Hilan: Why Cash Flow Softened Even as the Balance Sheet Strengthened
Hilan ended 2025 with ILS 379.3 million of operating cash flow, down ILS 43.5 million from 2024, even as net cash jumped to ILS 327 million. This follow-up shows that the gap was driven mainly by weaker customer advances and working-capital timing, while the balance sheet strengthened partly through lower client-trust balances, lighter acquisition cash use, and enough flexibility to keep paying dividends.
The main article already argued that Hilan finished 2025 with higher profit and a stronger balance sheet, but with cash flow that still needed another round of proof. This follow-up isolates that exact paradox: how operating cash flow fell to ILS 379.3 million while cash, net cash, and liquidity ratios all moved higher.
The central point is that this gap does not look like weakening demand or a broad collection problem. It looks much more like a reversal of working-capital tailwinds. In 2024 Hilan benefited from customer advances and a friendlier working-capital setup. In 2025 that layer contributed less, while part of the balance-sheet improvement came from a lower trust-funds layer in US payroll, a layer that reduces both assets and liabilities but does not create new shareholder cash by itself.
That matters because the market can misread the year in both directions. It can focus only on softer cash flow and miss that Hilan still increased cash and cut debt. Or it can focus only on ILS 327 million of net cash and miss that earnings-to-cash conversion actually weakened. The right read sits in between.
Where 2025 Cash Flow Lost Altitude
Net profit actually rose to ILS 262.8 million from ILS 247.6 million, and P&L adjustments rose to ILS 230.7 million from ILS 203.8 million. In other words, before working capital and before cash payments, Hilan built a pre-working-capital cash base of ILS 493.5 million in 2025 versus ILS 451.4 million in 2024. The problem did not start in the accounting core.
The problem started one layer below that. Changes in assets and liabilities moved from a positive ILS 35.7 million in 2024 to a negative ILS 31.8 million in 2025, a deterioration of ILS 67.5 million. The cash-paid line also widened to ILS 82.4 million from ILS 64.2 million. Those two lines explain almost all of the ILS 43.5 million decline in operating cash flow despite higher earnings.
| Component | 2024 | 2025 | What really changed |
|---|---|---|---|
| Net profit | 247.6 | 262.8 | Up ILS 15.1 million |
| P&L adjustments | 203.8 | 230.7 | Up ILS 27.0 million |
| Changes in assets and liabilities | 35.7 | (31.8) | A negative swing of ILS 67.5 million |
| Cash paid during the period | (64.2) | (82.4) | Higher cash outflow |
| Operating cash flow | 422.9 | 379.3 | Down ILS 43.5 million |
This is where the company's own explanation matters. It says the main driver inside working capital was a decline in customer advances received versus the prior year. That is a material distinction. It means 2025 did not weaken mainly because profitability broke down. It weakened because 2024 carried a stronger cash tailwind from the customer side.
More than that, revenue recognized during 2025 from balances previously included in customer advances was ILS 82.0 million, almost unchanged from ILS 80.8 million in 2024. So the advance-payment mechanism did not disappear from the model. What weakened was the pace at which fresh advances refilled that layer.
This Does Not Look Like a Collections Problem
If cash flow had weakened because collections deteriorated, the trade-receivables book should have shown clearer stress. That is not what appears here. Net receivables did rise to ILS 919.6 million from ILS 828.2 million, but the composition actually improved: balances not yet due rose to ILS 854.0 million, or 92.9% of the book, versus ILS 731.7 million, or 88.3%, a year earlier. Total overdue balances fell to ILS 65.6 million from ILS 96.5 million.
Average customer-credit days barely changed, 90 days versus 89 days. So this is not a story of weaker customer quality. It is a story of a business model that naturally carries a lot of working capital. In 2025 the group extended average customer credit of about ILS 896 million, against average supplier credit of only about ILS 273 million. That is still a very profitable model, but it also means cash conversion is highly sensitive to the timing of advances and balance-sheet movements.
There is another indication that the activity base itself did not break. Remaining performance obligations rose to ILS 2.594 billion from ILS 2.562 billion, and the portion expected to be recognized within one year rose to ILS 1.923 billion from ILS 1.838 billion. In other words, the future work base widened even though 2025 got less help from working capital.
