All right. Good morning, everybody. I think we stand between them and lunch later on, if I'm not mistaken. Glad everybody could join us. My name is Shaul Eyal. I lead TD Cowen cybersecurity research. We are very happy to host Dave Barter, CFO of Cellebrite, Andy Kramer, VP of IR. This is, as you can see, a fireside chat. By all means, any question, just raise your hands. Feel free to keep it as interactive as possible. Gentlemen, thank you so much for joining us. Andy or Dave, maybe for those slightly less familiar with Cellebrite, can you maybe provide us an outline about what is it that Cellebrite does for a living?
First off, thank you so much for having us.
Always a pleasure.
This is a great highlight for us.
Yeah.
Cellebrite is really a company that's focused on public safety. As a vertical software company, we really do pride ourselves on our ability to ultimately serve nations, communities, and quite frankly, all the citizens in terms of being able to really provide and support lawful and ethical investigations. That's kind of the heart of what we do. Our products now represent a full platform. Our ability, both with a very unique hardware-software solution in terms of how you might gather information, whether it might be from a cell phone, whether it could be from a computer, could be something like SD, or a SIM chip, or now drones.
Yeah.
We have the ability to extract information. Our superpower is the ability to do a full file extract. What that really means is we can ultimately go through over 500 applications, over 50 cloud services, and be in a position where we can garner a tremendous amount of intelligence. We have an evidence management platform that allows customers to maintain full chain of custody. It has an AI viewer that allows not just an examiner to interact with the data, but also allow a detective or potentially even a district attorney, and t hat platform is growing at a very high rate. We call out over roughly 100% for the last six or seven quarters. Adoption of cloud is a big deal within public safety, particularly the governments are probably laggard adoptions in terms of that type of technology and delivery model.
Then we have a variety of products around intelligence. We can ultimately help you take 200 or 300 phones and bring them together. We ultimately have a GenAI product that we recently launched, and we also now have a purpose-built solution for investigators and detectives that incorporates AI. It has full workflow, task management, but it really accelerates lawful investigations. Really kind of working end-to-end, we're pretty excited about the platform we've put together.
That's great. You just two weeks ago reported a solid set of results, Q1 2024 results with affirmation of your 2026 targets. Maybe Dave, can you double-click on some of these items? Also, how do you see 2026 potentially unfolding from that perspective?
That's a great question. I think what we were excited about from a customer perspective, first and foremost, we kind of highlighted it because it was really nicely captured in our EMEA theater, where we saw growth accelerate about 10 points year-over-year from mid-teens to mid-20s. What that really reflects when you think about national-level public safety, is a series of programs. In the U.S. we're a little bit unique, but overseas, if you think customs and borders as an example, it's a national program, but it usually goes into the defense side of the equation for many countries. We saw that defense side of the business within Europe. We also saw within Asia-Pacific as well, continue to accelerate as nations become much more protective of their borders, much more protective of their citizens, particularly given all the geopolitical tensions.
I'd say commercially speaking, we've always had a lot of strength in state and local government, and that continues to be a really great part of our business. At a national level, having national programs, whether it's around safety, whether it's around organized crime, in some cases it's around narcotics, all of those are actually becoming much more important to the company. The other piece that we alluded to on the call is we're starting to get to that point with a broader array of products. Platform agreements are starting to become a part of how we engage with states. Having a broader product roadmap, we'll have nearly 12 releases of product this year.
All of that is ultimately stimulating that relationship where it wasn't uncommon for a customer previously to only buy one or two, but now you're getting to a situation where we're having a much more holistic conversation.
Yeah.
Yeah. I think just to build on that, ironically a year ago, compare where we are now versus a year ago, new leadership. Tom's taken the permanent CEO role, Dave's joined us as CFO, t he product portfolio is much more robust. We're fulfilling the vision we had when we talked about an end-to-end platform and being able to deliver a solution for storing not only Cellebrite data, but all evidentiary artifacts, being able to analyze all of that rich and robust information. I think we're at a really interesting inflection point. I think the other dynamic that's starting to change is the level of spend from the U.S. federal government.
Yep. Well, we'll get to that. Thank you for that, both. Back in late February, clearly Anthropic released Claude's security code. We've seen a couple of other frontier AI-related models. Sending shivers on the one hand, we've seen nice correction actually over the course of the past say, four, five, six weeks or so. Curious as to how you see AI as a tailwind, as a potential headwind to your platform.
As you also think about your product evolvement longer term.
Sure. I think we'd be happy to share some perspective. We think of AI as a tailwind and a force multiplier and a net positive. I think it might be helpful because you published some research pretty recently.
