Alarum: AI Pushed Revenue, but Working Capital Swallowed the Cash
The main article showed that Alarum’s 2025 growth came from new AI and data products. This follow-up isolates why the cash did not follow: trade receivables jumped by $8.6 million, while contract liabilities and trade payables barely helped, and cash taxes plus lease cash pushed the bridge deeper into negative territory.
Where The Cash Actually Got Stuck
The main article already set the broad frame: in 2025 Alarum proved that it has new demand tied to AI and newer data products, and revenue rose 28% to $40.7 million. This follow-up does not revisit the margin story or the customer-concentration issue. It isolates the pipe between reported revenue and cash, because that is where the precise explanation sits for why a growth year ended with negative operating cash flow.
This is not simply a case of weak accounting profit. Before working-capital movements and taxes, Alarum generated about $5.0 million in 2025: $963 thousand of net profit plus $4.1 million of non-cash adjustments. Almost all of that was then absorbed by the balance sheet, mainly through trade receivables. After all asset and liability movements, only $270 thousand of cash generated from operations was left, and after $2.284 million of cash taxes paid, operating cash flow fell to negative $2.014 million.
This is also not a negative-working-capital story in the classic sense. The balance sheet still ended the year with positive working capital. The problem is that the next dollar of growth flowed first through the receivables line, not through the cash line. That is why the headline that “AI brought growth back” hides the more important question: who financed the bridge period.
| Cash checkpoint | 2024 | 2025 | What really changed |
|---|---|---|---|
| Revenue | $31.8 million | $40.7 million | Growth returned at 28% |
| Trade receivables, net | $3.231 million | $11.796 million | A jump of $8.565 million |
| Cash generated from operations before tax | $8.891 million | $0.270 million | Almost the entire cash engine was eroded before tax |
| Cash taxes paid | 0 | $2.284 million | A real cash outflow that does not show up in the revenue headline |
| Actual lease cash payments | $0.405 million | $0.690 million | A recurring drag that grew in a weak cash year |
That chart is the reason the right framing here is all-in cash flexibility rather than stopping at profit or EBITDA. On an operating basis before working capital, the company did generate cash. On a basis that asks how much cash was left after working capital, taxes, and actual lease cash, 2025 was already negative.
This Looks More Like Timing Than Visible Credit Stress
The first red flag is that the receivables jump looks much more like customer-term financing than like an obvious first sign of credit deterioration. The revenue note says the company either gets paid upfront or extends credit, mainly to large customers, with payment terms generally up to net 90 days. In the same year, the company says the $8.9 million revenue increase came from strong demand by a large-scale customer building foundational AI models and from sales of new products. That is the connection that matters: the more 2025 growth leaned on large new customers, the more that growth sat on customer credit rather than immediate cash.
The receivables note reinforces that reading. Gross open accounts rose to $12.040 million from $3.291 million, while the expected-credit-loss allowance was only $244 thousand, up from $60 thousand a year earlier. This does not yet look like a classic bad-debt story. It looks like a year in which invoices and customer credit ran ahead of the cash actually collected.
That gap is the core of the issue. Revenue added $8.9 million. Trade receivables alone added $8.565 million. In other words, almost the entire top-line increase found its way, at least temporarily, into the asset side of the balance sheet.
There is a second layer here as well. Starting in 2025, Alarum also generated revenue from ready-to-use datasets, and dataset revenue is recognized at a point in time when the data is delivered to the customer. That does not mean every new dollar of dataset revenue landed in receivables. It does mean that the shift toward newer growth engines increased the chance that revenue would be recorded faster than the cash would arrive.
Customers Did Not Prefund The Growth, And Suppliers Barely Helped
If this were only a routine timing issue, one would expect some meaningful offset from customers or suppliers. That barely happened. Contract liabilities rose only to $2.431 million from $1.987 million, an increase of just $444 thousand. On the cash-flow statement, trade payables added only $51 thousand.
That point matters. In 2025 Alarum did not benefit from a picture in which customers paid heavily in advance or suppliers financed a meaningful part of the step-up. The cash simply stayed on the customer side for longer.
| Offset layer versus receivables | 2025 change | The right read |
|---|---|---|
| Trade receivables, net | Negative $8.565 million | This was the main cash absorption |
| Contract liabilities | Positive $0.444 million | Customer prepayments grew far too little |
| Trade payables | Positive $0.051 million | Suppliers barely carried the financing |
| Other payables | Positive $4.564 million | There was some offset, but not one that solved the issue |
What is more interesting sits inside that partial other-payables cushion. On the year-end balance sheet, other payables and accruals consisted mainly of $4.127 million of accrued expenses and $3.677 million of employees and related institutions. By contrast, the provision for income taxes fell to $126 thousand from $1.409 million. So even the layer that did soften the hit does not read like comfortable supplier financing or tax deferral. It reads more like expenses and employee obligations building up while receivables had already expanded.
That is why the sentence “working capital weakened, but payables rose” sounds more comforting than it really is. Payables did rise, but they did not create a symmetric financing mechanism against the receivables jump. They only reduced part of the hole.
After Tax And Lease Cash, Even That Cushion Was Not Enough
The easiest thing to miss is that the story does not stop at working capital. After all balance-sheet movements left only $270 thousand of cash generated from operations, the company then paid $2.284 million of taxes in cash. That is what pushed operating cash flow to negative $2.014 million.
Then comes the lease layer. In 2025 financing cash flow included $102 thousand of lease interest and $588 thousand of lease principal repayments, or $690 thousand together. In a normal year that is manageable. In a year when the operating bridge, after working capital and tax, was already negative $2.0 million, even a lease layer of that size matters.
The issue is not only 2025 itself. The lease note shows that in June 2025 the company signed a new office lease that started on September 1, 2025, with monthly rent of about $58 thousand until the end of February 2027 and about $75 thousand until the end of August 2029. Accordingly, lease liabilities rose to $2.639 million at the end of 2025 from only $620 thousand a year earlier. That means lease drag was not just a small hit to 2025 cash. It also created a higher cash burden for the coming years.
It is important to be precise about the cash framing. On a normalized / maintenance cash generation view before working capital, 2025 still looks reasonably solid, at about $5.0 million before balance-sheet movements and taxes. But on an all-in cash flexibility view around operations, meaning after working capital, taxes, and actual lease cash, the year was already negative $2.7 million even before reported CAPEX, the purchase of a new intangible asset, or any other capital-allocation choice. That is the exact gap between accounting growth and real cash flexibility.
Conclusion
This follow-up does not change the main article’s thesis. It makes it more precise. AI demand did push Alarum’s revenue higher, but in 2025 it pushed revenue through a heavier working-capital layer, not through clean conversion into cash.
The stricter read here is not that Alarum lacks customers or near-term liquidity. It is that the company financed the new growth wave through receivables while customers did not advance enough cash, suppliers gave almost no counter-financing, and tax and lease payments took what was left. If 2026 shows stabilization in receivables, better operating cash flow, and proof that the new engines can work under tighter collection discipline, 2025 will look like a bridge year. If not, it will remain a year in which revenue grew much faster than cash.
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.