Yet Another Value Podcast - Firebird's Steven Gorelik on why Akamai's security business makes $AKAM interesting

Episode Date: June 13, 2024

Steven Gorelik, Lead Portfolio Manager at Firebird U.S. Value Fund, joins the podcast to share his thesis on Akamai Technologies, Inc. (NASDAQ: AKAM), the cloud company that powers and protects life online. Chapters: [0:00] Introduction + Episode sponsor: YCharts [1:54] Overview of Akamai Technologies and why its so interesting to Steven [13:09] $AKAM valuation [15:39] Cloud computing and delivery segments [29:10] Why $AKAM now and is $AKAM being run for stockholders or being run for the employees? [34:12] CEO insider transactions [36:44] As a generalist investor, who isn't solely focused on cloud/security companies, why isn't $AKAM just some growthy business competing with major players? What does Steven know that specialists aren't seeing with $AKAM [44:52] Ways $AKAM could benefit from AI, or as an AI play [51:32] Does cloud computing, delivery and security businesses belong together? Today's sponsor: YCharts This episode is sponsored by our friends at YCharts. With all the various job functions that advisors are tasked with, your time is extremely valuable—and often scarce. YCharts is a platform centered around efficiency, and built with speed in mind. With an intuitive and user-friendly interface, YCharts helps save advisors 29 hours per month while uncovering better, and new, investment ideas. Need a way to help clients visualize their financial future, reinforce the importance of consistent investing, and guide them toward informed decisions? Tools for scenario building, portfolio construction, and proposal generation can be the missing piece to your service, so you can act on an idea right when the light bulb flicks on. Start your free YCharts trial and see how YCharts is a one-stop shop for growing AUM with fewer hours spent on investment management: https://go.ycharts.com/yet-another-value

This episode is sponsored by our friends at Y Charts. With all the various job functions that advisors are tasked with, your time is extremely valuable and often scarce. Why Charts is a platform centered around efficiency and built with speed and mind. With an intuitive and user-friendly interface, Y-charts helps saves advisors 29 hours per month while uncovering better and new investment ideas. Need a way to help clients visualize their financial future, reinforce the importance of consistent investing, and guide them toward informed decisions, tools for scenario building, portfolio and proposal generation can be the missing piece to your service, so you can act on any deal right when the light bulb flips on.

Click the link in the show notes to start your Y-charts trial for free and see how Y-charts is a one-stop shop for growing AUM with fewer hours spent on investment management. All right, hello, and welcome to the yet another value podcast. I'm your host, Andrew Walker. If you like this podcast, we mean a lot if you could rate, subscribe, review wherever you're watching or listening to it with me today. I'm happy to have on for the first time. Steve Gorlick. Steve is a portfolio manager at Firebird. Is it just Firebird USA value or is it Firebird

overall? How do we say it? So the firm is Firebird management and we do, there's two kind of sides to what Firebird does. The traditional one, something we've been doing for 30 years of this is through European investing. So, and that's kind of the beginning of Firebird management. And then for the last 10 years or so, I've been running the Firebird US Value Fund under the same umbrella where we've taken the same fundamental part of what we do in Eastern Europe, which is focusing on cash flows, figuring out how companies make money, how companies spend money, and applying

it to a different market we feel comfortable with. Perfect. Perfect. Well, we've got a U.S. focus name to dive into today, but before we get there, the same way I start every podcast, a disclaimer to remind everyone that nothing on this podcast is investing advice. Please consult a financial advisor, do your own work. Remember, nothing here is financial advice. Steve, the company we want to talk about today is a fascinating one.

It is Akamai Technologies. The ticker for anyone listening is A-K-A-M. And there's a lot to dig into, so let's jump right into it. I'll toss it over to you. What is Akamai and why are they so interesting? Absolutely. And thank you for the opportunity to speak. So Akamai was started in the 90s as a CDN, which is a content delivery network business.

The way that I would think about the content delivery networks is that this is a network of servers and ISP relationships around the world that allows content creators or websites, whatever you want to call it. So any kind of internet information allows it to reach the final consumer quicker. So quite often, this information is coming from one place and that has to be distributed around the world

so the people, customers, so the consumers in Asia, Africa, Europe, US, wherever, get the information quicker, so you're reducing the latency. by having the servers and the local relationships all over the world. For listeners who are familiar with Netflix, like I remember, I think it was the book Netflix.

One of the things they discovered was, you know, if they stored everything in the central database out in Nevada, there'd be, you know, 50 milliseconds of lag when you and I are loading up in New York City or Florida. And they discovered actually by putting the data on a closer local center to eliminate some of that lag, their viewership times, their customer retention, everything actually went up.

And that's such a silly one because it's just streaming, but you can imagine, I mean, the company does video game leaderboards where that needs to be like up to the absolute millisecond. So you can imagine things where there's even better demand, but that's just a great example that always lost in my mind. And that's exactly their top customers, companies like Netflix or the gaming companies. And then if you think about where, if it's Netflix,

where you're just downloading, okay, you're comfortable maybe with that initial delay, even though as you're correctly pointing out, it's still uncomfortable. But if you play in a video game and there's this latency between your action and what you actually see on the screen, that's a big problem from a point of view of user experience.

