Logo of jester cap with thought bubble.

Image source: The Motley Fool.

Date

Thursday, Feb. 12, 2026 at 12:00 a.m. ET

Call participants

  • Chief Executive Officer — Thomas E. Hogan
  • Chief Financial Officer — David Barter
  • Chief Revenue Officer — Marcus Jewell
  • Vice President of Investor Relations — Andrew Kramer

Takeaways

  • Annual recurring revenue (ARR) -- $481 million, up 21%, including $16.1 million from the December Keryllium acquisition; organic ARR grew 17%.
  • ARR growth by geography -- Americas grew 19% to 53% of total ARR, EMEA increased 24% to 35% of ARR, and Asia Pacific rose 23% to 12% of ARR.
  • SaaS/cloud ARR -- Surpassed 50% growth and reached 22% of total ARR, reflecting significant adoption of cloud offerings.
  • Guardian Forensics -- Achieved its sixth consecutive quarter of more than 100% year-over-year ARR growth.
  • Insights conversion -- Fifty-five percent of the digital forensics installed base converted to Insights, exceeding the 50% target.
  • Q4 revenue -- $128.8 million, up 18%, including approximately $1 million from Keryllium.
  • Full-year revenue -- $475.7 million, an increase of 19%.
  • Gross margin -- Fourth quarter margin at 86%; full-year margin at 85%.
  • Adjusted EBITDA -- $127.6 million for the full year (26.8% margin); Q4 adjusted EBITDA was $38.3 million, with margin expanding to 29.8%.
  • Free cash flow (FCF) -- $160 million for the year (34% margin), up 30% from $124 million in 2024 (31% margin).
  • Cash position -- $535 million at year-end, up $52 million after a $147 million net outflow for the Keryllium acquisition.
  • Keryllium acquisition impact -- Immediate one-point compression on margins, with impact expected to wane by year-end as revenues scale.
  • SCG Canada acquisition -- Pending closure by Q1-end; transaction size $15 million to $20 million for a business with low single-digit millions in ARR.
  • 2026 initial guidance -- Expected ARR growth of 18%-19% to $567 million-$573 million; revenue of $565 million-$571 million (19%-20% growth); adjusted EBITDA of $149 million-$155 million (26%-27% margin); FCF margin above 30%.
  • Guidance philosophy -- 2026 guidance range narrowed (e.g., ARR spread reduced from $15 million to $6 million) based on renewals, pipeline, and robust forecast accuracy.
  • U.S. federal segment outlook -- Flat in 2025, expected to exceed overall growth in 2026 with anticipated pent-up demand, funding increases, and imminent FedRAMP Level 4 authorization for Guardian.
  • Go-to-market expansion -- Sales executive headcount increased by 20%, with new investments in enablement and training, particularly targeting defense and intelligence (D&I), and the enterprise segment.
  • Platform leadership -- Significant R&D and partnership investments to reinforce leadership in both Android and iOS forensics capabilities, with related product launches slated for the next six weeks.
  • AI initiatives -- Parallel efforts to embed AI-driven productivity into products and to develop agentic applications for specialized investigative use cases, with monetization opportunities expected in 2026 but not included in current guidance.
  • Rule of x metric -- The company now focuses on ARR growth plus FCF margin as its “rule of x,” targeting a range in the upper forties and aiming for 50% or higher.

Need a quote from a Motley Fool analyst? Email [email protected]

Risks

  • CFO Barter said, "Our profitability also reflects two transitory headwinds that weigh on margins. The first item reflects the absorption of incremental Keryllium costs we have added following the acquisition. We expect this impact will dissipate by the end of this year as top line expands. The other factor is foreign exchange, most notably the continued strengthening of the shekel against the U.S. dollar."
  • Management noted approximately a one-point margin compression from Keryllium, and more than a one-point FX headwind, particularly from the strong shekel, above initial expectations.

Summary

Cellebrite (CLBT 0.31%) delivered 21% ARR growth with $481 million in total ARR, maintaining strong free cash flow and gross margins despite integrating Keryllium and navigating currency headwinds. Management emphasized robust SaaS and cloud adoption, highlighted by Guardian Forensics’ sustained triple-digit ARR growth and a majority (55%) Insights conversion rate. The upcoming closure of the SCG Canada deal will extend the company’s reach into the emerging drone forensics segment, while a sharpened guidance philosophy narrows ARR and revenue ranges for 2026, reflecting improved forecasting. U.S. federal segment demand is expected to accelerate with an anticipated FedRAMP Level 4 authorization, and product innovation in forensics access and agentic AI applications is positioned to further elevate competitive positioning. Strategic talent expansion and focused platform leadership investments round out the company’s multi-year growth drivers, while management remains attentive to transitory margin headwinds from acquisitions and foreign exchange.

  • Cellebrite is preparing to launch new mobile unlock capabilities for both Android and iOS, expected to boost related revenues starting in Q2 2026.
  • The company expects Guardian Collaborate and Guardian Forensics are well positioned to sustain their 100% plus year-over-year growth rates.
    • SCG Canada’s drone forensics ARR is currently in the low single-digit millions.
    • Management stated the total addressable market for digital investigations and drone analytics is about five times larger than Cellebrite’s historical core TAM.
  • Ratable revenue recognition and evolving product packaging are expected by management to reinforce ARR growth durability over time.

Industry glossary

  • FedRAMP Level 4 (ATO): U.S. federal government cloud security certification allowing the storage and processing of Controlled Unclassified Information at higher sensitivity, unlocking large-scale deployments in the federal sector.
  • Agentic applications: Specialized AI-powered software modules designed to independently pursue investigative tasks, such as analyzing cold cases or detecting cybercrimes, within Cellebrite's portfolio.
  • CFID: Cellebrite’s drone forensics extraction device, enabling detailed data collection from drone platforms, analogous to its UFED mobile device technology.
  • RPO: Remaining performance obligation, referring to contracted revenue yet to be recognized, used in assessing future sales pipeline coverage.