It is also worth noticing the other-debtors line. It increased to ILS 22.6 million from ILS 6.4 million, and the note explains that it includes operating services and working-capital financing provided under a cooperation agreement with a third party around the transfer of Gav Systems' consulting activity to Ness. That is not the biggest issue in the story, but it does show that not all cash use sits inside classic trade receivables.
Why The Balance Sheet Still Looked Stronger
In the same year that operating cash flow fell, the balance sheet looked stronger from almost every angle. Cash and cash equivalents rose to ILS 442.7 million from ILS 343.0 million. Bank credit and loans fell to ILS 30 million from ILS 52 million. Net cash rose to ILS 327 million from ILS 188 million. Equity attributable to shareholders rose to ILS 1.264 billion from ILS 1.140 billion. The current ratio improved to 1.25 from 1.13.
But one of the year's most important accounting optics sits exactly here. The company explains that the main decline in current assets came from lower client trust funds in US payroll activity, and that the main decline in current liabilities came from the matching decline in liabilities for those same balances. The trust-funds note makes clear that these are customer monies held in designated accounts for salary and tax payments, with a matching liability on the other side.
That means part of the balance-sheet "improvement" comes from a layer that is relatively neutral for shareholders in the first place. When trust balances come down, both the asset and the liability come down. That shrinks the balance sheet, supports the equity-to-assets ratio, and can make the year-end picture look cleaner without, by itself, proving better cash quality.
Still, it would be wrong to dismiss all of the improvement as optics. The current ratio excluding trust balances rose to 1.31 from 1.20, and net working capital rose to ILS 341 million from ILS 210 million. So the liquidity improvement is real. It just came from several engines working together: more cash, less debt, and a smaller trust-funds layer, not only from cleaner earnings-to-cash conversion.
The All-In Cash Picture Is Still Positive, But Less Clean Than The Balance-Sheet Headline
The right frame here is all-in cash flexibility. The question in this continuation is not how much cash Hilan would generate in a normalized world before all uses. It is how much cash was actually left after the uses that really occurred. The selected materials also do not disclose a separate maintenance CAPEX number, so there is no clean basis here for a narrower maintenance-cash bridge.
In 2025 Hilan generated ILS 379.3 million from operations, used ILS 41.5 million for investing activity, used ILS 229.1 million for financing activity, and absorbed ILS 9.0 million of negative FX on cash balances. Even after all of that, cash still increased by ILS 99.7 million.
The key point in that bridge is not only that the result stayed positive, but why. Investing cash flow dropped sharply to ILS 41.5 million from ILS 182.4 million. The company itself says that the difference reflects the fact that 2024 included the CBS, WideOps, and Europe Cloud acquisitions, while 2025 mainly included the investment in Muvao Boost. Put simply, Hilan enjoyed a much lighter acquisition cash burden in 2025. That is what allowed cash to grow even while operating cash flow moved backward.
On the other side, financing cash flow became heavier than in 2024, mainly because dividends rose to about ILS 115 million from ILS 69 million. That sharpens the contradiction rather than resolving it: management accelerated distributions in the same year that operating cash flow softened. After year-end the company also declared a dividend of ILS 2 per share. That is a vote of confidence in the balance sheet, but it is also a reminder that 2026 will be judged by cash conversion, not only by the cash balance at one reporting date.
The bottom line of this bridge is that Hilan is nowhere near a cash squeeze. Its balance-sheet flexibility is strong. But 2025 flexibility rested not only on clean operating conversion. It also rested on a sharply lower acquisition burden, a lower trust-funds layer, and a cash base that was already strong.
What Matters Now
The sharper read from this follow-up is the following: 2025 does not look like a year in which the business weakened. It does look like a year in which earnings converted into cash less cleanly than the final balance-sheet picture might suggest. That distinction matters because it separates business quality from cash-timing quality.
That is why the 2026 checkpoints are fairly clear. The first is customer advances: do they stabilize and start helping operating cash flow again, or does the 2025 gap persist. The second is cash distribution: can operating cash flow once again carry the dividend pace without another big decline in acquisition outflow. The third is liquidity excluding trust balances: if that keeps improving, it will become easier to argue that balance-sheet strength is improving at the accessible-cash layer as well.
That is also why the next discussion around Hilan should not focus on whether it has cash. It does. The more important question is how much of that cash pile is being rebuilt by repeatable operating cash flow, and how much was rebuilt through timing, a lighter balance sheet, and lower acquisition use. In 2025 the answer is still mixed.