So the CDNs, the companies that are making sure that that user experience is better. They've been around since the late 90s. There's nothing particularly, and it is essentially a network of servers and ISP relationships around the world. Akamai was started out in late 90s, I think in 1998, it probably has the best and the biggest network and standalone CDN,

and this has been their bread and butter for the last 25 years. That said, this business for the last few years has been stable to declining. And part of the reason why that's happening is because, A, you have some competition to merge. And so there are other players that are CDNs that are providing a cheaper service that is not necessarily as good, but it's cheaper, and for certain class of customers, they're absolutely fine with it. But the other bigger issue for them. Can you just provide an example of a competitor providing a cheaper, not as good service, and kind of what a not as good service would look like? So Fastly is a good example of a smaller competitor, and the reason why they're not as good

is probably because they have hundreds of servers and not thousands and hundreds of relationships and not thousands. So from the point of view of, I think they're a little more susceptible, to things like the DOS attacks because you just don't have as much of a distribution. If you're looking for reliability, like reliability of service, I think the fact that Akamai has thousands of servers and this grade of an infrastructure does make a difference. That said, if you are just simply hosting a website, you probably don't care as much. And so Akamai's product is more of a enterprise product. and is more designed for kind of larger customers.

They do have some of the smaller customers as well, but then if you go with some, like Cloudflare or Fastly, which are kind of other CDNs, they will have more an entry-level product that Akamai does not necessarily compete with. You said Enterprise, and I think that's right, but I also think it's a little bit performance, right? Like a lot of websites, I know, because I know when they go down,

it says, you know, trying to connect to Cloudflare or something, like websites where 35 versus 50, milliseconds who cares like a lot of them will go with that but as you said gaming video like especially streaming video live streaming video those types of things probably going to go more with a performance product because milliseconds matter there so sorry to interrupt no you're absolutely right and you're pointing out the right things but the other problem that I've been facing in CDN from point of view of the customers for this what they call this delivery segment is that some of their largest customers, I think Netflix

is one of them since you brought them up, have gotten big enough where that started making sense for them to invest into their own CDNs, to take at least a portion of the traffic that would normally go to Akamai and putting it onto their own network. So when we're talking about

companies, obviously companies like AWS or Microsoft or Netflix, they will have they now have a portion of the CDN they do on DIY basis. And that takes volume and revenue away from Akamai. So that's why if you look at Akamai's delivery segment, that's something that used to be

growing at nice double-digit rates, but for the last two years has been declining. And that's the really good dynamic. That's happening there. And I would argue that that's the reason why the company is cheap, but that's not the reason why the company is interesting. Because at this point, delivery is only 40% of their revenues. and the other 60% is coming from two segments and two businesses that they've built on top of the, first of all, the network that they have,

but also very importantly, the relationships that they have with the customers. And those two businesses is a security segment. So we're talking about cybersecurity. And the more recent one being compute. So this is essentially the same thing, infrastructure as a service that you would get from AWS, Asia, Google Cloud. but with slightly different use cases and also actually in part of it, optimizing the edge network that they have,

utilizing the edge network that they have. The security business, they started building out in mid-2010, and it's kind of being built through a combination of acquisitions and organic growth. And what they seem to do is they seem to buy a high-quality company with certain capabilities that they want to add and to be able to cross-sell to their customers. And they usually buy these companies at multiples that may look expensive on price-to-sales basis. But once they plug in this product into their cybersecurity offering and able to cross-sell to the customers that they have, there's a really rapid growth.

So what they have, and the security business has grown into to represent, I think, about 45% of their revenues. It's been growing by 20 to 25% per year. And this is something that they've built. Initially, it was a, I think, like, $400 million business back in 2016. Right now it's $1.7 billion. And in order to build that from point of view of the acquisitions that they have done through 2023, it doesn't include an acquisition that just made in 2024. But they spent probably about $2 billion on different acquisitions, and they've built a business with $1.7 billion of revenue.

And if you take a look at security companies that are growing at similar pace and that are trading on a standalone basis, those companies usually trade at six to ten times revenue. So you can pick your number and can figure out what is worth from a point of view of the valuation here. I think the reason why they're able to cross-sell, and I touched them initially, is that they have at this point thousands of over a thousand customers enterprise middle-sized businesses that are already paying over $75,000 per year

to them for various services, various infrastructure services that they provide. It could be just the delivery business, the CDN business, it could be a combination of delivery and cybersecurity, but we're talking about a fairly large pool of customers. where they're already talking to the decision makers, the CTOs and the CIOs, and it's relatively easy for them to go to the same decision makers and say,

look, we're already getting great service from us on the CDNs. Let us add something for you that is actually fairly synergistic. So when you have a delivery business and you add to it a protection against DDoS attacks, it's very synergistic. It makes a lot of sense. So you can cross-sell that. And companies like Cloudflare, they can put it. all to get into one offering. Akemae conveniently, from our point of view as investors and consumers, they separate,

so you can kind of see where that revenue is coming from. But you can see that this has been an effective cross-sell strategy. One example, yeah, go ahead. No, go ahead, go ahead. Go ahead. No, so one example of what they've done, because they've provided some data around it, is that they bought this company back in 21 called Wardicor, which is a zero-trust security. So this is more like an authentication offering, and I apologize because I'm a generalist.

