Anek Capital 的 Orel Levy：Cellebrite 投资逻辑 $CLBT

Today we're talking about Cellebrite with Orel Levy, who has done great work on it. Whether you like compounders, event-driven situations, or whatever else you like, I think there's going to be something interesting here. It's a really interesting company. I don't currently have a position, but I wouldn't be surprised if I have one in the near- or medium-term future.

This podcast is sponsored by TRDA. Look, I already know you're going to like TRDA. Why? Because you're interested in this podcast. You're listening to this podcast. And TRDA is just like this podcast, though unfortunately largely without the handsome host. TRDA is anonymized transcripts of buysiders discussing their stocks that they're following and kind of what they really think will drive the stock, what they really think the upsides are, what they really think the risks are. In fact, a huge way that I prepped for my Sellbreak podcast is there was an incredible discussion on the Tora platform where they were talking about, hey, here's the risk. Here's what I think the event path will play out. Here's what I've heard from experts who I've interviewed when I've talked to them about what they really think about the podcast. So, look, I think if you join TRDA, you're going to get huge value. But even better, you can join as a lurker and you can just go and read all the transcripts. But if you want to join and actually start contributing to the platform too, TRDA will take it one step further and it'll be an anonymized interview. They'll publish it and they'll pay you hundreds of dollars to go on the platform and do anonymize interviews and talk about the stocks you follow with other smart people who follow the stocks and long, short, all that type of stuff. Nothing investing advice anywhere, but it's a really useful platform. And again, if you're listening to this podcast, I already know you're going to love the podcast. Go to trytra.com or see the link in the show notes today to check it out.

With me today, I'm happy to have on for the first time Orel Levy. Orel, how's it going?

Good, good. Thank you for having me.

Thank you for coming on. I'm super excited to talk about this company today. As I was telling you before, this is rare, but it was hitting all my boxes, and I'm seriously considering a position, which is rare. Before we get into the company, a disclaimer: nothing on this podcast is investment advice. I don't have a position now, though I might in the future. There's a full disclaimer at the end of this episode, and you can see our legal disclosure.

The company we're talking about today is Cellebrite. The ticker is CLBT. This has been popular in value and event-driven investor circles on and off for the past couple of years for a bunch of different reasons that I'm sure we'll discuss, including a Japanese holding company that owns almost half of it. I'd love to start with the high level: what is Cellebrite, and why are they so interesting?

I think Cellebrite is one of those businesses where it's very hard to find them at a reasonable valuation. It's such a quality business, and the earnings are so high quality, but I think we'll get to everything.

In a nutshell, what is it actually doing? Historically, they started around 2001. The foundation of the business was as a telecom provider. They were trying to solve the pain point of telecom providers: how can you transfer data from one phone to another? Let's say you're acquiring a new phone and need to transfer all of your contacts. What would be your options? Cellebrite was your first option.

They brought in Yossi, who was founder-ish, after 3 years. He had previously worked in a telecom company, and, by mistake, one of their products was actually bought by a sheriff—I think it was in New York. The sheriff bought the product by mistake, and Cellebrite realized that there was huge potential within public safety. It was a very funny story.

They had 2 businesses: the telecom retail business and the public safety digital forensics software and hardware business. Around 2017, they spun out the retail and telecom business and sold it to one of their peers for—I would say it's not public, the numbers—but around $20–$40 million. Then they just doubled down on public safety.

Going back to around 2007, it was very funny that Sun Corporation, the Japanese company, actually bought them for the telecom business, not for the digital forensics business. Now Cellebrite is the market leader by far in digital forensics. Think about it in simple terms: you're trying to break into a phone. Let's say you get one phone from an incident. One of the most famous use cases with Cellebrite was when someone tried to assassinate Trump. You get a phone, and there's plenty of data.

The most famous narrative people have is that Cellebrite cracks the phone. It's not just that. You crack the phone, which is sometimes the easiest option, but then you have a lot of structured and unstructured data. They were actually using a lot of unstructured data in terms of the GPS: where was this guy, and who was he next to? Then they used the Cellebrite product to analyze the data. The second part is actually Guardian.

The company is growing 20-ish percent, with beautiful margins. The cash conversion is above 100%, and unlike most software companies, they never raised any money, so it's bootstrapped. The SBC is very minimal, so the company's culture is very aligned with shareholders.

That's a great overview. You mentioned phone cracking. When I read about this company, I hear “phone cracking,” and I want to dive into that a little bit. They obviously talk about the Trump incident: the shooter was killed, and they wanted to investigate him. They took the cell phone to a regional office—I believe it was in Butler, Pennsylvania—but that office didn't have the Cellebrite product, so they had to take it to the Pittsburgh field office to analyze it, and that took four hours.

That's a pretty nice selling point: if a terrorism event happens and your field office doesn't have this, you want to have it there. They've got other examples from their investor presentation. But when I hear this company, I think of breaking into phones. You said they do other things, but why are they so unique? Why can this little, tiny, bootstrapped company break into phones? I believe they have 1 major competitor. What's unique about these guys? The first thing I thought when I looked at this was, “Why isn't Palantir offering this or something?”

Personally, I think we need to understand that Cellebrite right now is spending the biggest amount in terms of R&D, of course. We need to remember that digital forensics is a cat-and-mouse game. You always have to crack the phone. The iPhone, iOS, and Samsung are always trying to prevent you from cracking the phone, so you always have to invest a lot in R&D.

But I think it's not just that. It's a lot of data. It's about actually capturing the data during those years. Once you capture the data and have the logos, it's very, very hard to penetrate and acquire those logos. Once you trust Cellebrite and become a user, it's very mission-critical.