Full Conference Call Transcript

Andrew Kramer: Thank you very much, operator. And welcome, everybody, to Cellebrite DI Ltd.'s Fourth Quarter and Full Year 2025 Financial Results Conference Call. I am joined today in Israel by our primary speakers, Thomas E. Hogan, Cellebrite DI Ltd.'s CEO, and David Barter, Cellebrite DI Ltd.'s CFO. Marcus Jewell, our CRO, is also participating. This call is being recorded, and a replay of the recording will be made available on our website shortly after the call. We will also add a transcript. Please note a copy of today’s press release and financial statements, including GAAP to non-GAAP reconciliations, is available on the Investor Relations website at investors.cellebrite.com.

In addition to the press release, we post a separate investor presentation that provides an overview of the business and our recent financial performance. I would also like to remind everybody that the slide in your webcast viewer is a placeholder only. There are no actual slides to accompany our prepared remarks. We also publish supplemental historical financial information for each quarter of 2025 and 2024 along with full year 2023 and 2022 on our Investor Relations website.

Additionally, unless stated otherwise, our discussion of our fourth quarter and year-end 2025 financials as well as the financial metrics provided in our outlook will be done on a non-GAAP basis only, and all historical comparisons are with the comparable periods of 2024. I would like to remind you that today’s discussion will contain forward-looking statements including, but not limited to, the company’s business operations and financial performance. All forward-looking statements are subject to risks and uncertainties and other factors that could cause matters expressed or implied by those forward-looking statements not to occur. They could also cause the actual results to differ materially from the historical results and/or from forecasts.

Some of these forward-looking statements are discussed under the heading Risk Factors and elsewhere in the company’s annual report on Form 20-F filed with the SEC on 03/18/2025. The company does not undertake to update any forward-looking statements to reflect future events or circumstances. I will now turn the call over to Thomas E. Hogan.

Thomas E. Hogan: Thanks, Andy. I will just jump right in. We closed 2025 with a solid fourth quarter that capped a year marked by meaningful strategic progress. We cemented our Insights offering as the gold standard in digital forensics, drove strong adoption of our SaaS and cloud-based offerings, extended our integrated AI functionality, completed our first material acquisition, and added important talent across the company. We grew ARR by 21% in 2025, which factored the combination of a four-point headwind from our U.S. federal unit's actual performance versus our original plan, and a nearly four-point tailwind associated with the close of Keryllium. Overall, our ARR growth reflects expansion across all of our major geographies and our flagship offerings.

We outperformed relative to guidance on both our fourth quarter revenue and our adjusted EBITDA. Our growth and ongoing spend discipline delivered strong free cash flow of $160 million in 2025, and a 34% free cash flow margin. I would like to quickly share some of our fourth quarter highlights and accomplishments that position us for accelerated growth in 2026. First, we have now converted 55% of our installed digital forensics base to Insights, exceeding our 50% target and reinforcing our market-leading capabilities. Second, we doubled down on our mobile research to ensure our unlock capabilities continue to keep pace with the major phone manufacturers.

We believe these investments extend our leadership in Android and will reassert our leadership in iOS. We expect this range of leadership capabilities will hit the market over the coming six weeks and position Cellebrite DI Ltd. as both the leader across each major segment as well as the clear leader from a comprehensive cross-platform perspective. Third, SaaS and cloud adoption remains outstanding. ARR for these offerings grew north of 50% and now represents 22% of total ARR. Guardian’s impressive trajectory continued with its now sixth straight quarter of 100% plus year-on-year growth. Guardian Forensics is rapidly becoming the industry’s de facto repository for evidence that matters and where chain of custody is critical.

Fourth, we completed our acquisition of Keryllium in early December while we continued our work to gain final clearance from CFIUS. Keryllium’s ARM virtualization technology remains an industry-unique and powerful asset. Customer interest across both defense and intelligence and the private sector continues to exceed expectations. We remain confident this asset will be highly accretive to our growth and will exceed our pro forma expectations when we announced the transaction in June. Looking ahead to 2026, we start the year well positioned to reaccelerate growth with initial guidance of 18% to 19% as compared with our organic growth of 17% in 2025.

We see several levers for further acceleration as our portfolio and solutions evolve over the coming quarters, but we chose to take a prudent approach to our guidance until these assets become generally available and we can confirm expected market adoption. Let me recap the primary contributors to our expected reacceleration. First, core demand for our solutions remains strong. Macro tailwinds around crime, population growth, the use of digital in both the pursuit and resolution of crime continue to climb as validated in our recently released industry survey. And the constraints associated with human capital persist. Unfortunately, these known established macros have been exacerbated the past year by increased geopolitical tensions around the world.

Second, the well-chronicled disruptions in the U.S. federal segment are thankfully now behind us. We expect the roughly flat growth performance of this unit in 2025 to reaccelerate and to exceed the company’s overall growth rate in 2026. There are multiple drivers that will contribute to this resurgence in growth: pent-up demand in core unit growth, increase in focused federal funding, and the final DOJ-sponsored authorization to operate for FedRAMP Level 4, which we expect to obtain before the end of this quarter after a lengthy 18-plus month process. Federal ATO will pave the way for Guardian and our cloud assets in the U.S. federal market.