So we're talking about technologies, and I feel kind of our place once we get deeper. But, you know, I do understand things like who you talk to within the company. I can understand the general market. And so they added the companies, the GuardiCorps, that provides zero trust security. They bought it for $600 million, $600 million back in 21. At the time of the acquisition, this was a company. about 30, I think, $30 million of revenue. Over the next two years, by cross-selling to their customers,

they've grown up to $100 million. So you're talking about tripling, and they provide the data on what percent of their customers had zero trust security back in 22. So at that point, it was 12% of their customers. We can probably extrapolate it to about 20% right now. But if you take a look at some of their older security offerings, like the DDoS protection, those are being crossed sold to like 50 to 60%.

Let me hop into that example real quick, because this has a question I had towards the end, but I think since you laid off the numbers nicely, this is a great question. So one concern, and I think you can see this, you know, again, you've studied this company a lot longer than I did, but I just read like the last earnings call on the last two. They go to a lot of conferences to the last two conference. But one concern people have is what they're paying. And even the company will say, hey, we pay big numbers. So let's use the numbers you just laid out, right?

They buy a company for $450 million that's got $30 million in revenue. And they grow it really quickly over the next two to three years to $100 million in revenue. I mean, that is a operationally, that is a success. But as a shareholder, I'm just using the metrics you use, right? If I think security companies are worth six times revenue, right? 100 million in revenue today, six exit, it's worth $600. million. I paid $450 million three years ago. I can guarantee you that there was a lot of cash, a lot of stock comp and everything there. When I look at that, I say, hey, you guys paid a huge

multiple. This went about as well as it could go, you know, triple revenue in about three years. It went really well. And I'm looking at this. I'm saying, did we really create like a lot of value here? Does that make sense? It does make sense. And it was $600 million. It wasn't $450 million. You bring up a very good point. That said, if you look at total spending on acquisitions since 2014 in security, and I went through it and I tried to understand which ones are security, which one or not. But most of the acquisitions, aside from Linode, which was compute, that was most acquisitions in security.

We're talking about total spending of $2.4 billion. At the end of 2023, this was a division with 1.1.4 billion. $1.7 billion of revenue. And margins, they don't show those margins specifically, but then they're in yesterdays, they say that the margins there are better than in the compute business. So now, if we're talking about spending $2.4 billion, on creating a $1.7 billion revenue business that is growing at 20% per year, if we put a 6 to 10 multiple in that, did they create value?

And I think one of the things that's interesting here, here is, as you alluded to, like, this is a, call it a 15 billion EV company. If you put a 6x multiple on, as you said, they'll do about 2 billion in security revenue this year, growing 20%. You're kind of saying, hey, is the security business covering the entire, is the security business covering almost for the entire EV? Now, I think people, reasonable people could push back. Six X revenue is still a big multiple, 20% growth, a lot of things. But I think that's part of the interesting. Why don't we cover, if we're done with security for now, why don't we quickly go through cloud compute and then we can jump into some of the questions

and everything? Sure, so cloud compute, and I'll give you a little bit of background, kind of my history with the company and with the cloud compute. I started looking at the company back in 2017. At that point, we didn't invest into it, but I started looking at it. And I talked to a fairly senior technical person at one of the gaming companies at the time about how they use the CDN business, so the delivery business. And he said a lot of good things about Akamai as a delivery network, but he said the thing that they're missing is compute. I would love for them to have compute,

because especially for a company like gaming, where, and I think you alluded to it slightly earlier, where there is this need to do some computation on the edge in order to make things quicker, he said they were just missing that. And back in 21, and Akamai acquired a company called Linode, which was a relatively small infrastructure as a service, computational compute program. And when they paid $900 million, when they acquired it, I think I need to check my numbers, but it wasn't a significant size company. company. This was, I think, they had about $200 million of revenue. So they once again,

they paid a fairly sizable multiple on sales there. But the idea there is exactly the same as what they have done successfully, I would argue, in the security. Is that they're adding a third leg to the stool, selling services that their customers want and need, selling services to the same type of customers that they're already talking to, and they've built another leg to the business that they're going to be able to grow at 20% plus for the foreseeable future. The idea of the compute business

is that they want to participate in the cloud computing, which is right now a $200 billion business. They're saying that within that, yes, obviously they're hyperscalers, they're AWS, Google, et cetera. And they will be the vast majority of that business. And they need to be, and they have the

infrastructure to do that. But they're not going to be all of it. And there are certain use cases in which a distributed, still cloud network, but distributed on the edge servers around the world, actually makes some sense. One example we talked about is

in gaming. Because you already have, you brought it up yourself as far as when you're talking about the immediate response, when you're talking about leaderboards, gaming leaderboards, those can and should be done on the edge.

Another example is Internet of Things. So all of the devices, when we're talking about the IoT devices around the world, those computations are not necessarily the most difficult ones. When you're doing a difficult, scalable computation, that's where you really want to go

with somebody like AWS, because they do have the scale. But for the simpler computation, where speed and location matters, edge network makes a lot of sense. Another place where edge network makes a lot of sense is when you're doing a computation on data that really can't leave where it is. So when something is, I'm in the UK right now, but when you're dealing with something where you have to be GDPR compliant, healthcare data, but it often cannot leave the country. so once again you want to be able to you want to be able to do the compute on location

and this is not something that now akamai can provide to their clients the clients that they already have this episode is sponsored by our friends at ycharts with all the various job functions that advisors are tasked with your time is extremely valuable and often scarce ycharts is a platform centered around efficiency and built with speed and mind with an intuitive and user friendly interface, Y charts help saves advisors 29 hours per month while uncovering better and new investment ideas. Need a way to help clients visualize their financial future, reinforce the importance of consistent investing, and guide them toward informed decisions. Tools for scenario building, portfolio construction, and proposal generation can be the missing

