Alarum Technologies Fourth Quarter and Full Year 2025 Earnings Call - AI demand fuels rapid top-line lift, but margins compress as company spends to scale
Summary
Alarum says 2025 was transformational. Q4 revenue jumped 60% year over year to $11.8 million and full-year revenue rose 28% to $40.7 million, driven by a sudden shift into large-scale AI customers and new AI-focused products that grew to roughly 30% of revenue from 4% a year earlier. Management frames the margin hit as deliberate: heavy investment in infrastructure, doubled headcount, and product expansion to capture a bigger, longer-term opportunity in AI data infrastructure.
The flip side is volatility. Gross margins slid sharply as AI workloads require costly capacity and third-party services, adjusted EBITDA and net profit both fell versus 2024, and the business saw quarter-to-quarter swings tied to customers timing model development cycles. Management remains profitable, with a clean balance sheet and guidance for Q1 2026 around $11 million in revenue and roughly $1.4 million adjusted EBITDA, but warns that near-term revenue patterns will stay lumpy while the company scales.
Key Takeaways
- Q4 2025 revenue $11.8 million, up ~60% year over year; full-year 2025 revenue $40.7 million, up ~28% vs 2024.
- New AI-focused products accelerated to ~30% of 2025 revenue, up from ~4% in 2024, making AI customers the primary growth engine.
- Platform workloads exploded from about 3-4 petabytes per month at the start of 2025 to as much as 70 petabytes per month by year end, creating major operational complexity and scale requirements.
- Gross margin compressed materially: Q4 gross margin 53.8% vs 72.4% a year earlier; full-year gross margin 58.5% vs 75.1% in 2024, driven by higher infrastructure and third-party costs tied to large AI projects.
- Adjusted EBITDA fell to $4.4 million for full-year 2025 from $9.4 million in 2024; Q4 adjusted EBITDA was $1.0 million vs $1.5 million in Q4 2024.
- Net profit declined to $1.0 million for the year from $5.8 million in 2024; Q4 net profit was $0.2 million vs $0.4 million in Q4 2024. EPS also contracted (Q4 basic ADS $0.03 vs $0.06).
- Management says margin compression is intentional, driven by strategic investments made to win large, enterprise AI customers and build platform capabilities, with margin improvement initiatives underway and expected to progress through 2026.
- Balance sheet remains clean: cash equivalents and debt investments approx. $22.5 million at 12/31/25 (down from $25 million), zero debt, shareholders equity $32.1 million. Outstanding share count about 72.5 million ordinary shares (7.25 million ADS).
- Q1 2026 guidance: revenue about $11 million ±7% (implies ~46% YoY growth) and adjusted EBITDA roughly $1.4 million ±$0.5 million, guidance based on current orders, backlog and consumption trends.
- Company is transitioning from a proxy-centric business to a diversified data infrastructure platform, investing in global proxy network, platform capacity, self-service, datasets, and enterprise support functions to win deeper, stickier customer relationships.
- Management emphasizes revenue variability driven by timing of model training cycles and dataset refreshes, not a structural demand decline. They describe the current pattern as lumpy but positive long-term secular demand for training data.
- Operational progress includes doubling headcount, office expansion, and ongoing work to bring more infrastructure in-house and optimize network economics; management expects efficiency gains but says investments will continue through 2026.
- Large anchor relationships, including an Asian online marketplace and several major global AI developers, are described as stable and ongoing contributors, though customers intentionally split workloads across multiple suppliers, limiting single-customer concentration upside.
- NRR impact: legacy use cases slowed even as AI cohort expanded, producing NRR under 1 in places despite overall revenue growth, reflecting customer mix shift rather than broad churn.
- Management projects a pathway to a >$100 million revenue run rate over time if AI spending patterns persist, but offers no firm timetable and cautions that quarter-to-quarter swings are typical while customers move through development stages.