Augmenting these growth engines is our increased focus on more targeted defense and intelligence solutions, as well as the product fit of Keryllium in D&I. Given current mid-quarter visibility, we are optimistic this unit will get off to a fast start in the first quarter. Third, we have elevated the quantity and quality of our go-to-market organization with a roughly 20% expansion in sales executives and our increased investments in enablement and training. Fourth, within digital forensics, there are several important levers in terms of Insights conversions; the value proposition of our Insights upgrade cycle is now well understood and many agencies have incorporated their upgrades in their planning and budgeting cycles for 2026.

These dynamics position us well to drive conversions this year by an additional 25 plus percent. Just as important, based on our anticipated platform leadership, we expect accelerated growth in the unlock business as we enter the second quarter and the remainder of 2026. Fifth, we took an important and exciting step today with the agreement to purchase SCG Canada. The deployment of drones globally is not just growing; it is exploding. The drone market is expected to grow 20 plus percent annually and surpass $53 billion by 2026. Its constructive use cases are broad, ranging from surveillance and commerce, to the safety of local law enforcement and national defense. Unfortunately, drones also enable nefarious use cases.

The U.S. alone reported over 1.2 million drone violations in 2025. We believe drone forensics will rapidly become one of the most significant data sources for making our nation’s communities and businesses safer. Given our leadership in digital forensics, and our customers’ trust and dependence on our digital insights, adding drone forensic leadership was both a logical and, candidly, necessary strategic decision. This is a capability that will bring immediate value to defense, intelligence, and law enforcement agencies as well as to the private sector that is charged with securing airspace around critical infrastructure, prisons, and dense locations such as airports and sports venues.

This represents a modest but important move to address an emerging need and further elevate the impact of our AI-powered platform for multi–data source analysis. We expect to close this transaction by the end of the first quarter and we will share additional details upon closing. Sixth, with the recent closure of Keryllium, we enter this year transitioned from a reseller to a fully integrated selling motion. We are driving elevated education and training across the Cellebrite DI Ltd. go-to-market team and customer base and see meaningful growth opportunity across both the public and private sectors. Keryllium will also clearly exceed the company’s overall growth rates.

We are excited and optimistic about our progress in emerging leadership in digital investigations and analytics. These strategic assets grew 2.5 times faster than the overall business in 2025. Guardian Collaborate and Guardian Forensics are well positioned to sustain their 100 plus percent year-over-year growth rates. In addition to the important ATO for the U.S. federal market, we expect to obtain similar certifications in Australia, New Zealand, and select European nations later this year. We will also launch Guardian Investigate this spring.

This product is squarely focused on enabling criminal investigators, detectives, analysts, and prosecutors to build stronger case narratives, collaborate seamlessly in a secure workspace, leverage a diverse set of data sources and file types, including traditional smartphones, but also adding important sources such as call detail records, open source intelligence, video, RMS, ballistics, and license plate data, and ultimately leveraging the most powerful AI-enabled analytics in the industry to navigate and interrogate this mountain of important evidence. Feedback from beta and customer design partners has been exceptional, and we think is a harbinger for accelerated deployment and growth in 2026.

Pathfinder, our flagship analytics solution for multi-phone forensics, continues to grow and deliver important levels of insight and productivity to a growing percentage of our Insights installed base. And last, but certainly not least, is our progress in the thoughtful and ethical use of GenAI. We have been pioneers in the use of machine learning and AI for the past decade and plan to extend that leadership in 2026. While many view AI as a threat to software, we view AI as an absolute tailwind across three fronts. The first is applying it across our internal organization to drive productivity and efficiency, and this initiative is well underway.

Second, AI enables significant improvements to the productivity of users of the Cellebrite DI Ltd. portfolio, which ultimately elevates our value proposition and customer retention. And third, we see meaningful opportunity to monetize unique and focused agentic applications that bring rich capabilities that transcend a range of use cases from child exploitation and missing children to cybercrime to stagnant cold cases and major criminal investigations. I want to briefly expound on our constructive view on AI and why we view it as a force multiplier from both a business and a societal impact.

Cellebrite DI Ltd.’s mobile extractions are at the epicenter of the most valuable, complex, and difficult-to-obtain sources of evidence that are relevant to virtually every investigation, and the corresponding power and capabilities of any AI engine. Said more simply, our unique intimacy with the most complex evidential artifacts gives us a unique advantage in harnessing AI for good. That intimacy is then compounded by our domain expertise with investigative workflows leveraged by hundreds of man-years of law enforcement experience. And finally, Cellebrite DI Ltd.’s history is grounded in the quality, ethics, compliance, and security that has earned us the trust of thousands of the largest and most sophisticated public safety and government agencies around the world.

GenAI can and will be a powerful force for good, but the stakes involved in crime and sovereign defense demand that advanced analytics are complemented by full traceability, ethical use, and human verification. To conclude, I am proud of our progress in 2025. We navigated turbulence in the U.S. federal space while still delivering healthy growth in both the top and bottom line. Just as important, we made critical investments throughout 2025 that span organic innovation, strategic partnerships, and targeted acquisitions. Leadership and innovation matter, and we continue to invest in the long-term growth and leadership of this company.

We enter 2026 with a truly differentiated end-to-end AI-powered platform that delivers high-value insights and intelligence from an expanding range of data sources. We are already hard at work on where and how we can expand our value for 2027 and beyond. We have a bold aspiration to not just solve crime with efficiency, but to ultimately drive a material reduction in crime itself. We are proud of our impact in the world, and we are anxious for the future. With that, I will turn the call over to David Barter. He will do a click down on the details and add further insight to our first quarter and full-year guide. Dave?