The other side of the equation is that whenever we invest in software companies, we're always asking ourselves: what is the TCO, or total cost of ownership, for the customer, and what is the ROI? Right now, we usually try to look at it per case. Usually, a department will spend around $150 per case, and the ROI they're getting is so much higher. It's 10× to 100× higher than what they're spending right now. I think it's very hard to penetrate this kind of formula and acquire this logo and this trust already.

That's true. So, what are you seeing that the market is missing that makes this an alpha opportunity? To build on that, as part of the prep for this, I read a buyside call on Tora, and it was, by the way—

Say again.

I thought it was great. It was a really good one. I really enjoyed the product.

I read it, and I was like, “Oh, dang.” It was really good. But the thing that struck me is that it was 2 bulls, and they were basically saying, “Hey, you’ve got a company here that’s growing revenue 20% year-over-year, with everything you’re talking about—great margins, kind of a moaty product, 1 competitor that they seem to be kind of eating their lunch on, all this sort of stuff.”

Companies that grow 20% year-over-year in this space don’t trade for this multiple, right? Look at Axon. Look at Palantir. And it struck me that they said it. I think that’s what you’re going to say. I just look at that, and I’m like, well, it seems kind of obvious. So, what is the market missing? The bulls, everybody’s just kind of tearing their hair out. Why is it trading at this multiple?

So, right now, full disclosure, it seems like I’m the only bull here before Q2. But we will go into everything.

I have that question as well. Yes.

Let’s first look historically at what the market misses. They came via SPAC, and no one likes SPACs. Of course, the stock sold off from $10 when they came via SPAC to around $4.

So, back then, what did the market miss? The market missed that Cellebrite was doing a transition from perpetual software to a subscription. We all know that there’s a valley in the top line when you recognize revenue. With perpetual software, of course, you recognize everything upfront, and then when you switch the model, everything changes.

First, they missed the revenues by around 40% in 2022, as well as the margins. Back then, the company had never really provided guidance before. For them, providing guidance was kind of a first time. If they provide guidance for the public market, they just need to be close to that guidance, as a private company. But, of course, as we all know, if you miss by 1%, sometimes the market can actually punish you.

Also, some people lumped them into the same bucket as NSO Group, and, of course, we all know the story that Apple was trying to sue them. Beyond that, it was a very misunderstood story. Yossi and Dana were great operators, but it was very hard for them to share the story: Is it a one-trick pony, or are they becoming a platform?

Then, once they started to accelerate revenues again from, I would say, 20% to 25% or 28% ARR, with beautiful margins, the market was like, “Wow, we have a second Axon here.” The market fell in love with the company. Everyone was throwing around the narrative that Axon had bought 5% of the company in the SPAC, so Axon was going to buy them.

It was probably going to be 2025 because Adam—Adam was actually the owner of TWC, the SPAC—and Adam was going to get compensation if the stock reached $30. I think it’s probably around August 2026. Everyone was assuming that they would reach there, and then everything kind of changed. This is what the market is missing.

What changed for them was that they had a Q1 that was kind of soft. Net new ARR was lagging what the market was expecting. Then Cellebrite was also guiding for a soft Q2, and Cellebrite also embedded that H2 would include a federal ramp, and federal—which is around 70% of the business—would grow mid-teens or even 20%-ish, and now 2–3%.

So, you have, I would say, a lack of clarity about the federal business. You have management turnover. The founder just left, and the CFO just left. And now there’s the new narrative that Cellebrite didn’t manage to crack the new version of the iPhone.

We all know that if there’s bad news, everyone assumes some incremental problems. Right now, people are assuming that because they didn’t manage to crack the new version of the iPhone, they’re losing market share to Magnet. They didn’t manage to crack the iPhone because most of their employees were actually on reserve duty. So, there are many, many, many false assumptions, and that actually took the stock from $25 to around $14 right now.

That’s great. So, let me ask the first question. In your opinion, they haven’t cracked the new iPhone, but they can crack the new iPhone? So, that’s a risk that people are worried about that’s kind of been put to bed.

They didn’t crack it yet, but they’re going to crack it. It’s the same cat-and-mouse game that we spoke about. If they’re not going to crack the new iPhone with the highest R&D budget and the best employees, no one is going to do it.

But again, what the market is missing is that it’s not the core business of Cellebrite. People assume that if they can crack the iPhone, the new iOS, or Android, then why use Cellebrite? As we said earlier, it’s not the case at all. So, this is just the first point that the market is missing.

That’s so—let me just quickly divert. I’ve got a lot of questions, and you hit on a lot of the risks. I do want to come back to them, but let me ask one risk that kind of jumped out at me when I was looking at this.

I remember—I can’t remember how long ago—but there was some event, and I think Trump—this was Trump 1.0—leaned on Apple that they needed to help the FBI crack a phone or open a phone or something, and Apple didn’t. One risk that jumps out to me here, which I think is worth addressing, is: What if Apple and Google—because Google does Android and Apple does iPhone—just start working more with law enforcement and say, “Hey, if you need a phone cracked, we will crack it for you”?

What do you think about that risk? I could imagine all sorts of responses—positive, negative, reasons it won’t happen, everything—but I’d love to just pose that base risk to you.

I think in order for me to give you a good answer, we need to first understand what Cellebrite’s product is. I would say their platform, Inseyets, which they’re actually converting to a new platform right now—we’ll speak about it later in the growth algorithm, I assume—is a combination of all of their products that serve the digital forensics unit.

