Akamai Technologies, Inc. (AKAM) Earnings Call Transcript & Summary

All right. Good afternoon, everyone. Thanks for joining. Welcome back. My name is Rishi Jaluria. I cover software here at RBC. I'm delighted to have Ed McGowan, who is the CFO; and Tom Barth, who is the Head of Investor Relations at Akamai. Ed, Tom, thank you so much for joining us today.

Rishi, good to see you. Thank you for having us.

All right. And for any investors who have questions for Ed and Tom, please use the box, or e-mail me directly, rishirjaluriarbccm.com.

So I want to start by just thinking high level, right, about really, like Akamai was, in many ways, the one who invented this market. I always make the claim Akamai was really the first cloud company out there before cloud was a thing. But can you talk through the evolution of Akamai from the founding to where it is today? Because there's been a lot and a lot of changes in the business.

Yes. Sure. I mean going back to the early days, when Tom first founded the company, the basic premise was that the Internet wasn't designed for performance, right? The basic routing protocols for the Internet don't take performance into consideration. And Tom was a professor at the time at MIT and Danny, his student, set out to solve that problem. And through a series of algorithms and a theory of putting servers that are close to where the end users are and the edge of the network could actually solve that problem. And we've built a $3-plus billion business over the years by doing so. And the basic theory was that as the Internet grows and as people start doing higher bandwidth things, like watching video or downloading software, the pipes are going to get more congested. So Tom and Danny set off on this journey to start working with the carriers and putting servers in their networks. And sure enough, over time, that became a thing. It became -- it created the CDN industry. And so in the early days, it was all about moving files around. Most of our revenue, back in the early days, came from either streaming or from serving software downloads. Most companies used to send out disks way back in the day, and they moved to doing things electronically. And that created a big market for us. And then the theory originally was that, once you had this edge server out in the edge of the network, there are other things you could add on top. So we started off with static files with images, and then we moved to streaming and large file downloads and serving video. And on top of that, we've built a security business that's now growing at 26% last quarter and now over $1.3 billion. So this notion of creating a edge network, edge platform and adding services on top was an idea we've had from the very beginning 20-plus years ago.

All right. Wonderful. That's a helpful overview. Look, the big question I always get from investors on this one is you obviously had an anomalous 2020, particularly on the media side. How do we think about the sustainability of tailwinds that we saw during the pandemic and particularly as it just relates to the heightened traffic volumes?

Yes. It's interesting going into the pandemic. It was as if we saw, in 2 weeks, basically a year's worth of traffic growth. And it was sort of that last couple of weeks of March into April when everyone went officially locked down here in the U.S. And I remember being -- talking to investors in December. And we had just started to see some lockdowns in certain countries. And I said if this ever spreads to Western Europe and the U.S., watch out, it could be a lot of traffic. And sure enough, that's what happened. And what I was expecting to see is a big spike in traffic and then things start to normalize out. Every Q4, we tend to see a phenomenon where traffic will grow. You see new devices come on. There'll be new titles that are issued, whether it's gaming titles or movies or whatnot. And we seem to set a new sort of day-to-day level going into Q1 and out throughout the rest of the year. We actually did that with the pandemic, which was a bit surprising. So we set this new peak, if you will. And now that's our day-to-day traffic level. And I think what happened was, at the same time the pandemic hit, you had a lot of big OTT services launching. You had Disney, HBO and a few others. And you also had people that, for the first time, were banking online, shopping online, maybe buying a subscription to a television service online because of necessity. And we're not seeing that slow down. We're not seeing a big drop off. We're getting back to more normal growth rates, I would say, for the Internet, but we did not see the drop off, which I think, really, the way to think about it is it accelerated a trend, accelerated people watching video online. Gaming became more popular again. So gaming became a bigger part of the revenue. And then obviously, I think the video, the timing of those launches was perfect in terms of getting a consumer who was forced to look for other ways of entertaining themselves while they were sitting inside their house locked down for months.

Yes. Got you. That's helpful. All right. I want to drill down into the security business, which is the really exciting part. Maybe let's start with the acceleration that you saw on the security business. Anything in particular that you would call out that drove that upside?