David Barter: Thanks, Tom. I would like to briefly share highlights from the fourth quarter and full year. ARR grew 21% to $481 million, which includes Keryllium. We closed the acquisition December 1; Keryllium’s ARR was $16.1 million. Excluding this, our ARR grew 17% year-over-year, and sequentially ARR increased 6% over Q3. Perhaps even more noteworthy, after experiencing headwinds in the first three quarters, our net new ARR growth in Q4 was back to prior-year levels. This aligns with the remarks and the confidence we shared on our last earnings call that growth would reaccelerate in FY 2026. Geographically, the Americas represented 53% of total ARR, while EMEA represented 35%, and Asia Pacific represented 12%.

In terms of growth rates by geography, the Americas grew 19% with our U.S. state and local government and Latin America teams leading the way. EMEA grew 24%, and Asia Pacific increased 23%. Higher-growth solutions like Pathfinder, Guardian, and now Keryllium have become a larger percentage of our ARR mix. At the end of 2025, these solutions represented 14% of total ARR, and we anticipate that this mix will continue to shift closer to 20% by the end of the coming year. Turning to revenue, in Q4, revenue grew 18% to $128.8 million, which includes approximately $1 million from the Keryllium acquisition. For the full year, revenue grew 19% to $475.7 million.

Our software solutions drove approximately 90% of our fourth quarter and full year total revenue. Our fourth quarter gross profit increased to $110.8 million, which represents a gross margin of 86%. Our full-year gross margin was 85%. Fourth quarter adjusted EBITDA of $38.3 million increased 33% over the prior year, and the margin expanded by 340 basis points to 29.8%. For the full year, we generated adjusted EBITDA of $127.6 million, or 26.8% on a margin basis. We achieved this level of profitability despite a strong FX headwind as the shekel strengthened materially against the U.S. dollar.

As Tom noted, we have continued to balance the investments required to drive innovation and fuel expansion with our focus on giving our teams the AI-enabled tools to elevate productivity and efficiency. We ended 2025 with 1,285 employees, up 10% over 2024. Turning to the balance sheet, we ended 2025 with $535 million in cash, cash equivalents, and investments, up $52 million despite the outflow of $147 million in net cash used to acquire Peryllium in December. Free cash flow for the fourth quarter was $82.3 million. For the full year, free cash flow was $160 million, or 34% on a margin basis. This represents 30% growth over 2024 free cash flow of $124 million, or a 31% margin.

As a reminder, we remain very focused on reaccelerating ARR growth while maintaining a free cash flow margin of at least 30%. As a vertical software company, we believe we will benefit from AI. We are of the view that strong ARR growth combined with a strong free cash flow margin strikes the right balance and enables us to serve all stakeholders. Let us shift gears and take a look at our 2026 expectations. Before I review our guidance, I wanted to share a few thoughts around our guidance philosophy in response to investor questions on this topic.

We were very deliberate about not changing Cellebrite DI Ltd.’s guidance framework when I joined the company midway through 2025; we have modified our guidance philosophy for 2026. In particular, we focused on setting prudent ARR and revenue expectations around tighter ranges that are corroborated by our renewals, deal pipeline, and applicable RPO coverage. Accordingly, we use tighter ranges for our quarterly and annual ARR and revenue targets. As we execute over the coming quarters, we will reassess and revise those top line targets as appropriate, and the same is true for adjusted EBITDA. Our initial view into 2026 ARR calls for a reacceleration in our growth rate versus the 17% organic expansion we delivered in 2025.

I would like to quickly revisit the framework from November on the 2026 drivers. First, winning new logos and increasing price or mix on existing offerings is expected to generate several percentage points of growth. Second, Insights—through conversions, more pervasive deployments, and ups on unlocks—is anticipated to support growth in a meaningful way. Our third growth driver involves Guardian and Pathfinder, the cornerstones of our digital investigation and analytics offerings. We expect this will contribute mid–single digit percentage points to our ARR growth. Keryllium, our fourth driver, continues to experience healthy customer interest and demand. While it is still early days, we expect a contribution of at least a couple of percentage points to growth.

And finally, we expect to improve gross retention. In terms of our planned acquisition of SCG Canada, we have not yet incorporated any contribution into our outlook since the deal has not yet closed. It is worth noting that while SCG is currently a small business, it will bring innovative technology that we believe is highly complementary to our platform and will benefit greatly from our global distribution. Looking at the first quarter, we expect ARR growth in the range of $491 million to $493 million, or 20% to 21% growth.

The combination of our recent Q4 ARR and our anticipated Q1 ARR demonstrate not only sequential stability, but an expansion motion in terms of absolute dollars versus the comparable quarters one year ago. We expect first quarter revenue in the range of $127 million to $129 million, an increase of 18% to 20%, and adjusted EBITDA in the range of $26 million to $28 million with a margin of 21% to 22%.

For full fiscal year 2026, we expect ARR in the range of $567 million to $573 million, or 18% to 19% growth; revenue in the range of $565 million to $571 million, or growth in the range of 19% to 20%; and adjusted EBITDA in the range of $149 million to $155 million with a margin of 26% to 27%. As Tom noted, we are in the early stages of evolving our products and packaging in ways that are intended to ultimately make it easier for customers to expand the range of solutions they subscribe to over a multiyear period as they take advantage of our cloud and AI-enabled offerings.

We anticipate this will serve as a stronger foundation for durable ARR growth. We also expect that a byproduct of this transition will be more ratable revenue recognition over time. As a result, we continue to view ARR as the most relevant top line KPI. As you consider our outlook for profitability, I would like to highlight a few elements. We anticipate, in line with prior fiscal years, approximately 60% of our adjusted EBITDA dollars will be generated during the second half of the year and will be accompanied by stronger adjusted EBITDA margins. Our profitability also reflects two transitory headwinds that weigh on margins.