First, you have the cracking. I’m trying to simplify it. They have the iPhone or Android cracking. The second part is that they actually analyze the data for you. We need to remember that a lot of data was actually removed from the phone, such as deleted data or unstructured data. So, even if you have the phone, it’s going to be very hard for you to actually analyze this data.

Let’s assume the worst. Again, we know it’s not part of their DNA. Apple, as we all know, is very proud that its iPhones are secure. But let’s assume the worst. Even if this is the case, all of those agencies will continue to rely on Cellebrite no matter what.

I think a great example of that is border control. All of the phones are always open. It’s mandatory for the phones to be open, and they still use Cellebrite. They are still using Cellebrite even when the phones are already open. So, if they had, let’s say, a more budget solution, they would still go into Cellebrite.

That’s great. Let me ask one more risk that jumps out to me. Cellebrite says, “Hey, there’s this huge demand for our product. We’re continuing to sell, and we’re growing 20% a year.”

A big part is this huge market of security. The pitch is that this is a play on cybercrime, but not just cybercrime as in Russia hacking crypto accounts. This is literally a play on, “Hey, there’s a shooter. He has a cellphone. All the data is there. We need to go get the cellphone and analyze the data,” as you’re saying.

And if you believe that more and more criminals are going to have cellphones, data, and connections, this is a beautiful play on that, right?

That’s true, that’s true. But then I’d also say, hey, why are they growing so quickly? They already have—we mentioned the FBI with the Trump shooting. They already have the FBI. You could say, “Oh, there are thousands of logos out there,” but in a lot of fields, there are 20 big logos and then 1,000 small logos, and those big logos move the needle.

I looked this up before: The New York Police Department—I live on the Upper East Side—employs 40,000 police officers. The Baton Rouge Police Department, where I’ve spent a lot of time—I’m from Louisiana—employs 900. So, you literally need 50 Baton Rouge police departments to make up 1 New York Police Department.

It sounds great to say, “Hey, we’ve got 20 logos, and there are 5,000 out there,” but they already have the FBI. They already have New York. Is there this much growth out there for them?

I think another part of the bear case is that Cellebrite has already captured all of the logos. Then, once they manage to—first, 3 years ago, let’s give some background—they tried to shift all of the legacy customers to the more advanced product called Inseyets, which combines 2 or 3 legacy products into 1.

Every time you switch to Inseyets, there’s an uplift. It depends on the customer, of course, but it varies between 25% and 40%. So now the bears are saying, “Your growth actually accelerated because of the one-time Inseyets conversion. You already captured all of the logos.”

But what we actually spoke with those customers about is that Cellebrite might own all of the logos, but in terms of penetration within the departments, it’s still very, very, very low. We still see a ton of room for runway to at least 20% growth going forward, I would say, for the next 3 to 5 years. Then it will decelerate.

I think another question here is why it wasn’t faster if digital crime is actually growing, according to management, at a 30% pace. We all know that crime, unfortunately, is just on the rise, and there’s so much backlog in those cases. Those digital cases are increasing all the time, and there isn’t enough capacity. So again, the question is: Why is Cellebrite only growing at 20%?

I think historically the answer was actually go-to-market issues. Cellebrite didn’t really have, I would say, a quality go-to-market strategy. That’s why they changed the go-to-market strategy and combined a lot of the departments inside Cellebrite that they bought. You see more executives—one came from Juniper—and now they’re actually starting to execute all over again.

For example, Guardian, which is the second product, is actually used to share data. Right now, they’re replacing a legacy tool. So, if you’re an agent right now, you’re sharing evidence by USB or SanDisk. Sometimes, we’ve even spoken with agents from Canada who are flying all over to the U.S. to gather the evidence and then going back to Canada. It’s crazy how inefficient it is.

Guardian’s penetration, according to our numbers, is around 3% to 5% with agents. We spoke with agents and asked, “Is it mission-critical for you guys? If it’s the only solution, what is the barrier to massive adoption?” I think one barrier for Guardian—which is growing, by the way, 100% and is approaching 6%—is that historically, you could only share data with Cellebrite products. But if you were using Cellebrite and Magnet and trying to share data from Magnet, you couldn’t do it with Guardian.

Now that’s actually been solved, according to our research. Historically, Cellebrite didn’t put too much effort into it, so I think the main risk here, if I’m being even more precise, is execution and the new management: the new CEO, the new CFO, and the new CMO. The entire management team has changed over the last 18 months.

Well, let’s dive into that, because I think that is a risk. Again, I’ll reference the protocol that we talked about, and I’ll reference other conversations I’ve had. You even started to mention it. The bear case here—and there are other bear cases—but the immediate bear case is that when you listen to bulls, they say, “Hey, the CEO—let’s just basically call him the founder, because he joined a year after, but he grew this business—”

The founder left last November-ish, and the executive chairman took over the role as interim CEO.

They just hired a new CFO—we’re still in July, this month.

And so the bear case is kind of, “Hey, you’ve got an interim CEO.” What CEO of a company that’s going to grow 20% forever—what founder leaves when things are just really getting spicy? That’s number 1.

Number 2 is, “Hey, the CFO just left.” They had a soft Q1, and on the Q1 call they were saying, “Hey, DOGE and everything—the federal government is kind of hesitant to deploy.” Not that that affects the longer-term or medium-term outlook, but you combine an interim CEO with maybe a new CEO—they even said, “We’re going to have the CEO search settled by Q2 earnings”—and you bring in a new CFO. You talk about Q1, bring guidance down, and talk about softness due to DOGE.

That’s a recipe for a guidance cut. They’re cutting this year, and maybe they had an investor day in 2024 that was very interesting. Maybe they bring down a longer-term CAGR. So I don’t think that impacts the business or the valuation, but it does create the feeling that everyone thinks they’re going to wake up in August and have a 15% down day hit them in the face if they buy the stock.