Full Transcript
Operator: Ladies and gentlemen, thank you for standing by. Welcome to the Alarum Technologies fourth quarter 2025 earnings results conference call. During today’s presentation, all parties will be in a listen-only mode. Following management’s presentation, the conference will be open to questions. If you have a question, please press the star key followed by the number 1 on your touchtone phone. If you would like to withdraw your question, please press the star key followed by the number 2. If you’re using a speaker, please lift the handset before making your selections. This conference is being recorded. I will now turn the call over to Kenny Green, Investor Relations at Alarum Technologies. Kenny, please go ahead.
Kenny Green, Investor Relations, Alarum Technologies: Thank you. Good day to all of you, and welcome to Alarum’s conference call to discuss the results of the fourth quarter and full year 2025. I would like to thank management for hosting this call. Today, we are joined by Shachar Daniel, Alarum’s CEO, and Shai Avnit, CFO. Shachar will begin the call with an overview of the fourth quarter, followed by Shai, who will review key elements of the financials. Finally, we will open the call for the question and answer session. Before we get started, I want to highlight the forward-looking statements disclaimer. This conference call may contain, in addition to historical information, forward-looking statements within the meaning of the Safe Harbor Provisions of the Private Securities Litigation Reform Act of 1995 and other federal securities laws.
Forward-looking statements include statements about plans, objectives, goals, strategies, future events of performance, and underlying assumptions and other statements that are different than historical facts. For example, when we discuss a strategy of prioritizing long-term leadership and market share capture over short-term margins and profitability, expected trends in market demand and AI-driven growth, our business momentum and pipeline, our expectations regarding future revenue patterns and margin improvement, the anticipated impact of our strategic investments and product mix, and our estimates regarding fourth quarter 2025 revenues and adjusted EBITDA, we are using forward-looking statements. These forward-looking statements are based on the current management expectations and are subject to risks and uncertainties that may result in expectations not being realized and may cause actual outcomes to differ materially from expectations reflected in these forward-looking statements.
Potential risks and uncertainties include those discussed under the headings Risk Factors in Alarum’s annual report on Form 20-F filed with the Securities and Exchange Commission on March 20, 2025, and in any subsequent filings with the SEC. All such forward-looking statements, whether written or oral, made on behalf of the company, are expressly qualified by these cautionary statements. As such, forward-looking statements are subject to risks and uncertainties, and we caution you not to place undue reliance on them. On the call, the company will also present non-IFRS key business metrics. The non-IFRS key business metrics the company uses are EBITDA and adjusted EBITDA, non-IFRS gross margin, non-IFRS net profit or loss, and non-IFRS basic earnings or loss per share or ADS.
The exact definitions and reconciliations of these non-IFRS key business metrics are described in the company’s financial press release, which is available in the investor’s lobby of Alarum’s website. These non-IFRS measures may differ materially from similarly titled measures used by other companies and should not be considered in isolation from or as a substitute for financial information prepared in accordance with IFRS. Now I’d like to turn the call over to Shachar Daniel, Alarum Technologies CEO. Shachar, please go ahead.
Shachar Daniel, Chief Executive Officer, Alarum Technologies: Thank you very much, Kenny. Good morning, everyone. Thank you all for joining us today. We saw our 2025, which was a key year for us, for Alarum, and Q4 capped it with another strong quarter. Fourth quarter revenue were $11.8 million, up 60% year-over-year, while full year revenues reached nearly $41 million, up 28%. Even as we invested a lot in increasing our capabilities throughout the year, we remain profitable. We reported fourth quarter net profit of $0.2 million and adjusted EBITDA of $1 million. For the full year, we reported net profit of $1 million and adjusted EBITDA of $4.4 million. This strong growth reflects the shift we saw during the year as demand from AI-focused customers become a key growth driver.
In 2025, our new AI-focused products, which are our growth engines, accounted for about 30% of revenues, growing from around 4% only last year. As AI systems becomes more complex, demand for large scale, high quality public web data continues to grow. At the same time, collecting reliable data at scale becomes more difficult, raising barriers to entry and increasing the value of our infrastructure, technology and execution. During 2025, we significantly expanded our work with several major global technology companies developing AI systems. These relationships involve large scale data collection for model training, data set creation and ongoing refinement. We also saw a dramatic increase in workloads across our platform, with volume rising from between 3-4 petabytes per month at the beginning of the year to up to 70 petabytes at the end of this year.