Yes. I guess I was surprised when I looked that it was the larger franchises that really drove the acceleration. It's the Kona Site Defender web application firewall bot management, even DDoS. We're seeing ransomware attacks, volumetric ransomware attacks again. So some of the older franchises that people always question to me, like how big can that market actually be, continue to grow bigger than most people think? And we -- most of our new customer acquisition is through security now. I talked about having 67% of our customers now have a security product, 34% have 2 security products. So the team is doing a really good job of selling security into the base. Probably the fastest-growing area for security right now is in our media vertical. The early adopters were commerce and banking. And a couple of years ago, we set off on a journey to have our media sales team focused on selling security. That's becoming just a regular way of business for us. So I'd say those were really the things that just jumped off the tables, selling to more new customers as well as just continuation of penetration of the installed base and more customers adopting more solutions from us.

Got it. So one big question I get from investors a lot on the security side is it seems like your traditional or, I guess, bigger franchise, as you said, at motion was going after a lot of legacy players, right, getting the on-premise hardware-based security systems into cloud-based. And you've been phenomenally successful with WAF and DDoS and Bot Manager. Now it's on the newer products that you are going head-to-head, not just against hardware-based on-premise guys, but against the next-gen security vendors, right? You are, in many ways, going against Zscaler; Palo Alto, which is increasingly cloud native with their solutions. How does the competitive dynamic change? And how does the go-to-market motion changed for you as well?

Yes. Good question. So you're right, we are -- it is a different buyer, typically. Typically, you're selling to an IT buyer instead of the person who is concerned about the website and the motion for our sales force of, if you're selling somebody an OTT service, they obviously care about the site being defamed or being -- coming under attack. And now we're saying, we're protecting all of your public-facing applications and websites. Now we want to protect your employees, right? It's a very different discussion. Our relationship with the CISO has opened up a lot of doors. We've done a really good job for a lot of these companies. If you think about an e-commerce company at this point, online is a significant part of their business, so you build up a good reputation. A lot more banking and trading and that sort of thing is being done online. So that certainly has helped open the door for us. But we do have made some investments in overlay capabilities, people from traditional industry that sold hardware services or hardware solutions to IT buyers that are now moving to cloud-based solutions. From our acquisitions, we've largely maintained the sales force that traditionally sell into this world. But it is a different motion, and it's also a different set of partners. The telcos are one of the partners that we work with, but we're also looking at acquiring new partner capabilities. Guardicore brought on a pretty good sales force, along with a pretty good partner ecosystem, both from a technological perspective in terms of who they integrate and work with as well as who they partner with from a go-to-market perspective. And part of the thought with Guardicore also, if you think about the landscape today, you mentioned Zscaler as an example, they really are sort of best known for Secure Web Gateway. And Guardicore is in that micro segmentation space. There's really no clear leader there. And one of our thoughts with the acquisition of Guardicore is that no matter how good your perimeter is and how good you are trying to stop things from getting in, something is always going to get in. Somebody is going to click on a bad link or a device is going to be infected or someone is not going to have the latest patch on the machine. And that really does -- like I said, no leader in the micro segmentation space -- I think this gives us the opportunity to become the de facto standard. There's nobody out there right now that has a complete SASE or Zero Trust solution. There's piece parts that you have to put together. And part of our theory here is that, Guardicore, we evaluated every vendor in the market and felt like they had the best technology, the easiest to implement. And we think that this can really give us a leadership position. And that will drag along a lot of our other security solutions, whether it's our access product or our Secure Web Gateway.

Got you. That's helpful. And I was about to ask about Guardicore, so you got well ahead of that question. At the Analyst Day earlier this year, right, you kind of gave us these really helpful targets, right, of different areas of security, what sort of growth rates to expect and how that all rolls up into 20%-plus security growth. But I guess what is it that's giving you confidence in those longer-term targets and ultimately, this idea of being able to sustain 20% organic security growth, right? Because as that grows as a mix shift, it seems like the overall business could accelerate from here, which I think all investors would love to see.