100%. So first, let’s break it down. I’ve spoken with him many times, and he knows me personally.

This is the basic founder that we’re talking about right here.

Yeah. I think if he’s going to hear this podcast, I have so much respect for him. I really appreciate him so much. As he told me personally, taking the company from 0 to, let’s say, $400 million to $450 million was so much for him. He had to sacrifice so much of his personal life, and I think he was in the company for more than 15 years.

Once you have to sacrifice so much and realize that you’re not the right person to take the business from $400 million to the $1 billion that they’re guiding toward, I think it actually makes tons of sense. We also need to remember that besides Yossi, there are great operators who are still in Cellebrite. For example, the CPO—the chief product officer—came from SAP, and he knows the interim CEO is going to be permanent, so I’m just going to call him the permanent CEO. He actually met Tom before.

I think right now, if you ask me what Cellebrite was lacking 2 or 3 years ago, I would say probably 2 things. One was scaling a software business from $500 million to $2 billion. Two was probably M&A capabilities.

Tom, as we’re probably going to speak about him, was a CEO for more than a decade at software companies. He was leading Thomson Reuters. He was leading a company called Kony, which was sold to Temenos. He was a managing director at Vista, another private-equity firm. Before that, he was working at SAP and HP as well. So he understands a lot about how to execute and scale a business.

I think what the bears were probably saying in the transcript was that, unfortunately, he was sick. But according to our conversation with Tom, he has fully recovered. The way we see it, Cellebrite is probably going to get acquired in a year or 2, and Tom is going to be the permanent CEO for the next 2 years, with the new CFO.

Again, this is full speculation, but we actually think the new CFO, who came from New Relic and sold the business afterward, is a perfect combination. Again, speculation here: the new CFO didn’t sell Model N to Vista, but he was the CFO of Model N, and Vista acquired Model N 2 years after he left. So they probably knew each other before.

I think the person who is actually putting the pieces together here is Adam from True Wind. He owned probably $10 million to $12 million of Cellebrite in his personal account, and he also owns 5% of Cellebrite through the SPAC. He also owns 20% of the Japanese corporation Sun Corporation. I think he’s actually going to push Cellebrite toward a transaction in a year or so.

Then the question is, who is going to acquire them? Again, speculation: there are 3 candidates here. It’s either going to be Vista—they’ll try to copy or compete with Thoma Bravo—or it’s going to be, I would say, Motorola, IBM, or Axon.

When we spoke with Axon many times, it felt like Axon was trying to go more into the drones and hardware space, not applications, because Evidence.com is actually crushing it. Again, saying that Axon is competing with Cellebrite is another misunderstanding of the market.

Cellebrite is actually the perfect fit for Motorola because Motorola is trying to increase its recurring revenue, accelerate its top line, and buy more applications with higher margins. Usually, they acquire companies between $1 billion and around $4 billion or $5 billion. So, if you put, I don’t know, $30 on Cellebrite, it probably fits the higher end of Motorola’s range. But again, 100% speculation—100%.

No, look, you hit all the things. The thing I thought was interesting—and the reason I said up front, “Hey, I’m really interested in the stock”—is that, A, you pitched it and everything, but B, I read the earnings call again. As soon as I read one line, I was like, “Oh, no. This is catnip for me.”

You had a CFO who was a one-time CFO of an Israeli company, and that company did really well. He retires, and you hire a CFO whose last job was selling New Relic. You bring in a Thoma Bravo partner who becomes the executive chairman and then looks like he’s going to become the CEO. By the way, True Wind has about a year left until the last 1.5 million earnout lockups expire.

You roll it all up and you’re like, “Oh, this is a company that’s getting packaged to be sold.” And, by the way, I hear you. Axon invested 5% into the SPAC, and Axon trades at—I'm looking at it—let’s call it 100 times EBITDA. You’ve got this business here that’s growing 20%, should have synergies with Axon, and is a new product line that they could probably push through a lot of their platform. It’s trading for, I don’t know, 25 times, even while growing very quickly.

You roll it all together and you’re like, “Oh, I’ve never seen something better designed to be sold.” So, let me just—

I think it’s also—sorry to interrupt—I think it’s a classic private-equity playbook because Cellebrite can grow 20%. There are so many tailwinds. It’s the market leader, and it can generate so much cash.

I think right now the contribution margins are around 25% to 35%. Again, speculation, but one of our friends actually told us the numbers for Magnet, and they’re generating much higher margins than Cellebrite. It makes sense because Cellebrite is trying to invest so much. By the way, they just did layoffs.

There’s speculation that they just did layoffs before Q2 in order to show, “Okay, we’re going to take down the ARR by around $10 million to $25 million—let’s say 3% to 4%—but the 20% EBITDA margin was a floor, as we said at the Capital Markets Day 2 years ago, and we’ll show you much higher margins.”

So, we’re still at a Rule of 40 score of 45, even with one-time lower revenues. Then, probably—again, speculation—around Q3 or Q4 they’ll give you guidance for 2026 showing acceleration in revenue. If those federal budgets are delayed and don’t come into the 2025 budget, it’s going to just push.

It’s not like those guys can just cut Cellebrite. It’s a matter of when, not if. This is why we are such a big fan of the company: it’s so mission-critical, so you always know it’s a matter of when.