piece to your service, so you can act on an idea right when the light bulb flips on. Click the link in the show notes to start your Y charts trial for free and see how Ycharts is a one-stop shop for growing a UM with fewer hours spent on investment management. Let me just, a quick question on the cloud compute strategy, because that does sound interesting, right? You've got these clients. You're at the edge. You've got, you've already got them, and then you just sell, hey, we can do this compute right at the edge for you, right? That sounds interesting. And I think they're obviously like, this is an hour-long podcast, three division, we can't cover everything. But one strategic question that jumps to my mind is,

hey, the core business, the legacy business here, we started this podcast off by talking about how it's declining. And a lot of the reason it's declining is because these large customers, your Netflix of the world, are going and building their own locations right at the site. So I think kind of a natural question is, hey, if the core business is declining and you're going to your customers saying, hey, we've got this delivery business, we're already with you, let us do your cloud compute at the edge. are you going to are you trying to sell like a growth market with a unique thing into a lot of customers who are already pulling spend away does that make sense it does make sense but the customers i would say that customers is pulling spend away but they're spreading the spend so in a way that so if you have somebody like netflix it does make sense for them to take on some of the data volume that is going through and putting it under their on their own servers.

It doesn't mean that they stopped using Akamai completely. It's just the it's no longer, and I don't think it ever was 100% because you never want to use just one CDN. You actually want diversity. So, but you want you they have a lower market share

of this customer, but they still deal with this customer on daily basis. So you still have the window into that customer and you can still talk to them and providing them a compute solution, once again, it's more often than not, people are going to multi-cloud solutions. And we didn't talk about it yet, but the solution from Akamai and from other people like DigitalOcean is quite often cheaper than it is from the large cloud providers like the AWS or Google.

And by the reason why it is cheaper is because you don't have the ability for that unlimited scale that exist with AWS, but most people don't need it. And more often, and quite often that has got people into trouble. Like I, my wife is, and she's a VP of software engineering, so kind of like I hear stories through her about it. And there are more and more people just realizing that they run a query overnight and then they went up to a $50,000 AWF bill. I remember especially when the AWS first came up.

Let me ask, I have two questions I want to ask there. First, cheaper. That's a little bit surprising to me that, you know, an AWS or a Google wouldn't be cheaper than this. Because, you know, you think about AWS and Google, like they have such huge scale. I would think they would be the cheapest for most baseloads. And then like an Acomy or a smaller person would actually be more expensive, but they would be more specialized. So you would get whether it's more handholding or more custom cloud compute. I'm just a little surprised to hear or because they're closer to edge.

you know, you get lower latency. I'm a little surprised to hear that they're cheaper. So why are they cheaper? So I was trying to figure this out as well. It's a very good question. And so one area where there is significant cost savings is that somebody like Google or Amazon, whenever they have the relationships with the ISPs, like that final leg, the final mile of data either coming in and coming out,

they are not necessarily price sensitive when they, because they want to have as much allocation as possible and they're paying up for it and when they're paying up for it they are then passing the cost of that to the customers because they can and you as a customer if you need to put terabytes of data very quickly you can you cannot do that with somebody like digital ocean or linup or akamai there is some scalability that doesn't exist but you're trading on that for a lower price. So you're training off unlimited scale optionality for a lower rate. One more question of delivery.

Think about it. Like if you need an Uber and it's just you and you have an options and you can get an Uber Excel, you can get an regular Uber. You can get Uber with a car seat with a baby car seat. They have different prices. And if you have a baby, you need one thing. But if you don't, you don't need to pay for it. Yep.

One more question on delivery. and then I want to go to back up and go to some higher level questions. When a Netflix takes, and I just use Netflix as just a great example, everyone knows it, when they decide to take a little bit of the delivery business in-house, right? Is this a case where they're saying, hey, New York, Seattle, Los Angeles, these huge markets, there's enough scale there where we can build our own internal solution there, but Montana, right? We're not going to go build our internal solution, Montana, and maybe

there because there's not enough scale and not enough people, it still makes sense to outsource and let, you know, Akamai roll up Netflix, EA sports and 500 other companies and put that in. Is that more what's happening with the delivery or would a Netflix be like, we're large enough, we've got enough data, let's just go direct everywhere we can. That question makes sense? Yeah, and that's my understanding as well. So that's where you will have, I mean, Akamai has, I think,

they say something like 4,000 services around the world, some of them are in New York. Some of them are in Montana. And if you are, and this is exactly it. So where Netflix, when it knows that it has a few hundred thousand, if not millions, customers in New York, and they know how much they use the data, how much they use the data, they know that they can put some of their own hardware and save money. I mean, Akamai is a very, we didn't talk about it, but Akamai is a very profitable business, where they have their contribution margin is, the cash contribution margin is like 75.

percent. And the overall profitability is very strong as well. So it's a premium product. And if you can do some of this in house reliably, you will. But it's only an option for Netflix, for EA sports maybe. It's not an option for somebody if you have a website that has a few million customers. Like, it's like the top five, 10 customers, and they're big, have that option. But the hundredth largest customer probably doesn't because the infrastructure, the money that you need to spend is significant. Does delivery ever, does it ever

stop being a declining business and go back to growth again? Because I could see a scenario where like the hundred largest internet companies are so big and so large that eventually they do a lot direct. But the long, honestly, I'm kind of surprised like the long tail of internet continues

even while the big, bigger like the long tail of internet grows so much. I could see an area where a future where the long tail just continues to get so big, eventually they start growing again, or you could tell me, hey, like, a worry would be the hyperscalers start offering some of this delivery on their own or something just because they grow to be all encompassing. So they do offer some, so somebody like AWS does offer their own CDN service as well, and that may make sense for some customers.