Yes. I mean, just to be clear, what we said at the Analyst Day is that we thought we could grow 20% for the next 3 to 5 years, and that would include acquisition. And today, we obviously overachieved that, last year and this year. If you look at the last 2 acquisitions we've done, Asavie was about 3 points of inorganic growth. The guidance that I provided for Guardicore for next year will add about 2 points, 2.5 points, something in that range. So we're not adding transformative size acquisitions to do that. We actually had 20%-plus organic growth rate this year. I think -- number one, I think people just underestimate the size of the markets that we're in today on the web security side. And we're continuing to innovate. We came out with Page Integrity Manager. We're coming out with Account Protector, which, think about bot management has solved the fraud problem of machines that are coming to your system that are trying to gain access and do malicious things, credential stuff, that sort of stuff. And now with Account Protector, it's really trying to understand that, "Hey, it's someone with Rishi's credentials, but it's not him. So how do I identify that it's a fraudulent user?" Opens up a whole new market for us, huge problem that our customers are trying to solve. So the continued innovation, just on the web security side, gives us some level of confidence there. If you look at a product like Asavie, very early days there, you look at where the telcos are from a 5G perspective. The neat thing about that product, we don't have to bring that to market. The carriers bring it to market. From an integration standpoint, very simple integration. It's integrated at the SIM card level so that the carrier doesn't have to do a major overhaul of their OSS/BSS systems, and they can take that right to market. So that's really just about landing and expanding with carriers. We've got a lot of carrier relationships. And again, this is something that they're looking to do because for an extra couple of dollars per subscriber, you can have device management and you can have security on the devices, which is a huge upsell and a way for them to monetize the huge investments they're making in 5G. So you got great alignment there, and you can get some scale. And then obviously, with the enterprise security space, huge market, much bigger than anything that we're playing in today. You see some of the growth rates of some of our competitors, which you talked about. We think micro segmentation is very early days. It's going to be critical and stopping ransomware, and that's where our primary focus is. And when we start getting traction there, that's a much bigger market than we're in today. So when you put it all together, the growth and innovation in the web security space, along with some of the investments we're making with the telcos, including Asavie and also SPS, I did talk about, which is sort of think of it as a small-medium business don't have an IT shop, being able to provide some level of security, carriers take that to market for us, so reaching a market that we don't reach today with our direct force. And then obviously, the enterprise side is a significant market.

Got you. I'm glad you brought up 5G. As we think about the opportunity in Edge, it's definitely an area with a lot of buzz right now, especially with just the amount of 5G rolling out there. At the Analyst Day, you sized Edge applications as being a $150 million business last year and expecting 30%-plus growth in that business. How do you think about how, a, how real that opportunity is in Edge, b, how to kind of separate what's truly Edge versus marketing and then see what makes Akamai well-positioned to benefit?

Yes. So we tried to do that in our Analyst Day, where we were getting a lot of pressure where people are saying, "Break out your edge compute. How much is just simple delivery versus other things that you're doing in terms of use cases, whether it's API management or image optimization, personalization." We had rolled out our Vaccine Edge, which essentially was this notion of being able to do compute at the edge for sites that people are rushing to go registering. I mean that has applications for a lot of different things. But running -- basically somebody running code on our network, where you want to have close to the end user at a low cost with very low latency, those type of applications we broke out separately. And as you mentioned, I mentioned on the last call, will be over a $200 million run rate going into next year. What I'm finding is a lot of our customers are coming to us with more and more use cases and saying, we've got low latency applications or we're looking for an alternative. We're spending a ton of money with the traditional clouds. And we have applications that we can run there, but it would be better suited if we were running in more locations. And I'm hearing from a lot of really big customers about this. And I think really the question for us is how far do we want to go down in this journey? How much -- what direction are we going to go in terms of compute? We want to make sure that whatever we do is repeatable and we're not doing just custom solutions for folks. But obviously, the whole Edge compute business is a massive market, compute in general. And just look at the -- how much money is being spent in that arena. But I think what's unique about us is the fact that we have, by far, the biggest distributed platform in the world. And as more and more devices come online and you need to get immediate access, do it at a low cost, low latency-type applications, do that in a secure way, I think we've got an advantage over most of the players in the market.

Got you. That's helpful. I want to maybe turn to the competition right now. You've obviously had no shortage of competitors throughout the past 25-ish years. But now all the ones that we obviously get questions about, the most are some of the next-gen ones that have recently come to market, namely Fastly and Cloudflare. How do you think about the competitive environment from really those 2 in particular?