Let’s say another company, ZoomInfo. When you buy ZoomInfo and speak with a ZoomInfo customer, it’s always like, “Yeah, I can manage without them.” It’s not so critical for me. It’s more project-based; I’m not relying on them on a daily basis. No, as we said up front, if you have a terrorist event on your hands and you need to crack the phone, and you don’t have this product, you’re out of luck.

Let me ask you a question on that front. I have so many questions left to ask, but these guys sell to the FBI, right? We’ve mentioned the Trump example multiple times. They don’t have it in the Butler office; they do have it in the Pennsylvania office. Why, when they sell to the FBI, does the FBI have the most updated version in some offices and not in others? Is it that they haven’t paid the upgrade cost for those offices? Why haven’t they paid the upgrade cost?

When you say you sold to the FBI and it’s a subscription-as-a-service product, I think, “Cool, everyone’s got access, and they’ve got updated versions at all times.” I buy Excel for my company, and everybody’s got Excel updated all the time. Why did the FBI not have it in all their offices, updated?

I think it’s a matter of budget and budget allocation. But again, for us, it’s always a matter of when, not if. We spoke with many customers. Some customers can push back a little bit on Inseyets. By the way, next year is a very important milestone: the shelf life of their legacy product is going to end, so all of their customers have to migrate to Inseyets. It’s going to be another 50% migration. Some customers can push back that the price uplift is a bit aggressive.

We have to remember that Cellebrite’s 2nd half is so much stronger, both in profitability and ARR. It’s usually, again, rough numbers, 60% of the ARR, because of the way the fiscal budget of federal agencies works. Usually Q2 or Q3 is when the year ends for them. Usually—and again, this is speculation—maybe there’s going to be a budget flush in Q3, because everyone was waiting in Q1 and Q2, as you mentioned, waiting for DOGE, waiting for the BBB.

You’ve used the term “it’s a matter of when, not if” a few times, and that’s one of the things I like here. Mission-critical, unique software—it’s almost an insurance product. Hopefully, we don’t need it, but if there’s an event, we’re going to need it. If you’re the NYPD, you’re going to need it.

They’ve mentioned a few times that in 2027, the U.S. is hosting the World Cup, and in 2028, there are the Olympics in Los Angeles. So those are 2 big events, and they think that flushes out the budget a lot. On 1 hand, I can understand why: you’re worried about terrorist events and everything. I get that. But they’ve also mentioned that both of those events are huge human trafficking risks, and I don’t understand why they’re human trafficking risks or why they should really increase the budget that much. Again, the Olympics are in LA, and I think the World Cup final is going to be in the New York area.

These are people who already have Cellebrite. As we mentioned, usually if they have 1 of their products, they don’t have enough of it. So they may have it in 1 of the offices and need more, because sometimes there’s another element of the pricing: it’s consumption-based.

The 2nd product is Guardian. If you want to share the evidence, you have to use the Cellebrite software in order to maintain the chain of custody. No one is allowed to tamper with that evidence, so the evidence is actually admissible in court.

The 3rd product—I think there was a shooting, I forgot where it was, in Oakland—and there was a massive shooting. They used Cellebrite Pathfinder. We didn’t speak about Pathfinder; it’s their 3rd product to actually analyze what happened there.

What Cellebrite is trying to do is become a full platform instead of a one-trick-pony solution, covering a case from case to closure. They call it “from case to closure.” First, you get the phone, crack the phone, and analyze all of the structured and unstructured data. Then you use Guardian to move the data while maintaining the chain of custody. No one is allowed to tamper with the evidence, so the evidence is admissible in court.

Then you’re approaching a very different customer base. It’s the same logo but a very different customer base. We call them investigative units. Those guys actually need to build the case and present it to the judge. The penetration of Pathfinder and Guardian is still between 2% and 5%, and I think it’s now ramping up. Those products are around 25% of their growth.

Pathfinder’s CAGR over the last 3 years is 50%, which is incredible, and Guardian’s is actually 100%. By the way, their cloud product is approaching 20% of the business and is growing 100%–150% every year. So I think it’s just a matter of time—sorry to repeat the same formula—but those agencies are buying.

I think what Cellebrite was lacking more was its go-to-market. They were focusing on Inseyets. They didn’t put as much focus on Guardian as they needed to, and then they had to realize how to sell to a very new group of customers that didn’t even know them.

For the DFU guys—the digital forensics units—the technology in general is very penetrated. It’s a logo-play duopoly, in a way. Everyone knows Cellebrite, but in the other departments, no one really knows them. Then you have 4 or 5 other new players that we can actually speak about. It’s Cellebrite; sometimes it’s Axon with Evidence.com, sometimes it’s Motorola, sometimes it’s Palantir with a much higher range of customers, and then it can also be Cognyte and NICE, 2 other Israeli companies.

But we think the market is always mistaken because it thinks Axon is actually competing with them. As we know, Axon’s DNA was always more internal investigations, not external. It’s about monitoring the police or monitoring the agency, not monitoring outside of the agency. So Cellebrite and Axon are not really competing. Some of the customers we spoke to are actually using both of them.

But it takes time to actually teach the customers. This is why, by the way, around 250 employees at Cellebrite are actually teaching customers how to use the Cellebrite products. We need to remember that public safety’s software penetration is still, I don’t know, 15 years behind all of the other software companies. For other customers, it’s very hard for them to adapt, and you can also see it with Tyler, which is selling ERP sometimes also to public safety.

That’s great. I have so many questions left. Let me just pause here. We kind of addressed it, but I always like to make it clear: the market is a really competitive place. Again, I think we’ve hit on a lot of different things here, but what do you see in Cellebrite that the market is missing or discounting that makes this a risk-adjusted alpha opportunity?