The way that company is talking about is that delivery is kind of zero growth business. They say, like, low decline to zero growth business, and that's how. I'm thinking about it as well. It used to be a nicely growing business, like I already mentioned. For the last, I think since 2019, the revenues have been declining. And I think that's what creates the opportunity in the stock, because people are uncertain how long is going to decline. We had a fairly weak Q1.

This is, you know, company has gotten more interesting because they had their delivery revenues drop 10% year over year, so people are really concerned. They attributed to a couple of very specific factors, include one of them being they didn't name the company by name, but they said a social network company facing regulatory oppression kind of sounds like TikTok. So, you know, it does, but at this point, all the social networks face. Yeah, right, but let me back up a little bit. If we zoom out, right, the company, which was a dot-com bubble company, which it's fun to look at their stock chart,

it because it starts with this enormous peak. I think it's like $350 per share in 1999 or something. And you and I are talking and it's, you know, the highest has ever hit since then is like $120, $130. So that's kind of fun. But ignore the dot-com bubblebee, but I think we can put 25 years behind us. If I zoom out in 2015, this is a $70 stock. And that's kind of right when they start doing the security, the going into security.

2021, it's $120 stock. Like, you know, six years, 70 to $120 stock. not the end of the world, but not exactly the best returns ever. As you and I are talking today, June, what is it, June 4th, June 5th, June 6, 2004. It's a $90 stock. And I guess what I'm asking is two things here. Why now? And the second thing I'm going to ask, which we could get into is I kind of look at some of their commentary

and they go on these conference calls, they go on their earnings call and they say, hey, we generate a lot of cash and we can do a lot of things with it. We can do, we do some buybacks, we do some acquisitions, like we can do a lot of things. And that sounds great, right? It sounds like you've got a manager team really focused on capital allocation. But one thing I worry about is a lot of the buybacks go to offsetting stock comp. And there is a lot of stock comp here. And I guess what I'm going to try to drive that is, I do kind of wonder, is this a business that's being run for the stockholders? Or is this a growthy, sexy business that, you know, it's nice to get some return.

But it's really being run for the employees. You know, I'm just looking at my numbers, 2003, 1.6 billion in adjusted EBITDA, 330 million in stock comp. And that actually ignores another 60 million of amortized dot com that they'd exclude. So a lot of this is going to stock up and it's largely offsetting the buyback. So I did throw a ton at you. I'll just pause there and right here. And those are all very important points. And I'll try to address them one by one.

If I forgot them, please bring me back. From the point of view of why now and what's been happening to the coming? to the company between, let's say, for the last four years or so where the stock price has stagnated, is that you had a, back in 2019, two-thirds of the business was delivery, and since then, as we already discussed, that's been a declining business. Right now, we're facing, we have a company where delivery is 40%. So at this point, it's already 60% that is fast, I would argue, fast-growing businesses that are growing at 20% or so per year. And you have, have 40% that is a declining business. Because of how the math works, over the last four or five

years, you had a company that was delivering mid single digits growth rates. So call it 3 to 5%. You have the dynamic of the declining delivery, and now the business is growing at 20% per year. From here, because delivery is smaller, even if it continues to decline at like 7% per year, And I think it's going to be a little less, but let's say 7% per year. We are now, we and other businesses grow at 20%. We now go into the high single digits, low double digit growth. And the multiples that the market pays for businesses that are growing at 10% are very different than the multiples that the market is paying for the businesses that are growing at 5%.

And we, I think, at this point, are actually very close to. to that inflection point where you're going to go from mid single digits or low single digits to high single digits and we never play for the multiple expansion but you are in the conditions where multiple expansion can can start happen so that's why the reasons for why now as far as the stock comp is concerned and i can share i do the stock comp slightly differently is what i try to figure out is how many shares they issue per year and you can do you can figure it out if you look at how many they're buying back and how many net share the net share change so they do two to three million shares per year of uh issuance to the

employees and you know that's not insignificant that's 2% of the company per year but even if you take that dilution into account going forward then let's keep it the same about 2% of the company per year, given the amount of cash that they generate, and the growth that they're generating in a free cash flow, I still think we as shareholders are reasonably well compensated. We also have, I mean, the CEO, who is the founder, he has a salary of $1 per year, and all of his, he's already a sizable shareholder, he has over $100 million worth of shares, and all of his companies in shares as well. So from that point of view, I do think that we're more or less a lot.

If I can add on, so the CEO, I was just making sure I had to pull it up and write, Leiden Thompson, you know, one of the reasons for why now, when you say why now, is May 14th. We're speaking June 6th. So last quarter, right after earnings, May 14th, he makes an open market buy of $2 million worth of stock at, let's call it today's prices, right? And that alone would be interesting, especially, getting paid a dollar per year. You know, he's doing $2 million worth

two million times his salary is the open market part. But the other interesting thing here is, you know, with open market buys, I think the buy is one thing. The size of the buy is one thing. But another good signal is what is the history of the open market buy here? And when you look at the history of open market

buys, there was the general counsel bought two shares at 111 in 2021, so $200 worth of shares. We can probably ignore that. I think, who knows what happened there. This is just what my rep said. But aside from that, there was not a single open buy since 2016, 2017.