The first item reflects the absorption of incremental Keryllium costs we have added following the acquisition. We expect this impact will dissipate by the end of this year as top line expands. The other factor is foreign exchange, most notably the continued strengthening of the shekel against the U.S. dollar. We continue to thoughtfully manage our overall cost structure while also taking pragmatic steps to limit the impact of FX volatility. In terms of free cash flow, we are expecting 2026 to be another strong year with anticipated FCF margins in excess of 30%. And finally, I would like to offer a thought on Cellebrite DI Ltd.’s rule-of-x performance.

Historically, we have calculated our rule of x by adding our ARR growth rate and our adjusted EBITDA margin. As we have scaled our business and matured our execution, we have delivered adjusted EBITDA margins at levels well above the original floor of 20%. Accordingly, we now view 25% adjusted EBITDA as our new floor on profitability, which also correlates at a high level with a free cash flow margin of at least 30%. Since more ratable revenue will impact both top line and bottom line rates of expansion, we will be using ARR growth and FCF margin to measure our rule of x. We feel this will provide investors with more clarity and insight.

Building on Tom’s comments around rule of x, we begin the year with an outlook in the upper 40s and an objective to drive performance to 50 plus. Overall, the team delivered a successful 2025 despite the transitory headwinds in the U.S. federal market. We are moving into 2026 with optimism around our prospects to further reaccelerate ARR growth while delivering attractive profitability and free cash flow. We look forward to sharing our progress in 2026 with you as we execute on our plans over the coming quarters. Operator, that concludes our prepared remarks.

Operator: We will now open the call for questions. The floor is now open for questions. At this time, if you have a question or comment, please press star 1. If at any point your question is answered, you may remove yourself from the queue by pressing star 2. Again, we ask that you pick up your handset when posing your questions to provide optimal sound quality. Thank you. Our first question is coming from Bhavin S. Shah with Deutsche Bank. Your line is open. Please go ahead.

Bhavin S. Shah: Congrats on a solid year and a strong 2026 guide. Maybe first on the acquisitions. You kind of announced SCG Canada, expecting to close, kind of two deals in quick succession. Are you guys thinking about ensuring that you can execute against the strategy for both of these deals along with maintaining a focus on the core? Do you feel like you have to make any internal changes as you fold these companies in, and how do you guys think about allocating resources amongst the core relative to Keryllium and SCG?

Thomas E. Hogan: This is Tom. I will take it. So first, the Keryllium transaction took longer to close than we anticipated. The good news with that is we have now had essentially seven plus months since we announced the deal to get into a rhythm and a cadence. We inked the reseller deal quickly after announcement given some of the delays, and so from a training and go-to-market perspective, there were limitations, obviously, where you cannot fully operate as one entity pre-close. But while it may feel concurrent given that it closed early December and we expect to close SCG by the end of this quarter, realistically, from an executional challenge perspective, there was pretty good spacing between the two.

Then the second thing—and I will probably anticipate the question that somebody is going to ask—is, how big is the breadbasket with the SCG deal? We are super excited about it because the drone world, as I said, is exploding, and having market-leading drone forensic capability is hugely compelling. We think the growth trajectory of that business is also going to be—I will just say, and I hate to use exaggerated terminology, but “breathtaking” is an apt description. But for our stakeholders, it is currently a small operation. They are the market leader in our view, but we are talking about a business that is low single-digit millions.

The complexity of that business—and by the way, this is what we do. Instead of using our devices on UFEDs to extract forensic data from phones, we are now using their CFID to extract forensics from a drone. This could not be more core and complementary to who we are and what we do. So the dive difficulty here, relative to a standard acquisition, is low.

Bhavin S. Shah: That is super helpful there. And just a quick follow-up. On the last point you are talking about, the drone opportunity—understanding it is still very nascent here—but how did this come about in terms of looking at the asset? Was this something that customers are asking for? Or is this something that, as you look two to three years, five years down the pipe, this is going to be more meaningful? What is your remit and how we can help, but are coming from as well?

Thomas E. Hogan: Yes. So good news is the answer is both. In particular, in the short run—and Marcus might comment on this—but in the defense and intelligence world, they are already using the technology. As they look to their plans and procurement, and the need to expand given the proliferation of drones, the demand from our mutual customers is loud and clear. And then as we do strategic planning, one of the things you would expect from us is to always look out—kind of skate to where the puck is going—from a TAM perspective, and say, are there adjacent markets that bring big TAM to the Cellebrite DI Ltd. value proposition?

The moves we are making in the whole investigative world in analytics now combined with the moves we are starting to make in drones, the TAM of those two added markets is actually about 5x the core TAM that we have been chasing for the last 19 years. So both from a strategic planning and customer demand standpoint, it was one of the most obvious strategic moves I have seen in my 43 years.

Marcus Jewell: I will add to Tom’s comment there as well. Yes, there is customer demand. We are frankly taking data from sensors. The biggest sensor out there is a cell phone, but a drone deployed in borders and those areas is one of the other sensors which is definitely required. So yes, there is already demand, and there is already training for those solutions.

Operator: Thank you. Our next question comes from Jonathan Ho with William Blair. Please go ahead.

Jonathan Ho: Hi, good morning, and let me echo my congratulations as well. I wanted to start out with maybe a little bit more color on the investments that you made to extend your mobile forensics leadership that we will see later on this quarter. Could you help us understand what this could mean from either improving net retention, win rates, or product expansion perspective?

Thomas E. Hogan: Jonathan, the investments were basically doubling down with our internal research team to both extend our leadership in Android and to also ensure that—we love for people to just standardize on us for the front-end unlock and access from a platform perspective, but we do not want them to feel like they are making any sacrifice. The goal is to have a leadership offering in both of the major OSs, Android and iOS. The investments were made in our internal research team to make that happen, in conjunction with several external partnerships which, by the way, is not new.