Okay, sure. Yes. So I'll start with Fastly. I'd say with Fastly, it was interesting. If I were to go back 4 or 5 years ago, I would say I would see them show up from time to time in my commerce vertical or in some of the more traditional web business verticals. They made a pivot probably 2 years ago, at least from what I can tell, where they started showing up in a lot of the big volume media accounts, which makes sense if you're trying to maintain a hefty growth rate. You can grow pretty quickly if you take 10% or 15% share in some of the big video players or some of the big software download players and run your traffic up. The challenge with that is, as you know, you've been around this industry for a long time, there's a lot of competition there. You've got Limelight, EdgeCast, Level 3 that have been around for a long time. And it becomes a price game. It becomes a scale game, tends to be a bit heavier on the CapEx side. I think we've done a pretty good job there of maintaining our share, growing our share. And the media business has obviously turned around. But it is a pretty tough business from a competition perspective. That's primarily where we see Fastly. Cloudflare, we haven't seen very much of. So they don't participate in that market. So I don't see them. If I'm in the lobby at, I'd say, a Disney or another big video player, they're not going to be in the lobby waiting to go in and try to get their share. They have focused primarily on the SMB business. We don't focus on that part of the market. We do have some partners there, too, some of our telco partners, some of our partners that will OEM our solutions, some of the video players and that sort of thing. So we haven't really run into them there. Occasionally, we'll see them in a few accounts, but they have not been a significant source of competition. And as I said on our call, we actually had the lowest churn quarter in 14 years, I think. So it's been over -- maybe since 2014, sorry. So it's been a long time. We haven't -- we don't lose a lot of customers. The challenge that you have in the CDN business, the traditional CDN business, it's all been about price. I'd say in the media space, it has become pretty efficient if you were to run an RFP and you get a quote from me or from Fastly or from Limelight or Level 3, you tend to see prices within a reasonable range in terms of -- there's not a huge disparity anymore. Again, pretty efficient market based on volumes.

Got it. Got it. All right. And then what about the whole DIY, right? How big of a threat do you think it is longer term? We've obviously seen that happen over time and, I think a couple of years ago, kind of ring-fenced it around the Gann fan companies or what you call the Internet platform customers. Is it really those that have the ability to in-house CDN capabilities? Is it other? And maybe taking that 1 step further, is there any worry over time that some of the more CDN security offerings like a WAF, a DDoS can be in-house as well?

Yes. It's a good question. Obviously, that was a big concern going back 7 or 8 years ago. And then you saw the Internet platform or Gann fan, I think, was a term you used, stabilized. They're actually growing. They're actually growing quite rapidly. And before I took the CFO role, I was running the media sales team. And one of the strategies we've put in place was to figure out how do we operate in a world where someone like that, who's a massive scale player that has their own capabilities, how do you find your niche? And how do you sell into that world, whether you're selling security or you're doing different type of applications or finding areas where they're not going to grow their CDNs. And I think the team has just done a phenomenal job there. Obviously, that's been growing pretty well. So I think we know how to operate in the DIY world. Others are capable of doing it, sure. Some of them -- some of the larger players that deliver lots of traffic, some of your big OTT players could look to do that. They all have multi-CDN strategies to begin with, so I don't think it would be a significant issue if they were to go DIY. I don't know that they could necessarily do it cheaper on their own. You got to think about you need to develop a team, build out an Edge and keep the technology up and running and that sort of thing. So it can be costly. And the cost of people go up every year, where we can drive our costs down every year through innovation and whatnot. And also we get -- with the relationship we have with carriers, we get a lot of our bandwidth and co-location and things like that for free. So tough to replicate that. On the security front, I don't know. I wouldn't think that people would want to do that necessarily. You look at the size of the threat research team we have, the pace of innovation, the amount of data that we see, something that you probably wouldn't want to take a risk on. I'd hate to be the IT professional or the person who owns the website that decides to move away from a solution that works and says, "I built it myself," and then got attack. So I don't think there's much of a threat there. And then like I said, with the DIY, it's possible that some of the big video players do it, but I would assume that it would be part of the mix. I don't -- just traditionally, they don't single-source anything. So I would highly doubt that they would single source 100% of their delivery. So I don't see it as a big concern, and it hasn't been much of an issue for us.

Great. Great. All right. I got an investor question in my inbox. I want to pivot to that, which is with Rick McConnell leaving to take over as Dynatrace's CEO, should we expect any changes with Monte taking over his role?