2017, one VP buys $200,000. In 2016, the CEO, by $6 million worth of stock in the low 50s, right? So low 50s to, it's not the greatest return of all time, but 2021, as I mentioned, the stock was 120s, who had more than a double inside of about six years. That's actually a pretty nice return. But I'm just saying it because it's not like this is a CEO who buys a little bit on the open market every now and then. This was a sizable insider buy, and it was his personal first buy in almost 10 years. And I think the combination of those two, when I ask you why now, I think the CEO is kind of signaling, he sees exactly what you're seeing, if that makes sense.

I know you had a seat on your show recently, and one of the favorite things that he says, because I know he tracks a lot of the buybacks. And like the insider transaction, like he says, there's multiple reasons for why people will sell shares and, you know, life gets in a way. But there's really only one reason why people buy shares. You know, I love that quote. And I generally agree with it, though I can't tell you how many times I've seen a CEO who gets paid, you know, $500,000 per year. And he's been there for six years and the stock is down 70% and he goes and he buys 50,000 on the open market, which I'm kind of like. But different, we took a different moment. Yeah, but people say one, I think there's like maybe four reasons.

But in this case, I think that that quote certainly applies. Let me go to a different one that I think our listeners will be interested in. You know, one of the tough things for me, because I'm with you, I'm a generalist, one of the tough things for me, a lot of people want me to look at, I spend a lot of time in cable. A lot of people want me to look at like Enterprise Fiver or the B2-BeLacom. And whenever I look at them, I'm like, look, I, I, I understand the story. I get what you're seeing.

I want to get here, but I just worry. You know, I'm a generalist. I worry that there's people, and I can't tell you how many times there's been this sexy fiber, dark fiber, B2B play, where all the customers have all of a sudden fled. You said earlier, I'm a generalist. And I think a question would be, hey, Steve, you are competing like security is one of the hottest, toughest, most money going into its base. Cloud computing. It sounds great. Huge growth industry.

But you're competing with AWS. you're competing with Google, you're competing with all these startups that are coming in. I think people might just say, hey, like Steve, Andrew, me as a random listener, why are you guys the ones to look at this and say, it's cheap, it's hitting that inflection point, but we really understand the strategy and we're not going to wake up to, you know, six months from now, hey, all of our cloud computing customers are going with the largest or they're going with this hot startup or something. Does that question make sense?

It does make a lot of sense, and that's, I mean, you would be stupid if you're not asking that question. but the thing about it is that I think as a generalist we the advantage we have plenty of disadvantages that you just name but the advantage that we have is having the ability to take a look and having the mental models that are applicable the things that we've seen in different industries and apply them to other industries as well so this is a kind of a different example but in the past one of the investment one of the most successful investments that we've had in a past was in this company called aristonetrics And Ireland's the network, it's a really, really kind of successful company recently because they are, they've been grown at 20% per year. But they have this, and they make switches that go into the data centers. And it used to be grown at 20% per year. And then it had a period of about a year and a half where the growth stopped. And the share price went down, if I remember correctly, by 60 or 70%.

And the reason why the growth stopped was because, not because they didn't have the right product, but because their major customers, the methods of the world, the AWS of the world, will go through the period where they were designing the next generation of data centers. And there was a lull in purchasing of materials that they needed for those data centers. But at the same time, if you read the company transcripts, they said, we are with those customers participating in the design of the next generation of data centers. And I, as a generalist, I can understand the relationship between a customer and a supplier, because I've seen it in many different industries. So one of the things that we're looking for

is we're looking for the companies that make the whole ecosystem better. They take a rent on it, but they make the whole ecosystem better. If the customers benefit, if the suppliers benefit, and you as a company can benefit from kind of spreading that wealth around the world. And then what happened with Arista is that once you went through that law and the purchasing for data centers has restarted, and the growth rate has restarted 20% that actually accelerated recently because of the excitement about machine learning and artificial intelligence, well, the multiples went back to where they were before and even expanded for it.

And so I think it's not the same thing here, but once again, I think this is a company where what I'm betting on is that they have relationships, valuable relationships with thousands of customers that know the company well, that believe that they provide a good quality product. And the salespeople from Akamai have the foot in the door, any relationship. to be able to come in and say, guys, we have the right security product for you as well. We have the right compute product for you as well. We compute as a $200 billion market growing at 15 to 20% per year. What Akamai says is that, like, if we get 2% of that, well, 1%. I get 2%. 1% of that.