This sharp increase created greater operational complexity, and our ability to perform at this scale become an important competitive advantage for Alarum. Strategic investment and profitability. Alarum invested significantly in its capabilities and scaled the business during 2025. We doubled our headcount, expanded our offices, and strengthened our organization across R&D, products, support, customer success, and account management. From a technical standpoint, we strengthened our global proxy network, increased platform capacity, and enhanced our data collection abilities to support massive AI workloads and the growing complexity of enterprise-scale deployments. This investment now allow us to better support customers operating at massive scale, including both existing major customers and new target customers. The higher investment levels have led to lower short-term margins, although we remain profitable. These margin pressures were planned and are addressable.
As I noted last quarter and even before, we have several initiatives underway to improve margins over the coming quarters. These strategic investments have positioned Alarum at the center of the AI ecosystem as we move through 2026. The global AI market is still in its early stages, and that naturally creates some near-term variability in demand. Large AI developers shift spending sharply based on where they are in their model development cycles and the timing of dataset refreshes. While the year-over-year growth trend in spending on data is clear, revenue from major customers may vary from quarter to quarter. At the same time, we believe this remains the early stages of a much larger opportunity for Alarum.
As demand broadens, our customer base expands and becomes more diversified, and model development moves into more structured training cycles, we expect Alarum’s revenue pattern to become smoother and more predictable. Product platform expansion. Our business continues to evolve from a proxy-focused company into a diversified multi-product data infrastructure platform. During 2025, this evolution progressed across websites and blockers, self-service solutions, datasets, and our core proxy infrastructure. This shift from a single product proxy model to a broader platform is expected to support stronger long-term margins and healthier unit economics. Strategically, it also reflects move from a traditional zero-touch online model to a deeper enterprise relationships that requires dedicated support, performance monitoring, documentation, and long-term infrastructure and collaboration. While this increases operational complexity, it also creates stronger customer relationships and a more strategic role in customers’ operations.
We believe this evolution is expanding our value proposition and improving the quality of our revenue base over time. In summary, 2025 was a transformational year for Alarum. We observed growth, expanded customer relationships, scaled our infrastructure, broadened our product platform, and further positions the company to serve the rapidly growing AI data infrastructure market. At this stage, we view Alarum as being in a company and a platform building phase. Our primary focus is on strengthening leadership, expanding infrastructure and product capabilities, and deepening enterprise customers’ relationships. We believe this positions us well for the larger market opportunity ahead. I believe Alarum is on a path to achieving a revenue run rate exceeding $100 million as the market continues to develop. I will now hand it over to Shai for the financial details and guidance for the current quarter. Shai.
Shai Avnit, Chief Financial Officer, Alarum Technologies: Thank you, Shachar, and hello, everyone. I will start by reviewing our key financial results for the fourth quarter and full year 2025, comparing them to the same periods last year, unless otherwise noted. Following that, I will provide our guidance for the first quarter of 2026. Detailed definitions and reconciliations of our non-IFRS key business metrics can be found in our fourth quarter and full year 2025 financial results press release. One final note before I begin, the figures I will be discussing are rounded for clarity and ease of reference. Turning now to our financial performance. Revenues. Revenues in the fourth quarter of 2025 reached $11.8 million, compared with $7.4 million in the fourth quarter of 2024, an increase of approximately 60% year-over-year.
For the full year of 2025, revenues reached $40.7 million, compared with $31.8 million in 2024, an increase of approximately 28%. As Shahar mentioned, the growth was driven mainly by strong demand from large-scale customers building foundational AI models, as well as increased sales of our new AI-focused products. Gross profit and gross margins. Gross profit in the fourth quarter of 2025 amounted to $6.4 million, compared with $5.3 million in the fourth quarter of 2024. For the full year, gross profit was $23.8 million, compared with $23.9 million in 2024.