It is sort of a standard operating procedure for us for the last 15 plus years to leverage the combination of our internal badged researchers with some of the best and brightest researchers in the world to help complement our capabilities. We have doubled down on one of those partnerships and added another one that helps bring new attack vectors for exploits and vulnerabilities. When you roll those partnerships together with the innovation, that is why we feel so optimistic about us having a very clear leadership position across the board.

If that comes to fruition, given the importance of access, it will clearly drive accelerated growth on the front end of our unlock access and our Insights penetration and market share and year-to-year growth.

Jonathan Ho: Got it. And then in terms of your comments around the U.S. federal government spending environment, where are you seeing maybe the most pent-up demand? Where are you seeing improvement in terms of the malaise that we saw last year? And what gives you the confidence that this can return to a stronger growth rate?

Thomas E. Hogan: I covered the macro categories, but I will let Marcus, because he is really dialed in and close to this, give you maybe a better answer. I think you will hear there is empirical data and meat behind our enthusiasm. Marcus, why do you not take a shot at that question?

Marcus Jewell: It is as we have said over the last six months. What we are seeing is—reminding everybody of the use cases which are used—there are the defense and intelligence use cases, which, as the world continues to get stranger and more threatened, in forward deployment areas where data collection in the new kind of war situation is incredibly important. We are seeing those use cases build out and a lot of confidence build around those, not only on a local level, but also on a global national level. You saw that strength in our EMEA results as well. Border security continues to be, across the world, a big area, and a lot of money is going into that.

There are also some external things. There is a FIFA World Cup coming to the U.S. That is going to be potential for a lot of serious crime, and that has to be prevented as much as possible, and we are used in the deployment of those areas. So that is the confidence that we see. The agencies, remember, under the big, beautiful bill, are giving a two-year budget. So they are able to get ahead now and start thinking and plan more strategically. Our competitiveness with our product means that we feel comfortable in those positions. The same story of those hardening use cases with more stability from that.

And then the final point, as Tom mentioned in opening remarks, will be the ATO—authority to operate—for our Guardian solution, which means we are unique in a position, the only people that can actually store and share forensic data under the FedRAMP approval, and that opens up multiple petabytes of opportunity for us to capitalize on once we get through the ATO process in the next few weeks.

Operator: Thank you. Our next question comes from Shaul Eyal with TD Cowen. Your line is open. Please go ahead.

Shaul Eyal: Thank you. Good afternoon, everybody. Congrats on solid 2025 completion. Tom, maybe just for clarification—I've been getting some emails from investors. On that small drone tuck-in acquisition, the scope, the low single-digit millions, I think you have indicated, is that the price paid or is that potential ARR contribution, I do not know, like, first half or maybe even first quarter once you close it? And maybe any headcount number you can provide us with as it relates to this acquisition. And I have a follow-up.

Thomas E. Hogan: Okay. Good question. Let me be clear and helpful. We are inheriting a low single-digit ARR run rate. It is a small-scale, although market-leading, solution. Price paid—we have not closed yet, so there are a couple of conditions that we are chasing down here in the next week or two—will be in the $15 million to $20 million range for the company. And then we would be disappointed in the midterm if the ARR growth potential for that business is not well north of that $15 million to $20 million. We are hitting this at the right time, which is part of why we moved fast here.

We think asserting ourselves as a first mover at scale to address drone forensics is going to pay big dividends for the company.

Shaul Eyal: No question about it, makes whole sense. Maybe with respect to the model, how should we be thinking about second half versus first half linearity? Should it mostly resemble 2025 trends, or should there be any little deviations as we think about second half versus first half?

David Barter: Great question. Thank you for asking. I would actually model the top line pretty close to 2024 in terms of that split, which was largely a little less than 40% in the first half and 60% in the second. That kind of maps to what I think Tom highlighted in his remarks around the number of offers that are coming to market.

Shaul Eyal: Thank you so much. Good luck.

Operator: Thank you. Our next question comes from Jeffrey Van Rhee with Craig-Hallum. Your line is open. Please go ahead.

Jeffrey Van Rhee: Great. Thanks for taking the questions. Congrats on that free cash flow margin in particular. Tom, on the AI side, spend a second and talk a bit more about these opportunities. Specifically, you talked about agentic applications—cybercrime, child exploitation, etc. Just talk a bit more about exactly what those would be and how you monetize them, if you would.

Thomas E. Hogan: Yes. We are parallel-tracking our efforts in AI. There is a lot of work being done by the core product and technology team to integrate advanced AI capabilities that drive productivity—things like media classification, summarizing text chats, report generation, identifying conflicts in testimony—just a list of things that turbocharge our current stack and deliver a lot of productivity to our users. We are going to continue to elevate and identify more opportunities to do that real time.

In parallel, we have an AI innovation center that has been missioned with developing more specific agentic applications and use cases that, based on feedback from design partners and some early evaluators, are getting—we are pleasantly shocked at people’s enthusiasm to begin to deploy these agentic apps. I gave you some examples. I will not name the departments, but in talking to the chief of the investigative unit in a large U.S. police department, his response was, “So wait a minute.

I can deploy this app and just turn the engine loose on a stack of missing-children cold cases and see what this application can surface that might provide added insights that we had overlooked and then merit or make it worth the time to circle back and pursue the investigation.” In terms of your question, Jeff, about monetizing, the candid answer is we are working on that in real time, as we speak, to figure out what we bundle as an enhancement and a value add to the product’s core capability—which we are doing with Guardian Investigate—and then what are the things that you can monetize, and then what are the price points.