Yep
On Mythos and what Anthropic is trying to do and with that particular capability, maybe share your perspective on how it impacts the security community more broadly and from a Cellebrite perspective, we can share how we think about it.
Absolutely. I'll put it as a note. It's a good point, happy you brought it up. Just over the weekend, the 90-days update came out. I think the major takeaways are, on the one hand, 10X velocity with the ability to spot new threats. Couple of companies, couple of partners that participated, say like CrowdStrike, for example, flagged many concerns that they were able to spot, given a 10X velocity of scanning capabilities.
I think it also brings up the question, whether we're seeing the big platform providers as becoming more of beneficiaries. There's going to be pretty much, much more partnerships along these lines. It's not going to be a zero-sum game from that perspective. Yet, there's going to be couple of categories and subcategories that are slightly more, call it at risk, given Claude's capabilities, say application testing.
Which is a really kind of a smaller sub-sector. We don't have a publicly traded company in Palo Alto Networks. They do have a small asset that they acquired three, four years ago, but it's tiny. There's several private companies in the space. They could be seeing potential headwind. We don't have any standalone, call it public entity in that respect. Again, the takeaways, partnerships, platform providers, and probably more of a tailwind rather than a headwind as initially perceived back in late February into early March.
Okay. No, that's great.
Could that be how you guys be observing it from where you sit?
Yeah, I think we are just a pro-AI company in many respects. Pro-AI meaning similar to an Anthropic, we employ researchers. Similar to Anthropic, we have inference engineering. We are, I'd say, on the balls of our feet in the sense that we have a very unique ecosystem of people that represent our human capital.
in the form of what do we do around research or what do we do around the engineering. Our general viewpoint is our innovation, particularly around our product, and I brought it up earlier, we have a hardware appliance, we have a cloud. We have a very unique approach. From that standpoint, we look at AI as a net positive to our business in the sense of it certainly is changing our innovation cycles in terms of how we build product. Certainly changing how we run our company. It's certainly leading to new products that we're bringing to market. I think we remain positively encouraged. Again, we believe based on our calls with all the mega caps, we feel pretty comfortable with the path that we're on.
We're humble because we still view it as early days around AI. We, again, love the feedback we're getting on our agentic Project Genesis from customers. From that standpoint, I think we're probably in awe of the power of AI. I think we're excited about it, and we are running about as fast as we can. We take nothing for granted.
Sure. Maybe on that topic, on Genesis-
Yeah
Andy or Dave. Not even two months, about two months, give or take, that you launched it. Not GA yet.
Exactly.
Kind of letting some of your customers play with it, let us know. It would appear as if they're quickly becoming addictive, given some of the commentary made on the recent conference call.
Yep.
Maybe you want to share some thoughts and views.
Sure
on Genesis.
Andy, do you want to hit a couple highlights?
Yeah. Maybe I'll sort of set the table, and then Dave can build off of it. I think that when we launched Genesis, whenever we launch a new product, we typically sign up a cadre of customers, could be a half dozen to a dozen customers, to really help us, we call them design partners, test that solution, and their feedback will help guide future enhancements and what we need to get that product into general availability. We were overwhelmed with the interest levels that came from our customer base. Instead of six to 12 customers and users, we had over 500 users representing several hundred agencies globally. These are national agencies, these are regional, these are local police departments.
The feedback came in very quickly and was incredibly compelling relative to accelerating the speed at which these detectives, investigators, prosecutors could get to the signal from the noise in terms of all of the data that they were trying to analyze. Taking weeks of investigative, manual, painstaking, and tedious work and compressing that into 15 minutes in some cases, half hour, hours. You're talking orders of magnitude of productivity, and ultimately that is enabling them to get to prosecute and/or exonerate much faster than they would have otherwise been able to do. I'll pause there and maybe Dave, if you want to build on that.
I think you really hit on it. I think the pieces that really came to me were customers saying, "Gosh, I would have spent two weeks and incurred massive overtime in order to do something that now is a 15-minute exercise." We had another customer and this kind of got to our people, "Well, how will they pay for it?" One customer observed, "If I go to court, it will cost me $15,000-$25,000." They look at this and say, "Gosh, a DA would love to have access to this product." Perhaps that special analyst or a special investigator that works for the DA to be able to get to a point where they can say, "Gosh, I really have a great foundation of evidence. This is a case that naturally should be settled." Let's say save all of our taxpayers a lot of money.