I think it’s such a great question, but whenever you see 1 of your top picks—any investment recommendation, full disclosure—going down every day by 4%, you ask yourself, “What am I missing? What am I missing? Why is the market so bearish here?”

First, Cellebrite didn’t have its federal ramp, and everyone was assuming that the federal ramp was maybe going to be delayed to 2025. Then, a week ago, they got support for the stock. It was up 4% that day, and then the day after it was down 3%. So I’m like, “What is the market missing?” Then I was like, “Okay, maybe it’s missing the $20–$25 million of ARR.” Then I just realized it’s a combination of 4 big things.

1st, it’s Q2: everyone is afraid of Q2. It’s new management. Everyone is afraid that the growth algorithm doesn’t actually make sense, that they will decelerate as they move beyond 2026, and that they will have to pull the 2028 guidance. It’s more of a “show me” story for the Street. It’s not like people are skeptical about the quality of the business.

What we think the market is actually missing, and we need to remember, is that, 1st, it’s not new management. The chairman has been the co-CEO already for the last 2 years, so he’s very familiar with the story. 2nd, the CEO and the CFO will probably be aligned with shareholders.

I would say the 3rd thing, in terms of Q2, is that around a month ago they bought—how do you pronounce it?—Corellium, which will probably add around $15–$25 million of ARR on a full-year basis. So let’s say you just take half of the $20–$25 million of ARR; it’s going to be another $10–$12 million of ARR. Cellebrite is just lacking 1 or 2 big customers within the federal government, and once they go to the Fed, I think they can actually do it.

I think the market is more skeptical about the growth algorithm, Q2, and new management. Once you feel comfortable with everything, I think it will also be good to break down the growth algorithm at a high level. 1st, we need to remember that 90%–96% of the growth—the 20%-ish growth, let’s say—is coming from the installed base.

Then the question is how it’s divided within the installed base. 2%–4% of that 20%—let’s call it 10%–15%—is actually coming from the Inseyets migration. I would say the next 5%–7%, the majority of the growth, is actually coming from new products like Guardian and Pathfinder. The other part is seat expansion, with Inseyets again, which is another 3%–4%. They usually increase pricing every year, and it varies between 2% and 4%.

Usually, they can follow Magnet because Magnet raises prices by 5% to 10% every year. But we need to remember that, ballpark, the sector itself is growing 15% to 30%. So, Cellebrite has a lot of room to grow, and the sell side and the market in general are so focused on the idea that Cellebrite no longer has room to grow because it has penetrated all of those logos.

We actually see 2026 as a year of massive acceleration. If this year they pull down their guidance, as we mentioned, those federal agencies will have to buy even more of Cellebrite's products.

I think I tried to give the full picture there, and I think that was great. Let me ask a slightly different question. Let's put aside, as we said, the bulls and bears—everyone is terrified of the Q2 print, and possibly for good reason. Let's put that aside, because, yes, the market could be weak, but that does not matter for the long-term thesis, right?

When you think about the long-term thesis, we've laid out a lot of bullet points here: the acquisition, the growth, the execution. What keeps you up the most at night with the long-term position? Is it Axon as a competitor? Is it Palantir? You said they're selling very high-end software, and high-end can go to very low-end pretty quickly. What keeps you up the most at night? I mentioned Apple working with law enforcement to crack phones more. What's your biggest worry here?

To be honest, it's mainly execution with the new management. That's what keeps me up at night. I think Palantir is crushing it in the upper, upper, upper market, and we need to remember that. Every time Palantir takes a new customer, it's such a big contract. It could vary between $25 million to $100 million, or even more sometimes. Cellebrite doesn't really operate with those same customers.

If Cellebrite is not going to get acquired, I don't know, 2 years from now, then we are asking ourselves internally what kind of IRR we can actually achieve here.

So let me ask you that. If a company is growing 20% to 25% every year—let's assume 20% ARR with higher margins every year—and you assume a 20x free cash flow multiple, what kind of return would you expect here?

I would expect—

You'd expect a pretty good return, but let me put that back to you in a different form. I think there are 2 assumptions there, right? If I were a bear on this stock, I would say, “Hey, Cellebrite right now has a $3 billion enterprise value, and everything Andrew and Orel are saying is really interesting. But they're going to do about $120 million in adjusted EBITDA this year, based on their prior guidance, and it seems like they might take that down.”

There is $30 million—

Take out the stock compensation. I think the free cash flow will probably come up, but that's speculation.

I'm just using their Q1 guide of about $120 million in adjusted EBITDA. It's adjusted for stock compensation. It's not, you know, what we've seen at some other companies, but there is real stock compensation here—about $30 million. So if I were a bear, I'd say, “Hey, that's $90 million of pretax earnings after stock compensation. This is a $3 billion enterprise value. Yes, the 20% growth is great and it should be very high margin, but once you account for that stock compensation—and, eventually, no one grows 20% forever—you run into all the market-size issues that Orel kind of started the podcast with.”

Are you really playing for a lot of alpha here? I think that would be the other pushback.

Usually, when we analyze a software company, we always look at free cash flow because, as we know, most software companies will try to manipulate the numbers. For example, one of the HCM players, Dayforce, capitalizes the sales commissions of its customers for 10 years and also capitalizes R&D expenses. Then you're asking yourself, “How much cash is this business actually generating—real cash?” It's crazy.

You're preaching to the choir here. On the Tora call, both analysts said one thing that jumped out at me when I was just starting to research the company: “This is like the cleanest adjusted EBITDA you'll ever find in tech.”

The first thing I did was look for capitalized R&D expenses. None of that. It is a very clean business. I love everything you said. I'm sorry to interrupt you, but I'm so excited about what you said.