One of the messages I would give to the shareholder base is, candidly, the guidance that we have in place today assumes that we do not monetize any of that in 2026. Based on the feedback we are getting, I actually think there is going to be an opportunity in fiscal and calendar 2026 to start to monetize some of these agentic applications.

Jeffrey Van Rhee: Great. Fantastic. A couple maybe for Marcus. Marcus, it looks like—Tom called it out in the script—adding 20% more capacity in terms of sales heads. Talk about where you are allocating those heads. How are you thinking about the assignment and where you are pointing those folks? And then in particular, if you would comment on D&I—obviously, with SCG and Keryllium, you seem to have a deeper toehold and TAM opportunity in D&I. Does that require a different motion, different touch points? And is that an area of particular focus?

Marcus Jewell: Indeed. Let me break that down. In public safety, which is the largest part of the business, there is still scope to add heads. We feel that our coverage, even in SLG in the U.S., can increase, and so we have increased penetration especially into the medium and small agencies across the U.S. That is one area where we put a well-trodden motion, and we repeat and extend on that. The next area, which is probably getting missed, is through the private play. We have Viper, which is a pen testing solution, which is proving very interesting and successful with the enterprise. So we have increased our coverage in the enterprise business for our mobile app penetration testing.

Financial services is one of the largest customers we have. A lot of you guys on the banks there run multiple apps, and they need to be checked that they are safe and cannot be penetrated, and that has been a real uptick for us. So we have increased our coverage on the specialization there. Then the third area is D&I, and we are delighted that we brought back one of our leading salespeople who left us for a little break and has come back to head up a global development there.

He is an special forces person and is driving a new global standard for us around the D&I space, and we have added resources both in Europe and the U.S. to cover that.

Jeffrey Van Rhee: Maybe just one last for me then, Marcus. On the FedRAMP, is there a backlog sitting there that have been kind of waiting for this? And should you see a surge from that? And then maybe to put a finer point on it, I think the prior call—you or Tom, I believe—had mentioned a very large deal. I believe you had said the largest deal that you had ever seen that was looking like it was lined up for a 1Q26 close. If I have that right, can you give any update on that?

Marcus Jewell: Let me tackle the second part of the question first. We have multiple threads of large annual spend in the first and second half of this year. We feel confident about our position trending there, but I would like you to think about that—it is not one single customer. It is actually three or four different programs which are all in seven figures for us. We continue to track well, and we feel good about our technical evaluation position with multiple threads there. The second point—could you actually ask the first question again? Sorry, I lost my thread there.

Jeffrey Van Rhee: I think we were talking about the federal side. I was most interested in the big deal that you had talked about.

Marcus Jewell: The big deal continues to track well. We feel good about that. These big deals tend to scale up and grow. Procurement can be complicated, but we feel good. But I want you to think there is not one—there are multiple agencies both in state and local and in federal which have multiple big deal opportunities for us. We continue to track well against the one we highlighted. We are actually adding fuel to the fire on that. And then the first part—ATO—has come back to me now. When you say backlog, I have to be careful because backlog would suggest that we have a purchase order. What we do have is we have seen our largest bid go out.

We have scoped an initial contract for a large federal agency in their storage requirements, which exceeds nine petabytes. That would equate to a very large deal that would be certainly the single biggest transaction we would have ever done. We continue to track well. We want to get through the ATO process, which we are confident on, and then we will talk about turning that into real revenue as quickly as we possibly can.

Operator: Thank you. Our next question is coming from Mike Cikos with Needham. Your line is open. Please go ahead.

Mike Cikos: Great. Thanks for taking the questions here, guys, and I will echo the congratulations on the strong finish to the year. I wanted to come back to the Insights conversions and the unlocks for a second. Can you help us think about how the adoption of unlocks trended in Q4? And Dave, I know you went through a number of different vectors that give you guys confidence in that ARR reacceleration in the coming year. What are your assumptions as far as how Insights and unlocks play into that? I just want to make sure I was clear on that. And then I have a follow-up.

David Barter: Mike, let me break them apart between the Insights and the unlock because I think they will have different dimensions. We ended with Insights at about 55%. Overall, we feel good. I think that will follow our broader deal seasonality, just from the standpoint that the transitions will probably occur with deal contract expirations. So it will probably follow the linearity or the seasonality I described earlier around how things unfolded over the course of fiscal year 2024. I think the unlocks will be a little bit different just from the standpoint that, as Tom alluded to, there is work and new technology that will be hitting the market.

That really will start to kick in late Q1, but really Q2, and then I think it builds and ramps from there. Some of it will be linked to expiring agreements, some of it will be expansions of existing deals, and that is a little why we called out in his remarks we want to see how the market reacts to give you a little bit more color around how it unfolds. I have a thesis, but I need a little bit more data, and that was kind of incorporated in the guidance.

Mike Cikos: Thank you for that. And then for the follow-up here, just wanted to get a better understanding. I am happy that we have Keryllium in the rearview as far as the closing in December, and the expectations for a couple of percentage points growth in the coming year. Can you help us think about the cost base that Cellebrite DI Ltd. is now carrying for that asset and how that impacts the calendar 2026 guide? I know you cited specifically Keryllium as well as the headwind from an expense perspective tied to FX, but I just wanted to make sure I was thinking through that properly.

David Barter: Let me tether with some numbers. It is about a point of compression on margins due to Keryllium, and as we shared in the remarks, we will grow into it over the course of the year. It is a little bit heavier in the first couple of quarters; we will get some leverage in the second half. As we go into 2027, we will be cooking well. We picked up a wonderful technical team, and they are really excited with what they are already doing. FX is more than a point that is ultimately burdening the P&L.