The last part that really I'd say hit home for me was there was a case involving a younger person and the evidence really stacked up against this person. Through Genesis, they actually exposed a lot of interesting and unique insights and quickly actually came down to it became an exoneration as opposed to one of those situations where the prosecution geared up. In the world that I think we're very warrant-based, we're very lawful in our approach, very ethical in our approach to AI, I think we're very encouraged around how this is actually changing the game for law enforcement.
If you think about maybe just to build off of that further, this is a pretty meaningful opportunity for us. We'll be a first mover in this space. If you think about the traditional buying centers and users of Cellebrite technology, they've been largely technical in the digital forensic units and labs of various national defense intelligence, law enforcement, state, local police departments. This gives detectives, analysts, prosecutors, a very powerful tool to get to actionable intelligence, much faster than they could do through a manual review. Compressing the cycle times, we believe that it opens up meaningful TAM to us that previously we have been limited in our ability to tap into.
Maybe relating to that on your recent conference call, Tom, the CEO, made several bullish comments about investigative AI and that potential to expand Cellebrite's TAM opportunity. Do you want to go back and double-click on that? I don't know if we want to add even numbers around it.
I think the approach that we took on the call and the approach we took to sizing was really geared around the idea that within law enforcement, there's heuristic. For example, a customer might say, "Hey, we budget $1,000 for a gun." A handgun, that's where that sits in the line item. Maybe even for a taser, it's $10,000. Where we started to get to where with our customers, they said, "We can imagine a quarter of a detective ultimately starts to amount to AI spend." That was our approach that said, you may start off at $1,000 per detective per month, or you may start even at a lower level of spend commitment. The tipping point of, hey, this is a force multiplier, we think actually gets to be about a quarter of a detective.
That was actually our approach to sizing it, saying, okay, you have a heuristic in mind for every part of law enforcement. Let's lean in on that and let's figure out, okay, if that's where we're skating towards where you're actually thinking about it as a trade-off between human capital investment versus human capital productivity, that's an interesting way for us to size it since so much of your spend is already sized that way. That was a kind of a leading indicator for us.
Got it.
How we thought about the potential of this market. Our view right now is, quite frankly, very centered on adoption and using it, exactly the way Andy described it. There's not a frontier model that seems to be really scratching the itch for law enforcement. They seem to need something that looks more akin to a vertical language model. That's where we put all of our research, our training, and our engineering effort is really to meet that need. Our focus is let's drive adoption, and then we'll probably towards the end of June, we'll start driving monetization among some of the early adopters who have been using it for several months.
As you start to think about that, you described and I think our language on our call is this can be addictive for detectives. They won't go backward after they've tried a product like this because their productivity is elevated to a very different level. If you're a counter-terrorism professional, a counter-terrorism unit, and you're a detective there, and you're able to spend a couple of hours and untangle a financing network for a terrorist organization that would've taken several people, several weeks and or months worth of time, that's a game changer. I don't think that opportunity with that particular customer, there are 10 people in that unit.
They want to spend money, they want to buy that product, but more importantly, there are several hundred detectives across various other crime types who that initial foot in the door will hopefully open wider into those other opportunities.
Let's maybe touch on Inseyets.
Yeah.
All right, just for a second, maybe you can outline the Inseyets and where are we in the migration process?
Yeah
of Inseyets.
Yeah.
That fair? Yeah.
It's a great question. Maybe just for those who haven't been tracking, we ultimately, when we launched the ability to unlock and have an extraction process, we had a variety of tiers. What we've been running, I think last year we finished about 55% of our customer base had moved over to Inseyets. What Inseyets represented was actually a very simple offer. It was at a fundamental level, do you want a premium UFDR experience? The UFDR is the extraction that comes out of the phone. If you want a premium experience, this is probably for you. If you want better manageability and be able to manage your fleet and be able to manage all these upgrades that come through Android or iOS or other platforms, you want better manageability, this is probably an offer for you.
Finally, if you want more automation in the platform, you ultimately want to migrate. In the end, we've had something in there for everybody in terms of, do you want a premium experience? Do you want better manageability? Do you want automation? Ultimately for that benefit, there's certainly a modicum of cost, but there is a lot of inherent customer value in three different ways. Our view is, we've been taking down 20, 25% of the customer base in any given year. We think we'll certainly cross 80% this year, roughly, and maybe do a little bit better, but I'd say the value seems to be working well for the customer base.
One of the frequent investors question I know we get, which sometimes could be observed as a potential risk, is that ongoing ability to unlock, for example.
