Carmit: Was 2025 Cash Flow a Real Turn or Mainly an Inventory Release?

What This Follow-Up Is Isolating

The main article argued that Carmit's operating recovery in 2025 was real, but that credit was still the core test. This follow-up isolates a narrower question: does the jump in cash flow from operations to 25.8 million shekels mean profit finally started turning into cash, or was it driven mainly by a working-capital release.

The short answer is that both are true, but not to the same degree. The company did stop bleeding cash. Cash flow from operations moved from negative 5.2 million shekels in 2024 to positive 25.8 million shekels in 2025, and year-end cash rose to 18.4 million shekels from 9.4 million a year earlier. But most of the year-on-year jump did not come from a clean fix in cash conversion. It came from a favorable working-capital turn: inventory fell, suppliers funded more of the cycle, and receivables still consumed cash.

The important number here is not only the headline operating cash flow line, but the quality of the change underneath it. In 2024 working-capital changes used 19.8 million shekels of cash. In 2025 they added 2.9 million. That is a 22.7 million shekel swing, about 73% of the entire improvement in operating cash flow versus the prior year. In other words, most of the repair was balance-sheet driven before it became structurally cash-generative.

Where The Cash Flow Really Came From

Management says explicitly that the increase in operating cash flow in 2025 came from profit and from lower inventory balances and inventory days. That is not a small technical note. It is the center of the story. Once the working-capital lines are broken apart, the picture becomes much clearer.

* Analytical proxy based on receivables + inventory - suppliers, in line with the way the company presents those three items as the core of operating working capital.

This table does not support a reading that collection suddenly improved. Trade receivables rose by 5.6 million shekels to 82.7 million, and average customer credit jumped to 92 million shekels from 62 million in 2024, while customer days stayed at 80 to 90 days. That means Carmit did not free cash through meaningfully faster collection. It sold on broadly similar credit terms, but on a higher nominal base, so receivables still absorbed cash.

Inventory moved the other way. It fell by 10.4 million shekels, from 63.0 million to 52.6 million, and that was the single biggest source of release. That also fits the company’s own disclosure: it holds raw and packaging materials for roughly 45 days and finished goods for roughly 30 days, while material availability and logistical constraints still affect inventory levels. So the decline in 2025 is good news, but it is also not a lever that can be pulled again and again at the same strength. Once inventory normalizes, the next leg of cash improvement has to come from commercial discipline and operating execution, not from another large shelf-clearing exercise.

Suppliers helped as well. Suppliers and service providers rose to 41.1 million shekels from 33.7 million, and average supplier credit increased to 41 million from 34 million. But supplier days stayed unchanged at 60 to 70 days. So there is no disclosure here of a major structural improvement in payment terms. This looks more like higher routine funding on the back of activity and purchasing levels. That helps cash flow, but it is not the same thing as a durable improvement in earnings quality.

What Remains After Real Cash Uses

To avoid confusing cash flow from operations with real funding flexibility, the right lens here is all-in cash flexibility. In 2025 Carmit generated 25.8 million shekels from operating activity after interest and taxes. Against that, it had reported capital expenditure of 5.0 million shekels, lease principal repayments of 2.8 million, and long-term bank debt amortization of 9.0 million. After those real cash uses, roughly 9.0 million shekels remained before any change in short-term bank debt.

This is exactly where two opposite mistakes need to be avoided. The first is to say that 2025 cash flow was fake. It was not. Even after capital expenditure and scheduled repayments, the year still left positive room, and cash on the balance sheet increased sharply. The second mistake is to read the 25.8 million shekel headline as Carmit’s new clean steady-state cash generation. That is also wrong, because without the contribution from inventory release and supplier funding the picture would have been materially smaller.

Put differently, 2025 was a real repair year, but not yet a full proof year. Cash flow came back, but it came back through the balance sheet at least as much as through the income statement.

The New Covenant Turns Working Capital Into A Financing Test

This is where the continuation connects back to the point the main article only compressed. In the updated credit framework signed in March 2026, the company did not just get longer duration and lower credit spreads. It also accepted a new test with Bank Hapoalim under which short-term debt, net of current maturities of long-term debt, cannot exceed 90% of operating working capital.

The company does not disclose the full contractual calculation of operating working capital. But if the balance-sheet bridge uses exactly the components the company itself puts at the center of the working-capital section, receivables, inventory, and suppliers, the proxy comes out at roughly 94.1 million shekels as of December 31, 2025. Against that sits 84.5 million of short-term debt net of current maturities. That implies a ratio of roughly 89.8%.

This needs to be stated carefully: that is an analytical calculation, not a company-reported covenant ratio. Even with that caution, the direction is clear. The updated credit framework did improve pricing and maturity, but at the same time it tied short-term debt directly to working-capital discipline. That means inventory rebuilding, stretched customer credit, or weaker supplier funding are no longer just normal balance-sheet noise. They can become covenant pressure very quickly.

That is the real point of the follow-up. The 2025 cash-flow improvement bought Carmit time, but it also shrank the operating working-capital cushion just before the new credit test started looking directly at it.

What 2026 Still Has To Prove

The next step cannot come from the same lever again. If 2025 was the year when inventory was released and part of the balance-sheet burden was cleared, then 2026 has to show three different things:

• that the company can keep inventory leaner even if raw-material and logistics conditions stay volatile;
• that sales growth does not keep inflating receivables faster than cash collection;
• that the short-term debt to operating working-capital ratio moves to a level with real headroom, not just near-line compliance.

If those three things happen, the cash improvement can start to look more repeatable and less like a balance-sheet reset. If not, 2025 will still count as a good year of cash extraction, but not as full proof that the new earnings base already converts comfortably into free cash.

Conclusion

Carmit’s 2025 cash flow was not an illusion, but it was also not as clean as the headline might suggest. Operating recovery created the floor, yet most of the jump was built on lower inventory and more supplier funding while receivables still demanded more credit in shekel terms. So the real question is not whether 2025 was good. It is whether 2026 can keep cash positive without another meaningful working-capital release.

That matters even more now because the new covenant links short-term debt directly to operating working capital. Once the credit test is built on inventory, customers, and suppliers, cash-conversion quality stops being only an analytical question. It becomes an everyday operating discipline.