Me too. Every time you analyze a software company, you analyze the SBC. You speak with the CFO, and he's like, “No, it's not a real expense.” And you're like, “Yes, of course it's an expense. I'm paying this as a shareholder.”

Then you see a company like Cellebrite, which I love because of its DNA. It's a bootstrapped company. They're very efficient. Every dollar they spend, they monitor so carefully—every incremental dollar. So we think more about Cellebrite on a cash basis.

Right now, for 2025, it's a 5% free-cash-flow yield, and in 2026 it's around, I would say, 8% free-cash-flow yield. For a company that's growing double digits, is a market leader, is operating in a complex technical environment, and is so efficient, to me it sounds like a no-brainer once you get past the uncertainty.

Everyone I speak with on the buy side is especially focused on Q2. We have Q2 on our minds, and then we also have the new management team. We don't know how they're going to guide. So I'm taking the view that the new CFO has historically been conservative. In the first quarter after he arrived, he actually maintained guidance.

I think the way he was saying it was that the market is punishing us for crimes we haven't committed yet. And I'm like, “100%.”

So right now, let's say they pull down the second-half numbers by $25 million, and also pull down the 2028 guidance, because the CFO wants a clear playing field to beat and raise. Again, let me just speculate here: I think Q2 could actually be a clearing event because all of the uncertainty is kind of gone. The FedRAMP ramp and the sponsor issue are already behind you, and Q2 is already behind you. Now the guidance is very conservative. You know how the CFO is guiding, and maybe they're actually going to disclose some buybacks, which I think is not a must, but they can do buybacks.

By the way, even the rest of the position, I love the way they are stewards of capital. I love it. I met with the company in Chicago at the William Blair conference, and most of the buy side was saying, “They're going to do transformative M&A. They're going to do a big acquisition.”

The company said, “We're going to do between $10 million and around $150 million, and we're going to buy a company that we've already known for the last 5 to 7 years.” I love the company they bought. Why am I such a big fan of it? The company is growing 50% to 80% every year. It's very mission-critical, and it has the same logos. Cellebrite had already been partnering with this company for the last 7 or 8 years. The founder is going to stay with Cellebrite, and the deal was around $120 million.

So they paid about 5x for a company growing 50%. You can see the way they think about capital. They're such solid capital allocators. They're not just throwing money around like they don't think about shareholders. They are very aligned with shareholders—the way they think and the way they operate.

That's why I sleep very well at night. Of course, we have a new CFO, and we don't know how he's going to guide. But historically, he was always very conservative, and he treated capital like a shareholder would treat it. He is a very thoughtful steward of capital: buying back shares when the stock is down and using the company's currency to buy new companies when the stock is very high—let's say, around $25 to $27.

It's great. I love everything you said. I'll ask 1 last risk question, and then I have a few random questions I want to ask.

AI. The company is starting to use AI for some of its data analytics, and I'm sure they've already been using it for all sorts of things. But it strikes me that you could tell me this company is a massive AI beneficiary for a variety of reasons: more data to analyze, the ability to analyze it better using AI and language-model tools, and probably very unique data in terms of analyzing it.

You could tell me that, or you could tell me, “Hey, AI is a risk to every software company.” And you know what? It's a real risk to Cellebrite. You're trying to crack a phone and interpret that data. Well, once the phone is cracked, AI can analyze every cracked phone's data equally. All of a sudden, you can go to ChatGPT for discovery. I'm probably being a little facetious there, but you can see how AI could be a risk along a whole lot of vectors here as well.

I'd love to ask you: AI beneficiary or AI victim? Where do you think they fall?

I would probably say neutral. We need to remember that those agencies didn't even move to the cloud; they're just transitioning to the cloud. It's going to take them another decade, or 5 to 7 years, to adopt massive AI tools—if they adopt the AI tools at all.

Cellebrite has been investing in AI since 2015 or 2016, so they are the market leader by far. This is why, when we model them, all of the margin unlock will come from sales and marketing.

We need to remember that vertical SaaS companies usually have very high margins because the way you acquire a customer is more product-led growth. A customer will be your advocate. They will advocate for your products, and you will get a sticky customer because everyone knows you, in a way. But you will spend, let's say, 25% to 40% on R&D—Veeva Systems, for example.

I remember the same pushback we used to get on Veeva. We always got the same pushback on Veeva.

Yeah, but Peter, the founder, never really speaks about AI because Peter was always saying, “Listen, those pharma companies have much bigger problems than AI right now.”

So, if AI is going to be a thing, it’s probably going to be a thing in the next 5 to 8 years, in a way. I would say that before then, they have some other problems to solve. Again, just to sum it up, I think Cellebrite will actually be a beneficiary here, but I would say it’s going to move the needle in the next 3 to 5 years or so. Maybe I’m mistaken; maybe adoption is going to be much faster, which I don’t know. But if there is going to be adoption, remember that in order to penetrate those logos, it takes years of trust, FedRAMP, and classification. Again, it’s going to be Cellebrite, not some other player.

That’s fantastic. Last question, and then I want to give you the chance to wrap up. I think we’ve had a very comprehensive discussion, but obviously you’ve done tons of work here, and I want to give you a chance to hit anything just on ownership structure. We’ve already talked about some of it. This was a de-SPAC that carries warrants, but I think there’s a real company here.

We talked about True Wind, who was the SPAC sponsor. It still owns 5%, and they get a kicker at $30 per share. I just want to ask one last question: Sun Corporation, the Japanese corporation. This company’s ownership structure is so funny because they started as a US company, they get bought by Sun Corporation, a Japanese company, they get an investment from an Israeli PE firm, and then they go public through a de-SPAC. It’s just crazy.