Yep
the iOS landscape. What we've been telling investors from that perspective, if history is any guide, I know that past performance is not indicative of future performance, but it's still, if history is any guide, it'll always take a few months, and it will be sorted out.
Right.
By the way, I think we have seen when thinking about the iOS 17.
Yep.
Is that the right way to think about it? Could that longer term change, it would appear as if Apple has been trying to put more guardrails.
Like-
More obstacles from that perspective yet.
I think when you look at it macro-wise, you'd say, gosh, the leading competitors generate about $1 billion of revenue today, maybe a little bit more, and the OEM manufacturers generate hundreds of billions of dollars, yet they have great confidence in terms of, I'd say, the engineering they do, but also the engineering that we do, and the value of the service that we provide. I'd say Andy and I have a lot of conviction, as does Tom, in terms of our approach around R&D.
I think, having worked at Microsoft earlier in my career, it's one of those, I never doubt the power of one of the bulge bracket or big brands, but at the same time, I guess I have a lot of confidence in what we do and the role that we play. I think it is one where we approach the world with a lot of humility around the fact that we are competing against companies that are much bigger than we are, and yet, for a number of years, we've always managed to deliver that service that is so important to public safety. My general view is, the will will find a way in the form of this is incredibly important forensic information. Law enforcement can't work without it. The unlock is a key piece.
The ability to forensically extract all of that information might even be more important because, as you know, we're going after over 500 applications, over 50 cloud services, some of the most encrypted information. Even after the phone is unlocked, we still have to get all that information. That's the power maybe of our research team is not just doing a, I'll call it the magic trick of unlocking a phone, but really to be able to garner all that insight. I'd say we feel really good about our technical chops, but I would never want to say that, I would never want you to think that we're arrogant because we do compete against larger companies.
The level of investment that we're making today is, I think, meaningfully and demonstrably higher than where we were maybe 24 months ago. We now own an asset, Corellium, that specializes in vulnerability research. We work with multiple third-party specialist firms who build their IP around stringing together vulnerabilities and building exploits. They augment the teams that we have that are globally dispersed, and that are expert in the trade craft. I think that as Dave said, we don't want to be presumptive around our ability to continue doing it, but we've built a good track record and just the same as a portfolio manager has a track record for outperforming the marketplace. We have our own track record for continuing to deliver solutions that impact our customers.
As Dave mentioned, what's interesting is that even when a phone goes into a lab unlocked, it won't stay that way. The majority of the time it doesn't stay that way is because there's a victim, there's a witness, there's even a suspect who will readily provide their consent to examine that device, and they'll provide the password. In the minority of cases where there is no compliance, then technology like Cellebrite is impactful. The same exploits that we build to get access to the device are necessary and impactful in extracting the data and decoding that data.
Andy, I know you mentioned at the early part of the conversation FedRAMP. It has been a key focus for investors last year, without a doubt, into 2026. Can you guys update us on the progress achieved so far and maybe in that context, your May 6 announcement?
You're on a roll.
All right. Well, I'll start. Yeah, we were thrilled to receive the formal authorization to operate for our FedRAMP compliant cloud solutions, and it opens up a new opportunity within the U.S. federal government. The Department of Justice was our sponsor, and it was a multi-year, two-plus year, multimillion-dollar investment on Cellebrite's part. If you think about the opportunities to grow with U.S. federal, now they are multifold in that we can continue to support their needs for digital forensics. That next step, and think about the types of agencies that we work with, civilian, Department of War, Department of Defense. Those are our agencies, particularly on the civilian side, that interoperate and collaborate with state and local agencies.
Not only is the FedRAMP authorization significant and opens up new spend within the U.S. federal government, but that collaboration will undoubtedly create a halo effect where agencies around the world, not only in the U.S., will look to the U.S. federal government and say, "If it's good enough for the U.S. federal government, why aren't we looking at a solution like this?" I think that we've tried to set realistic expectations that we do expect to have some monetization to occur this year. Timing wise, we'll see whether that happens sooner rather than later in terms of first half versus second half. That having that authorization to operate now allows us to also identify opportunities that attach to the government's fiscal 2027, which starts on October 1st.
I think that we're very optimistic that there will be some initial signs of success this year, and that you'll start to see that scale in 2027 and beyond. I don't know, Dave if there's anything else to offer there.
Probably the only additive piece is this is a wave of digital transformation that we've seen play out before. Anyone who's kind of gone through the hyperscaler journey and then started to see how it would impact some of the laggards, whether it was financial services or even healthcare. I think what we see with the U.S. federal government is they're going through their own digital transformation around all the data associated with Just, again, it was no surprise that the DOJ was the sponsor. Criminal justice at a national level and then all the way down to state and local, has a tremendous opportunity to unlock value associated with chain of custody, but how they collaborate across all of these crimes, whether they're at a national level or a state level or a local level.