That's a 2 billion revenues for them. And we're talking about a company with current revenues. of what about almost 4 billion so like just like from a scale but magnitude of what it could mean like we have a lot of ways like I think the risk reward is in our favor

I hear what you're saying though you know the old if we can only capture 1% of the TAM is tough but I really like what you're saying on the general side because one of the things it was Chris Pabisi I think he was pitching it was AVTR on the podcast and one of the things that I've learned this year that just really stuck with me

as he was talking about, this is a life sciences company. He said, look, if you're, I asked a similar question, why can you as a journalist outperform, they're a life scientist, specialist? All they do is this. And he said, if you're a life science person, you've literally never seen cyclicality before, right? You've never seen the inventory bullwip. Whereas if you're a journalist, you've seen inventory bullwips, not in life sciences,

but you've seen them in memory cards. You've seen them in oil and gas. You've seen it everywhere. you get the industry is going through an industry inventory bullwip and these specialists have no clue how to handle it how to think about it and a journalist can apply that and make money here and what you're saying with that i think it's i think i love the story you just told because it makes sense to me right where you've got this sales force they've got a pretty sticky product they've got their fingers into a lot of different clients and it's a cross-sell now the counter to that is you know as you

know with everything these are competitive spaces Industry is very difficult. Everybody's trying to cross all everything, but you've kind of got some of the proof already where, as you mentioned, they've spent $2.4 billion scaling up the security side, and they've grown it into a $2 billion-ish revenue company growing 20, 20% per year. You want to say anything else on? So in terms of security and kind of going back to the point of 1%, 2%, etc.

I was trying to find data on the global cybersecurity market. I think it was about $30 billion back in 2016. it's around $7, $80 billion right now. If you take a look at the revenues that Akamai has in global, in cybersecurity, they had a 1% market share back in 2016, and they have a 2.3% market share today from those numbers. So when they're talking about getting to 1% market share of compute, I don't think that's a ridiculous statement.

You know, I don't disagree. let me one thing I do wish they did is I wish they gave out organic I know it's difficult when you're cross selling everything but I do wish they gave out a little bit organic because maybe they have at some point but as I was prepping for the podcast you know I went through like the last earnings call and two trip conferences in decent depth and then I kind of flipped through some of the other ones and I went through their disclosures they don't give any organic growth breakout and it's just as a generalist as somebody who's a little skeptical I just really wish they did because they are buying stuff all the time, and not huge deals.

Most of these are Bolton's, even if they're big dollar figures. But I just wish they would give us a little bit more organic. So I could go to listen and say, hey, they're going to grow 20% in security this year, and 19.5% is organic or something. I don't know if you've got better. I think what we talked about earlier in terms of the high multiple to revenue that they're paying for their positions suggest that most of the growth is organic. Very true.

Very true. Yeah. But it still just would be nice to get a little bit more. You know, Sunshine is the best disinfectant. Switch to something a little bit more fun. I've got the quote if you don't have it, but you could see a lot of ways this company could be an AI play, right? A lot of different ways. And I'll just turn it over to you.

If you don't hit the quote that the CEO gave it at a recent conference, I'm going to give it. But I don't know if you know this one or not. It's no big deal. It was just the thing I would have to hear it. So somebody, an analyst asked, hey, is. AI good for you? Like, would you guys be an AI beneficiary? And he said, look, we, we do everything on the edge. If people start wearing goggles, AI goggles all day, it's going to be really gosh darn good for us. And that quote when I read it just made me absolutely crack up. But ignoring that

tongue and cheekness aside, you could see a lot of ways this could benefit of an AI play. It obviously hasn't caught an AI multiple, but AI is hot. A lot of stuff is going into LLM investments, everything else. I just want to ask, do you see any AI upside or do you bake any of that into your thought process? So for the time being, I think the company is making a conscious choice not to invest into it.

And when we're talking about the compute part, so this is one of the areas, major areas of investment for them. It seems like for the time being, they're making choice of not to invest a lot into GPUs. And I think the reason why they're doing not investing a lot into the GPUs

because they figured that that's something that they can get to when the time is right. And so one of the metrics that they're giving on the investment, on the compute side, is that a dollar of revenue growth requires $1 of complex. The returns on it actually pretty good because you have about 80% gross margins, about 40%, net margins. So you're getting good return of that.

but it's one-to-one ratio. If you look at DigitalOcean, which is taken a slightly, so they are a compute provider, and it's an interesting, like if you are looking at Akamaya, I would strongly encourage to look at Cloudflare and to Fastly and to DigitalOcean

just to understand the other companies in a space that do provide a lot of useful information. They're spending a third of their top X this year on GPUs. But what they're spending a third of their top X this year on GPUs. But what they're spending a third of their third of their top X is, are saying is that for $1 of KAPs, they're going to get 50 cents already. Because NVIDIA has a $3 trillion market cap. They can charge whatever they want for the H100 chip or whatever this next generation.

They can charge whatever they want and people will pay it today. And there's going to be other people that will pay for access to those chips through the infrastructure as a service. But those end users right now, very few of them are making money. That's a completely different conversation that we're not going to get to today. So from the point of view of whether that company

that is actually trying to build an LLM will be able to make money and will continue to pay the amount of money that's required to the digital ocean or to AWS to access those chips, to access that capability, that's a question.

