interest asset, and through financing cost that is already rolling through the project. That is why the 20% gross margin at Sde Dov cannot be read independently of the funding map.

Two more frictions matter here. The first is land-related. In March 2026, the Ministry of Environmental Protection approved that PFAS concentrations in most of the site did not exceed threshold values, but said a small area still required additional investigation in order to define the boundaries and any excavation and disposal volumes, if needed. That is not enough to break the thesis. It is still one more reminder that Sde Dov has not yet moved into a clean, frictionless execution track. The second is the guarantee layer. In the guarantees note, the company shows unlimited parent guarantees for Sde Dov-related debt to Leumi and Discount, NIS 412.5 million to each bank, and in addition a NIS 398.337 million loan from the International Bank tied to the Sde Dov project. Even without fully unpacking every leg here, the message is clear: Sde Dov is not a sealed financing island. It still sits inside a wide parent-support layer.

What This Says About the Funding Map

Taken together, Bavli and Sde Dov hold nearly NIS 943 million of expected gross profit. That is the number that gives them strategic weight inside the Dimri thesis. But the path to that number is very different in the two projects, and so is the type of friction.

At Bavli, the main bottleneck is economic structure. The numbers are good, but not all of the value stays with the company. The project has not yet started converting its story into signed sales, and a large part of its funding map will only be tested once marketing starts to translate into advances and execution pace. This is less a question of whether there is margin, and more a question of how much of that margin actually remains with Dimri and when it becomes cash.

At Sde Dov, the main bottleneck is the price of time. Sales have already started, the land has already been financed, and capitalized financing is already building. But precisely because of that, Sde Dov is more exposed to a long bridge period: funding that is already running, profit that is still unrecognized, and execution that has barely moved. The question here is not only whether units will sell, but whether sales and build-out will move fast enough for financing economics not to keep eating into value along the way.

That is why the combination of the two projects matters so much for reading Dimri. Bavli still has to prove that what looks good in the gross-profit table can also survive the project structure and the funding path. Sde Dov already has to prove that financing will not weigh too heavily on what looks good in the gross-profit table. Together they send the same message: value at Dimri can no longer be judged only by the headline gross margin. It has to be judged by the quality of the path from land, funding, and execution to actual excess cash.

Conclusion

The big number, nearly NIS 943 million of expected gross profit, is real. But it is not clean. At Bavli it runs through a revenue and profit sharing structure with a third party. At Sde Dov it already carries an economic financing cost that does not look, at first glance, like ordinary interest. And across the two projects, more than NIS 2.08 billion of cost was still left to complete at the end of 2025.

That is why Bavli and Sde Dov are not just the biggest value engines in Dimri's thesis. They are also the place where the reader should be least naive. At Bavli, the market now has to see marketing turn the project structure into real cash. At Sde Dov, the market has to see execution and sales move fast enough for financing cost not to keep chewing into future profit. Until that happens, these two Tel Aviv projects represent less "profit that is already ours" and more a test of funding, timing, and structure.