We ended the year with a pretty healthy hedge, but nonetheless, we are taking on a healthy point of compression just due to the strength of the shekel—about two points of extra headwind than what we expected. I think it ends up being a temporary headwind in the sense that it actually passes through the P&L this year; we come out the other side with better profitability. Again, I think cash flow margins continue to remain at a compelling place. That help, Mike?

Operator: Thank you. Our next question comes from Brian Lee Essex with JPMorgan. Your line is open. Please go ahead.

Brian Lee Essex: Great. Thank you for taking the question, and congrats on the results. Great to see the stabilization, particularly in the Fed. Maybe, Dave, if I could have you unpack a little bit of the commentary you had on the guidance, particularly with regard to the prudence of guidance and the tighter ranges around ARR for your targets. Could you help us understand the philosophy that the company had over the past few years and how that has changed—how it is tighter? Where are you narrowing the guardrails? Is it around upside to give yourself more cushion? Help us understand the level of conservatism in the setup as we head into fiscal 2026.

David Barter: It is a great question. Thank you. If you were to rewind the clock and compare where we are today versus where we were, I think we had a $15 million spread on ARR last year. Right now, we are calling a $6 million spread. What you have probably seen is a pretty healthy degree of forecast accuracy. The business has a strong customer base motion, and it is complemented by a smaller new logo motion just given how the initial lands happen. We really used our customer base motion to inform it. We looked at the pattern of our expirations, looked at the upsell—this is where we went from the expirations to the expansion at the time of renewal.

That is what we started to model in. Then we built in some of those points that Tom talked about, but where we had a little bit less visibility, that is where we snapped the chalk line. That is how we ultimately got to a $6 million spread—looking at the 18% to 19%—and then we pulled it through the P&L. So maybe the variability was previously included in the guide, and we are pulling that back a little bit. Correct. In general, I look out and go for a high degree of visibility. As I get a little bit smarter, that is what I will update.

Brian Lee Essex: Super. And then maybe just to follow up on your comment around gross retention, the comment that you expect to improve gross retention—where have you seen the points of improvements, and how does that translate into what you have for expectations into fiscal 2026?

David Barter: Great question. I start to see it almost beginning with our flagship for our platform in the form of Insights and unlock. There are signs that is going to get better. Quite frankly, with the maturity of where we are at both on Pathfinder and Guardian, we see the signs where we have customers really operating on very current versions of Pathfinder and expanding on those versions, and then certainly the buildout on the Guardian platform. There are a number of compelling features. As Tom alluded to, the customers that are starting to experiment—these are our design customers—using AI naturally tag into both Guardian and Pathfinder. They will get, quite frankly, pronounced impact from the adoption of AI.

The fact that they are uploading more and more data and using it that way gives us increased conviction.

Brian Lee Essex: Got it. Very helpful. Thank you so much.

Operator: Thank you. Our next question comes from Eric Martinuzzi with Lake Street. Please go ahead.

Eric Martinuzzi: Yes, Tom, I wanted to dive a layer deeper on your comments regarding AI. In your conversations with customers, obviously there are tools that they use outside of your C2C platform. Are they pulling you in a direction as far as where they want to see enhancements made in the platform? Is it just, hey, we have got AI within the C2C platform and people are not straying from that?

Thomas E. Hogan: The good news is everybody is tinkering. What the early tinkerers are discovering, especially as they start to see and hear about what we are doing—candidly, they are seeing a huge gap in what we are able to produce in terms of the depth and quality of the AI output, which makes sense for the reasons I shared earlier given our knowledge and access to the complex and critical phone data.

Instead of them saying, “Hey, can you shift course 10 degrees here because this would help me or this is where I am going,” as they are learning and discovering what we are doing, what they are saying is, “I am going to stop, and I am going to adopt the products and the capabilities that you are bringing to market.”

Eric Martinuzzi: Understand. And then as you look at the pipeline—and this is kind of product development pipeline—you talked about forensic investment as a key area. How are you using AI in that forensic investment?

Thomas E. Hogan: If your question is, are we leveraging AI to extend or maintain the forensic capability, the honest answer is that what we are leveraging more right now from a technology perspective is the Keryllium asset. That has been a tool for us for the past five years, and now we own it. We use that to accelerate the identification of vulnerabilities and exploits. If the question is, is there a magic AI engine that can go surface vulnerabilities, maybe that will happen in three years or five years, but we do not see that right now. And just—obviously, all software stocks, and ours in particular, seem to have gotten hit here in the past few weeks.

With that assumption in mind, it is a head-scratcher for me. I am sure it is for you guys as well, but just had to ask the question.

Thomas E. Hogan: No. I think what we would say generally is, given the nuance of what we do, the specific use cases, the work, the complexity of the data on phones, this is not some standard research or task that a lot of these engines are capable of essentially outsourcing and driving. This is why we think AI for Cellebrite DI Ltd.—and probably a handful of other companies in very specific nuanced industries—is actually a tailwind and a force multiplier for us that we are going to harness to both deliver value to our shareholders and deliver more value in speed and insight to our customers.

We scratch our heads on the whole “AI is going to make software go away.” We have kind of the complete opposite view.

Operator: Thank you. This does conclude the Q&A portion of today’s call. I would now like to turn the floor over to Andrew Kramer for additional or closing remarks.

Andrew Kramer: Thank you, operator, and I would like to thank everybody for joining us this morning. If you do have any questions, please feel free to follow up with Investor Relations, and we look forward to speaking with our shareholders and prospective shareholders over the coming days and weeks. Thank you.

Operator: Thank you. This concludes today’s Cellebrite DI Ltd. fourth quarter and full year 2025 financial results conference call. Please disconnect your line at this time, and have a wonderful day.