But you can always build that capability a year or two down the road. Yes, you're not going to get the revenue today, but you're not getting the returns today either. So I don't think that Akamai is missing anything by not having that infrastructure today. Because they already have the servers and they have the customer relationships, I think they're going to invest in when the time is right and when the returns away. This episode is sponsored by our friends at YCharts. With all the various job functions that advisors are tasked with, your time is extremely valuable and often scarce. Y-charts is a platform centered around efficiency and built with speed and mind. With an

intuitive and user-friendly interface, Y-charts helps saves advisors 29 hours per month while uncovering better and new investment ideas. Need a way to help clients visualize their financial future, reinforce the importance of consistent investing, and guide them toward informed decisions. Tools for scenario building, portfolio construction, and proposal generation can be the missing piece to your service. So you can act on any deal right when the light bulb flits on. Click the link in the show notes to start your Y-charts trial for free and see how Y-charch is a one-stop shop for growing AUM with fewer hours spent on investment management. You know, I am not familiar enough with their assets to say it, but I will just throw it out

to you and you can, word smith, correct anything I miss. But the other thing is there is this enormous, enormous rush for data centers that can power with power access. And I would have to imagine if the crunch gets bad enough, you know, you're looking at Akamai with a billion-dollar cloud compute business. If the crunch got bad enough, even if they had absolutely zero AI exposure, I just think buying all that server and all that access to power alone would make them kind of an AI play if that makes any sense whatsoever. It makes sense. And I think what you're identifying is different ways that we can win with this investment. the our starting point and kind of going back to what I was talking about in the beginning where kind of how do we look at companies which is what do we pay for it our starting point is

a 6% free cash flow yield yes that includes some of the money obviously that's free cash flow from operations so that includes the money that they're going to you have to offset the comp but that our starting point is 6% free cash flow yield I think we have a company that is growing at high single digits, low double digits for a foreseeable future, even without the things that you just talked about. I think we're okay even without multiple expansion. We should be able to get returns in mid-teens without multiple expansion, assuming multiple doesn't shrink and assuming the company delivers. Two big assumptions, but it is what it is. But this company is trading at half of the multiple and not less of the companies that do have the AI

heaver to it or the cybersecurity or whatever it is. So if we have that multiple expansion, which is actually what happened with Arista networks, you get very different return. Let me ask you one last question here. And then I unfortunately have a hard stop at 10, so we'll probably have to get close to wrapping up. But I think you've detailed all the ways the sales force synergies and everything we talked about why these these three different segments belongs together. But another question is, do these segments belong together? Right? Because I do hear the Salesforce synergies and everything, but I've got a high growth cloud compute business, a high growth security business. You know, I know like Amazon offers, Microsoft offers

the combination, but most players start kind of going to the, it's such fast growth and it's so specialized. Most players are kind of focused on just one area. And, even ignoring those two to the delivery business, which is a, you know, it should be, it should probably be run as a cash cow. Marrying those three is a unique combination. I could see the synergies there, but I could also come to you and say, Steve, like, we've got one business that's about 40% of our earnings. That should be a cash cow.

We should be levering that up and we should just be returning capital. We've got the security business, a lot of stock comp, a lot of highly paid engineers. We should be, you know, pedal to the metal, going for it, stock for stock. M&A, all sorts of things, cloud compute, kind of similar. Do these businesses belong together in your mind? In my mind, they do. Because what you are providing, especially when Edge, so the delivery and compute, you can combine those two with an idea of infrastructure as a service.

What you're providing in both cases is internet infrastructure for people that need it. It could be just infrastructure to deliver. It could be infrastructure for you to be able to be. do scale computing on a size that you cannot do on your own location, but it is the same to me and also has a similar customer base, as we have discussed. Security, once you take into account that you already have the edge, so the delivery to the edge, security is super important there, because that's where the attacks happen. The attacks happen on the edge.

So if you can do the DDoS, you can start the DDoS attack by using that network, you have, and by spreading the DDoSotype, that's exactly what it's there for. Web authentication is much better on the edge than it is in the middle, because you just, even if somebody gets through, you have another layer of security there. So from that point of view, and I think from a customer point of view, it actually makes sense to be talking to the same company, if they, and that's a big it, if they can provide a good enough service or the best service in each one of these things, Why wouldn't you buy it from them?

If they don't, and that's the problem with a lot of these large companies that are trying to do everything, is that if they cannot provide a good enough service, then they have no business of being there. But looking at how effective the sales force has been across selling these services, it seems like they are, it seems like they are providing value for the customers. Perfect. Again, I have a hard stop at 10, so we're going to have to cover it. I'm impressed with how much we got through.

I'm looking through my notes. I think we got through everything. So I just want to turn it over to you before we wrap it up. I think we did a really nice job of covering everything, but is there anything you wish we had kind of talked about that we didn't manage to get into? No, I think we covered quite a bit, and I really appreciate your questions. And you're doing this podcast, I think we as an investor community benefit from the questions that you ask and from the light that you're shining on the companies.

I really appreciate the opportunity. Happy to follow up with anyone who's interested to talk about this or any other. other things, either in U.S. or in Eastern Europe. I know you posted this on Twitter, and you immediately got a question about Bank of Georgia, so it's a completely different conversation. We're not going to get to it within five minutes. But I'm on Twitter or X, whatever you want to call it. We're at our website is www.fbIRD.com.

So if you're always happy to reach us through that. And thank you so much for the time. No, I'm just laughing because, you know, when I announced these on Twitter slash X and I tag you, and you're just a blank face on Twitter. And then, as you said, the first response is Bank of Georgia. So I try not to check Twitter too much, though. I end up checking it quite a bit. And I'm like, Bank of Georgia?

What? Am I doing a bank podcast? Am I just like in such a haze? I can't remember everything. But this was great. Fbird.com. If you want to follow up with Steve, Steve, this has been awesome. And the unfortunate thing for you is I've got your Q1 letter and there's two companies.

I'm interested in there. So we'll have to, we'll have to do a follow up. in the near future but I really appreciate you coming on and we will uh we will chat soon sounds good thank you andrew a quick disclaimer nothing on this podcast should be considered an 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 advisor thanks