Sun Corporation owns, let’s just round it, 50% of the company. You look at this company and say, “Oh, cool. They’ve got Sun Corporation, a Japanese company, as a 50% owner.” Japanese companies—maybe it’s changing a little bit, but they’re not exactly known for being urgent with their capital, let’s put it nicely. When you look at Sun Corporation here and you look at the ownership structure, what do you think Sun Corporation’s motivations are?

Of course. I’m not trying to criticize anyone, but Sun Corporation actually sold to IGP, the Israeli PE firm, and they sold Cellebrite. A big portion of their Cellebrite shares are around 1x revenues—I don’t know, 10x-ish. So obviously, they need an activist. They knew what they were holding and what kind of a gem they were holding here.

Now I think Sun Corporation actually has around 3 activists that we spoke about, the majority of them. Everyone is pushing Sun Corporation to actually sell Cellebrite. There are so many funny campaigns there. There’s even a funny song about Sun Corporation. I forgot who the activist was, but there’s a funny song that every year goes something like, “A year goes by, and Sun Corporation again didn’t provide our shareholders any value because they don’t care about shareholders,” or something like that.

I think it’s a matter of time before Sun Corporation sells Cellebrite. I’m not saying they will sell everything; maybe they will just try to sell it piece by piece. I can just say that we can feel comfortable that Adam owns Cellebrite and also Sun Corporation. He will act, I don’t know, in our benefit, in a way. We should remember that Adam was formerly at KKR. He was a partner at KKR, working on software and fintech companies and big transactions, so he knows this kind of stuff and he’s very motivated.

For those who don’t know, Sun Corporation, the Japanese company, was a popular value-investor, venture-style play because they owned a lot of Cellebrite and were trading at a discount. True Wind offered to buy Sun Corporation, if I remember; there’s a lot involved. But I wanted to mention it because anyone who looks at the beneficial ownership will see a Japanese holding company with 40%.

I think it probably works out well, but it is something to keep in mind. Look, this has been great. We’re almost at an hour, but I just want to stop here. I think we’ve covered most of it—not all of it, because I had so many questions and found this so interesting—but I’d love to pause here. Is there anything we didn’t hit, or anything we kind of glanced over, that you think we should have discussed in more depth?

I think Cellebrite is a very, very, very complex story. It took us hours of field research just to actually understand the thesis. Every time, let’s say you do 20% of the research and you’re like, “Okay, what is this?” Then you do another 20%, and you hit some kind of pushback.

I think it’s a very complex story, but once you do the work, you understand Cellebrite—it’s such a gem. I think we need another 4 hours to actually expand on every product: Why is it mission-critical? Where is the penetration? We didn’t even speak about the private market, which is around 10% of the business and is also growing 20% to 25%. There’s a lot—a lot—to still elaborate on here.

No, it’s really—actually, I did have one other risk. You mentioned the private market, and you said, “Hey, tier-1 banks and insurance companies especially are looking into it—”

Companies in general, yes.

But there is a risk here. I think it was their investor day, or maybe it was their most recent earnings call. I can’t remember, but they said, “This is powerful technology.”

I don’t know if you’re familiar with the Batman movies, but in The Dark Knight, they turn everyone’s cell phone into a surveillance device. It’s not quite there. The ability to crack anyone’s phone—if you sell that in the wrong hands, that’s a potential criminal tool instead of a criminal investigation tool.

They mentioned, “We might have to cut off some customers because of what they’re doing.” I could also imagine a kind of dystopian police state where Cellebrite is used to crack open dissidents’ cell phones and arrest dissidents or something. Do you worry about that, or how do you think about that risk? Because it might sound over the top, but that is a real risk for this company.

100%. I think, first, I’m so happy that you brought this up, because in 2021 and 2022, they had to drop many customers in their Asia segment. So right now, if you think about their GRR, or gross retention rate, people can assume it’s actually 90%. They have 10% churn every year.

Then you can ask me, how is it so mission-critical if 10% are churning every year? You actually have to break it down: 2% to 4% is coming from those countries where Cellebrite doesn’t feel comfortable operating. They have to drop those customers in a way. Cellebrite also has an ethics committee that tells the company where it can operate and where it can’t operate.

I think it was a headwind, but right now it’s not a headwind anymore within the private sector. I think it’s not a headwind because Cellebrite doesn’t put so much effort into this segment, and Cellebrite is not the market leader there. There’s another private competitor that is the leader by far within the private segment.

If you ask me as a shareholder, I prefer the public sector. It’s so much more stable and so much more efficient, and the private sector can actually work with Cellebrite on a project basis. Let’s say they’re afraid that one of their employees stole some data to launch his own company, so they’re using Cellebrite on his phone. This is why the churn within the enterprise—the private side—is much higher than on the public side, and it’s especially critical for them.

No, that’s great. Well, I’m going to have to wrap it up here, unfortunately, but this has been—again, I said it up front, and all my podcasts are my baby. I tend to like all the ideas, but this has just been a fascinating idea to dive into.

It’s got a lot of event-driven angles and a lot of quality compounder angles. I’m honestly surprised it’s not more popular among the value-investor community. You’ve done great work here. I really appreciate you reaching out and coming on the podcast, and we’re going to have to have you on again because the software world is an interesting place, and there are plenty of other interesting companies.

Thank you so much for having me.

A quick disclaimer, nothing on this podcast should be considered investment advice. Guests or the hosts may have positions in any of the stocks mentioned during this podcast. Please do your own work and consult a financial adviser. Thanks.