Gross margin in the fourth quarter of 2025 was 53.8%, compared with 72.4% in the fourth quarter of 2024. For the full year, gross margin was 58.5% compared with 75.1% in 2024. The decline in gross margin reflects our work with large scale AI customers, which require data gathering at significantly greater scale and involve higher initial infrastructure costs, including a larger volume of servers and stronger, higher quality infrastructure. In addition, the new product sales have triggered related third-party costs, which we expect to decrease in the coming quarters. Overall, this is consistent with our strategy to pursue large scale, highly strategic opportunities that we believe can drive meaningful long-term growth and profitability, even at the cost of lower short-term margins.
I note that as our revenue continues to grow, the company will increasingly benefit from margin increases due to the operating leverage inherent to our business model. Operating expenses. Operating expenses in the fourth quarter of 2025 were $6.4 million, compared with $5 million in the fourth quarter of 2024. For the full year, operating expenses were $23.6 million compared with $17.2 million in 2024. The increase was driven by higher revenues and broader operations, primarily attributable to research and development expenses and to a lesser extent, sales and marketing expenses. Again, this increase is the key part of our strategy to invest in our infrastructure and capacity in order to position the company to capture the significant long-term growth opportunities we see ahead of us. Net profit.
Net profit in the fourth quarter of 2025 was $0.2 million, compared with $0.4 million in the fourth quarter of 2024. For the full year, net profit was $1 million, compared with $5.8 million in 2024. While we still maintained our profitability, the decline in profit reflects the increased strategic costs. Adjusted EBITDA. Adjusted EBITDA in the fourth quarter of 2025 was $1 million, compared with $1.5 million in the fourth quarter of 2024. For the full year, adjusted EBITDA was $4.4 million, compared with $9.4 million in 2024. Earnings per share. Basic earnings per ADS in the fourth quarter of 2025 were $0.03, compared with $0.06 in the fourth quarter of 2024.
For the full year, basic earnings per ADS were $0.14 compared with $0.87 in 2024. Balance sheet. As of December 31, 2025, the company’s shareholders’ equity increased to $32.1 million, up from $26.4 million as of December 31, 2024. Our cash equivalent, and debt investments, including accrued interest as of December 31, 2025, was approximately $22.5 million, compared with $25 million at the end of 2024. We ended the year with 0 debt, and our strong balance sheet continues to support our ability to invest strategically while maintaining a focus on sustainable value creation. The outstanding ordinary share count as of today is approximately 72.5 million shares, representing approximately 7.25 million U.S.-listed ADSs. Guidance. Moving on to our outlook for the first quarter of 2026.
Our guidance reflects what we see today based on customer orders, backlog, and current consumption trends, and is given as of today’s date. We currently expect that in the first quarter of 2026, revenue will be around $11 million with a range of ±7%, representing approximately 46% year-over-year growth. Adjusted EBITDA for the first quarter of 2026 is expected to be approximately $1.4 million, with a range of ±$0.5 million. To summarize, 2025 was a year of significant transformation, strong growth, and continued strategic investment. We remain focused on building long-term leadership in the AI data infrastructure market and on generating sustainable value for our shareholders. With that, we’ll now open the call for questions. Operator?
Operator: Thank you. We will now be conducting a question and answer session. If you would like to ask a question, please press star one on your telephone keypad. A confirmation tone will indicate your line is in the question queue. You may press star two if you would like to remove your question from the queue. Again, for participants using speaker equipment, it may be necessary to pick up your handset before pressing the star keys. Our first question will come from Kingsley Crane with Canaccord Genuity.
Kingsley Crane, Analyst, Canaccord Genuity: Hi. Thanks for taking the questions. Shahar, any update on the state of modern website data collection prevention techniques? I know that the goalposts are always moving. How do you feel like you’re keeping up, and what kind of investment do you feel like is needed to keep up specifically on that front? Thanks.
Shachar Daniel, Chief Executive Officer, Alarum Technologies: Yes. First of all, hi, thanks for joining. Yes, as I mentioned, just very shortly, just a few minutes ago. As data become really the oxygen of the AI and the LLMs products, and we need to scale and everybody needs data at scale, it become more and more challenging to collect data which we see today and maybe forever as an opportunity and not a threat because it basically you know it become a barrier for small players to come in. Of course, the results for this is positive.