The ability to be able to go from legacy on-prem storage solutions to a contemporary cloud solution is pretty exciting.
Questions from the audience before we proceed.
One back here.
Oh. Hi, hugh. Sorry.
You talked about your strategy of expanding from where you were at the beginning, extracting from phones, and it looks like you try to get more sticky by doing stuff. Previously, I heard at your presentations, officers talking about Cellebriting phones.
Yep
In terms of as you expand from just even Cellebriting phones, what's your strategy to be more sticky in that process?
It's funny. I'm not actually aiming to be more sticky as much as I'm aiming to go where my customer needs me to go. I think the drones represented a great case study for us where they said, "Gosh, when you look at the types of data and the data artifacts we need to ingest, drones are becoming just increasingly more common." I think in Andy's backyard, there was a situation with a drone that went into Fenway Park. The reality is drones are actually now one of the most persistent threats within the prison system around how contraband ultimately gets into a prison. My general view is as we extend the platform, we're really extending it where our customers need us to play, and that's opening up opportunities.
That really, I guess early on when I came in, I had the benefit of actually reading a report where we engaged with one of the state-level agencies, and they shared their viewpoint on their IT infrastructure and their footprint. It was a variety of siloed applications that was used for both operational and judicial data. That, for me, it really opened up my eyes to the opportunity of saying, "Gosh, I actually think this has a lot of the digital transformation aspects where you can say you have siloed applications.
You really do need to have the connective tissue between what we do as it goes from operational all the way into judicial, where we can ultimately, again, almost like what we're doing with Genesis, can I take weeks and months and compress them into minutes and hours, and really fundamentally just bend the curve in terms of how they're working. I don't think I need to work hard at being sticky. I'm already sticky. The question is how can I really unlock it as the leading software provider to this segment of the economy?
Thank you.
It's a great question. Please.
Is the barrier to this cross-functional work really more cultural in terms of between departments? I worked on some, you know, the 9/11, right, the challenge of the fire department and police department.
Yep
Not talking to each other or being on different systems. Do you see that more as a cultural barrier because the technology is less-
The technology is only in a barrier in the sense that people don't have yet the solution that actually meets the needs. There was a situation where somebody in one of the states said, "What we really would like to do is when a parolee actually comes in for their report, a parole officer would like to be able to run an extraction, use your platform, cloud-enabled, to find out have they actually been compliant with the terms of their parole?" It's a generational shift in technology where they haven't actually thought about using the cloud ultimately to automate those types of workloads. There was a member of Congress who recently said, "Gosh, internet crimes against children.
All of the judicial data is in siloed systems. That you're through ATO, we actually could have a national system that actually is focused on internet crime against children, or different attacks like that. The cloud-enabled layer, I think, has been missing. We have actually the connective tissue to work either at a national level or at a state level to bring the isolated populations of on-prem data together to affect a judicial outcome. I don't think it's as much cultural as it is they just didn't have actually the approved tooling in place.
Do you think the more advanced use cases will reverse that as opposed to some of the challenges they've had in the past?
Certainly, if I take Genesis as an example, our first agentic AI product, where we've had customers from over 15 countries say, "A sovereign data cloud is no longer all that important to me. If you can accelerate judicial outcomes with a cloud solution, I'm interested." I think that's why I think it has those characteristics that make a digital transformation almost a good analog, where this part of the economy has been on-prem for such a long time. Now that they're seeing either AI or cloud-delivered solutions, I think they're seeing lawful investigation shift in a pretty quick way.
I think our acquisition of SCG for drone forensics is also instructive for how we need to get better tools, better technology, more actionable information faster out into the field. Whether you're a parole officer, whether you're working the border and passport control, whether you're a homicide investigator, and there was a nightclub shooting and you've picked up 40 phones, the better we can help these different user groups determine the viability of, okay, I've got 40 phones, only six of them are relevant. If I can return the other 34 quickly and focus on those six, I'll save time, I'll save energy, I'll save budget, and get to my information faster. That is relevant to a homicide investigation. It matters to an officer or a soldier in the field if they're looking at a drone forensic.
Portability, faster access to information, and more intuitive usability are all driving our development activity.
Dave , Andy, thank you so much.
Thank you so much for having us.
It was a pleasure. Thank you.
This was wonderful.
Appreciate it. Thanks a bunch.
Thank you, everybody.