No. I think, look, Monte has got -- first of all, Rick's done a phenomenal job, so I want to thank him for everything that he's done, and he's got a phenomenal opportunity in front of him. I think it just shows you also how committed Tom is to the business that I get a lot of questions about Tom. He's got more energy than any one of us. So he's as engaged as ever. Monte has been around the business for a long time and also was in the CIO seat. So I think he's going to bring a lot of value in terms of his experiences as it relates to us moving into the enterprise security space. But no, I'd say continue the course. And again, Monte is someone who's been in many different roles, very customer focused and has that CIO background. So this could be very helpful.

All right. Wonderful. I want to then turn to the question of a macro thing, which is the supply chain constraints. Obviously, it sounds like you're in fine position now partly with all the capacity built over the OTT launches and then gearing elevated traffic last year. But how are -- have you seen any impact from this even when it comes to trying to procure new equipment? And at what point do you -- could it start to become a problem if it keeps going on for several more quarters or years?

Yes. Great question. Well, what was interesting during the pandemic, one of the exercises that we did was we looked at our supply chain. And obviously, you saw like Asia was the first part of the world that was impacted. So we diversified our supply chain immediately. And then we also ran some game theory and said, look, what if this pandemic were to last for a year, 2 years, et cetera? We should be building out, assuming that we're going to continue to see this elevated traffic, build out a significant amount of inventory and keep that pace of ordering things well in advance. We have several quarters worth of inventory. And if I look at backlog for things like routers and things like that, you could be 200 days or something like that, in some cases, with certain vendors. But we've stayed ahead of that. So what we're doing is we're really just now looking at ordering and trying to keep that inventory that we have at a stable rate so that we're not going to be disrupted. So if we don't need the equipment right away, the good news is we built out for that, and we have inventory to take that into consideration, albeit it was a different scenario that we were game planning for. We weren't predicting what happened here, but we were predicting a disruption in the supply chain. So I think we're ahead of the game there. The only thing I think that could be a potential disruption is if you were to see another massive growth in traffic, right? If we were to see another pandemic-like step-up where you see a year's worth of traffic show up, that would probably be the one thing that could be disruptive, but that obviously would come along with more revenue and that sort of thing. But it would be a problem for the entire industry, really. But I think we -- I give hats off to Tom and the rest of the team for really taking this one seriously and really starting to think, "Okay, this pandemic has changed our thinking. We need to bake it into the way we plan." And obviously, our CapEx was a little bit elevated. I took some knocks for that last year. Told everybody we'd get back to more normal levels. As a matter of fact, my guidance here was for sub-15%. Part of that has to do with the mix changing as well. The security is not as capital intensive as a traditional CDM business.

Yes. Got it. As we look forward to -- I mean, really this quarter, I mean, we're slightly over a week away from Black Friday and Cyber Monday. How are we thinking about the e-commerce season as it plays into Q4, especially, a, with supply chain constraints on the other side, right, not being able to produce enough goods for people to buy; and b, maybe just the tougher comps on e-commerce side, right? I mean everything was e-commerce last year. And now in Q4, we have option for more brick-and-mortar retail and non-e-commerce. How do both of those play to your expectations for e-commerce in Q4?

Yes. Rishi, great question and something that, obviously, I gave a pretty wide range for guidance, and we try to think about different scenarios. I think there's a couple, right? You can -- you could envision a scenario where people go to websites, and there's nothing on the shelf, so there's less shopping. You could also run a scenario that says, there'll actually be more shopping because people will do more searching on the web because they're going to keep looking until they find a site or some goods to find, right? So you could see elevated traffic in that scenario, lower traffic in the other scenario where people are just going to go with a gift card or just give up and not bother, give cash or something as a gift. So there's a lot of different outcomes. The other thing I would just say is that about -- was it the summer of 2019, we introduced our 0 overage offering, which we were hearing from a lot of our commerce customers, in particular. There's 12, 13 different retail seasons throughout the year, and they had traffic that would go up and down, especially in Q4, they wanted to flatten out their bills. So we have about 50% of our customers in commerce that around that. So the variability is not quite as big as it was in the past. It's still there, but it's not as big as in the past. But you're right, this is a harder than normal holiday season to predict for all the things you mentioned. Last year was a pretty strong commerce season because we didn't have the option of going to stores. And then who knows how the supply chain will impact the surfing behavior of folks.

All right. Wonderful. I think that's a great place to wrap up because we're already over time. But Ed, Tom, thank you again so much for joining us, and thank you, everyone, for tuning in.

Thanks so much, Rishi.

Have a nice day, everyone.