Yes, websites, not all of them, yes, but some of them are okay with the scraping because it’s bringing traffic and ranking, but it becomes challenging from you know products that are trying to prevent those players from coming in, sometimes false alarms because those products are looking for cyber attacks and not on scraping because some of them are not so sensitive for scraping. Our target basically with our product is to come and say, "Hello, we are collecting only publicly available data. We are not a threat.
We are not here for a cyber attack." If we cross this, let’s say, shaking hands in the beginning of the website as we are doing today, we can come in and collect available public data. Sorry.
Kingsley Crane, Analyst, Canaccord Genuity: Okay. Yep, that’s helpful. Just a couple more. I mean, you’ve highlighted last quarter and this quarter the benefits of the long-term infrastructure partnerships with enterprise customers. You know, revenue has declined, this quarter, you know, guiding to a decline in Q1. Can you just help us get a better sense of that improving visibility that you have in the business as a result of these longer term partnerships? I mean, do you see clear seasonalities that are emerging? Do you think that like the back half could be stronger just based on the timing of when these models, could need to be trained or fine-tuned? Thanks.
Shachar Daniel, Chief Executive Officer, Alarum Technologies: Okay. Kingsley, something that is very important for me to say, I think I said it over the last quarters, and I want to mention it again. If you take a look on 2025, you will see that suddenly, let’s say it like this, suddenly in the middle of the year, we jumped. You know, over the time, we jumped slowly, a digit after digit, meaning we were in $6 million for quarter, then we jumped to the class of 7, then to the class of 8. In the middle of the year, we jumped 3 numbers, yes, from quarter two, from the second quarter to the third quarter, from around 8 to 11, and then 10, and then 11 again.
Meaning this exponential growth took us to a totally different place, and it comes significantly, this growth from this great achievements of us to penetrate with this large LLM that are consuming data in huge scales. The only thing is that if you will measure it quarter after quarter, you will still see that the numbers are not sustainable because they are working on data loss. Meaning for example, they are now in the stage in the development and now the demand is for data or for video data for this kind. They have a huge demand. They are getting most of their, we are the suppliers, and we provided most of the demand.
They slow down a little bit and then come the next stage, and then the next stage. These workloads, because it’s not something in production that is sustainable, also, of course, has an impact on our revenues. Then you can see a quarter of $11.8, $11, then you can see $13, then you can see $12. But this, as we see it, is by nature. We don’t see it as a decline. We see this is the nature of the market now. We are going with the market. The opportunity is here, is endless and unbelievable. Nothing more that I can explain because it’s not a decline that comes from something, let’s say, something behind that we can explain and elaborate, and there is a trend here or trend there.
The trend is positive, but the movements can be, you know, sometimes a movement of two weeks can have an impact on the full quarter, but not on our strategy and tactics and over time. Make sense?
Kingsley Crane, Analyst, Canaccord Genuity: No, it makes sense. Like the trend is broadly positive, particularly on the, you know, the annual trend. Just trying to get a better sense of that seasonality, and it’s hard to predict. Last one for me, just, you know, by our math, nearly half of the revenue in Q4 came from customers that you didn’t have one year ago, just based on backing into that number from NRR. I’m just curious how you think that might trend over the course of the next year and just the potential to sign new customers, anything you can share with other engagements that are in the works.
Shachar Daniel, Chief Executive Officer, Alarum Technologies: Okay. I will divide my answer into two parts. First of all, just to put some light on this part of the NRR. Of course, as we just mentioned all the time, this year was, you know, a really it changed our lives. Suddenly very fast, a new vertical came in and become the first place of most of all our other use cases and verticals. For this, even though you see that the NRR is less than one, you see that the company is growing because those old use cases, let’s say, slowed down or dropped, but the new one came, and it’s much bigger than the others.
You know, it looks like that the AI is not a trend, that the data demand for AI is not a trend. It’s here to stay, it looks like forever. From this aspect, we think that this change, okay, took the NRR down, the new customers cohort went up significantly and will bear fruit from it. For your second question, yes, we are negotiating all the time. In the last two quarters, we have bids, negotiations, proposals, starting small with mid-level LLMs, the biggest LLMs of the world. We are, you know, we are one of the key players that they have. As I mentioned in the past, most of these customers are not as a strategy, yes, as a policy, they are not using one vendor.
Sometimes it’s two, sometimes it’s three vendors. Most of the times we will come in, sometimes we will get more demand, and we will get more data. Sometimes we’ll be the first option from their supplier. Sometimes we’ll be the third option, but it can change all the time. Yes, dramatically, we are investigating new opportunities, new engagements, new big customers, mid-level customers, and most of them are around AI and data collection for training models.
Kingsley Crane, Analyst, Canaccord Genuity: Thank you. Good to speak. Thank you.
Shachar Daniel, Chief Executive Officer, Alarum Technologies: Thank you very much. Appreciate it.
Operator: Our next question comes from Brian Kinstlinger with Alliance Global Partners.
Kevin, Analyst, Alliance Global Partners: Hi. Thanks for taking our questions.
Shachar Daniel, Chief Executive Officer, Alarum Technologies: Hi.
Kevin, Analyst, Alliance Global Partners: This is Kevin with Brian.
Shachar Daniel, Chief Executive Officer, Alarum Technologies: Hi, Brian.
Kevin, Analyst, Alliance Global Partners: You spoke last call about bringing in more infrastructure in-house, and optimizing the network to expand the gross margin. Can you explain a little deeper any progress on these initiatives over the last quarter? When should we expect to see those start to materially impact margins?
Shachar Daniel, Chief Executive Officer, Alarum Technologies: Just to make sure, I understand your question. You said that we mentioned last time and even before that we are investing in order to scale our infrastructure to support the huge scale and the huge demand that is coming especially from the AI use case. Yes, as I mentioned, most of our. You know, we are investing in both sides. First is to expand our network, in order to bring more endpoints, to improve our performance, and the infrastructure, which is costly.
From the other side, as it becomes a game now of huge scales, we are working around the clock, which one of our main missions all the time to make our infrastructure to be more efficient and to be in the optimum cost. Meaning, yes, we are not saving money now, but vice versa. Every time we see an opportunity, we are investing and buying more endpoints, more infrastructure, improving our infrastructure. From the other side, we have a team that is working around the clock with all our hosters and, you know, the cloud and the on-prem hosters, the servers and the endpoints suppliers in order to make our product to be more efficient and in the optimum price because we are scaling.
Both of them are working very well and we are progressing, and we would keep this progress at least in this all 2026.
Kevin, Analyst, Alliance Global Partners: Got it. Thanks. Are there any updates on your large anchor customer, the Asian online marketplace? How’s that partnership going, and what have they indicated from a demand perspective?
Shachar Daniel, Chief Executive Officer, Alarum Technologies: Okay. It’s going very well. It’s stabilized. We are talking with these customers. We have more than one that are basically, let’s call them, market leaders. We see that the demand is here to stay. They need this kind of data, and then they will need in a few months or quarters other kind of data. They are training this model, and then they are developing another model, and they need to train him. Then in production, they need to stay up to date, and they need to get refreshed datasets. It’s going well. They are satisfied. We are working with them already for a few quarters. Thank God, it looks okay.
Kevin, Analyst, Alliance Global Partners: Great. Thanks.
Operator: This now concludes.
Shachar Daniel, Chief Executive Officer, Alarum Technologies: Thank you.
Operator: End our question and answer session. I would like to turn the floor back over to Shachar Daniel for closing comments.
Shachar Daniel, Chief Executive Officer, Alarum Technologies: Thank you, and thank you all for your time today. We look forward to hosting you on Alarum Technologies’ next results call. Thank you very much.
Operator: Ladies and gentlemen, thank you for your participation. This does conclude today’s teleconference. You may disconnect your lines, and have a wonderful day.