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

Okay. I think we're live. Welcome back, everybody. I'm still Brad Zelnick, software analyst at Crédit Suisse. And today is Day 2 of the annual CS Tech Conference, typically in sunny Scottsdale, today, coming to you live virtually in your office, in your living room, wherever you might be. This particular session, delighted to be joined by the good folks over at Akamai. And I think we have with us today -- we've got 2 representatives from the company. We've got Ed McGowan, Enterprise Vice President, Chief Financial Officer; as well as Adam Karon, EVP and GM of the Media and Carrier division. Gentlemen, to both of you, thank you so much for making the time and joining us today for this year's conference.

Thanks, Brad.

Thanks, Brad. Good to be here.

Awesome. So just for logistics, fireside chat format. We've got a number of topics to go over. I'll try to keep my eyes on my e-mail if anybody who's listening wants me to weave in a question into the conversation. I will absolutely do my best to do that for you. But without further ado, maybe we jump right in. And I think the question that you're probably being asked over and over again, but I know that folks really want to tune into is just thinking about the impact of COVID on the business. So maybe to start, one for both of you. This year has brought a number of challenges and among them, is making sure the Internet didn't break as the entire world move to working, schooling from home, doing mostly everything from home. So thank you for that. But if I listen to your comments and your -- and I look at your most recent reported results, it seems like Akamai was prepared to handle the surge in traffic, which is great. But can you maybe walk us through the trends that you saw going into what I hope has been the worst of COVID behind us? And how you're able to handle it? And maybe for Ed, as traffic spiked, revenue accelerated from 8% year-on-year growth to about 13%. As we think about where to from here, how should investors think about the rate of traffic growth from elevated levels? And how that might impact your growth going forward?

All right. I'll start, and you can thank Adam for the Internet not breaking. It was his preparedness in getting us ready for all this surge in traffic. But it is funny....

His back seems strong. It looks like he was holding up quite a bit.

Yes. And I'm sure he'll get into a little bit in terms of the way that we're architected why we could handle the surge. If it wasn't for us, there'd be some pretty upset gamers and OTT video subscribers that would have had a very boring few months. But anyway, in terms of the traffic and the trends, I would say what, I guess, was somewhat surprising to us, not so much the surge. We knew that if people were all sitting at home and not having a lot to do, there'd be more gaming traffic, more video traffic, certainly more e-commerce traffic. What we thought, though, was come the summer, you'd see a big slowdown, but there was an artificial bump and not a new norm, but this really was an acceleration of a trend. And we're seeing these elevated traffic levels through our seasonally weak Q3 and starting off here in Q4, strong traffic as well. So really, the trend is continuing. And I'd say there's probably 3 trends that -- or let's say, 4, that really, I'd say, got exaggerated and are continuing here. Obviously, one is commerce. I saw there's some pretty interesting retail statistics that came out from Adobe just over the weekend for Cyber Monday, Black Friday, et cetera. Banking is another one. We're seeing an increase in traffic on our banking sites, just can't go to a branch, you got to do more online banking. I think people are going to continue to do that. And then within in Adam's world, within media, you've got OTT subscriptions going up and then gaming. Gaming seems to be -- and now with the new console releases, just a continuing trend. So how does that impact growth next year? Obviously, traffic, I think, will stay at an elevated level at sort of this new norm. Growth percentages become a bit of a tougher compare in Q2, that's when we started to really see the benefit of COVID. But I do expect these trends to continue. We did have 2 pretty challenged verticals that make up about 20% of our business travel and hospitality. Don't expect to see that recover anytime soon, probably another year or 2 before we really see recovery there. And then retail, it's really been a question of the haves and have-nots. So a bit of a mixed bag there. That will probably be a little bumpy as we kind of work that out in the first half of next year as well.

Got it.

Brad, just to jump in, just add to that. I think one thing to just chime in, in terms of the scale and being able to handle the traffic, I think it's just another credit, I think, to the folks who design the Akamai network originally. So I think the idea of us being able to localize traffic and our close-knit relationships with our carrier partners and ISPs gave us the ability to kind of distribute that traffic in a unique way and kind of scale with the -- with those carrier partners who have to kind of hold some of that traffic locally in different places all around the globe. So it's kind of a unique architecture and then the ability for us to do some of the interesting things on the edge to help people kind of moderate the traffic. So whether it's rate-limiting game downloads or doing some of the edge compute capabilities we have on our edge, things like our waiting room application, so that folks can kind of be held off before they kind of can engage with a platform until the traffic level subsides to a level which they can be handled.

Got it. That's very helpful. Maybe if I could ask you a question about 0 overages. I think it was back in June of 2019, I believe Akamai introduced this sort of packaging for Web Division customers and I think investors may be a little less familiar with exactly how it works. If you could maybe explain how that plan works and perhaps even what percentage of Web Division customers are currently on it?

Sure. So you're right. It is something we introduced at our customer conference in June of '19, and it was really in response to customers primarily in the commerce vertical that have many seasons. Some retailers could have 10 seasons, Father's Day, Christmas, graduations, you name it. And having that bursting bill from time to time was a challenge and they wanted to basically flatten out their spend throughout the year. So we came up with this concept of 0 overage, which essentially gives our customers the ability to burst the 2x to what they normally would push on the network, which pretty much covers most every holiday season, gives them a more predictable spend. It does flatten out our revenue a little bit. So I won't see as much bursting as I talked about on the Q3 call now that we've been in the market for about 18 months. The primary buyer has been the commerce customer, and about 1/4 of our commerce customers have taken the offering so far. I expect that to continue to tick up as we go next year. We don't offer that in media for obvious reasons. That's where you don't have a ton of traffic on the website. We offered it in media, it probably wouldn't be the smartest thing from a margin perspective, but -- been a pretty good response from our customers as a result of doing that.

Any directional sense or have you said in the past, and maybe I missed it, commerce as a percentage of Web Division or any way of sizing commerce to the...

Sure. Yes. So yes, on the Q2 call, I called out commerce was 16% of total company revenue, about 900 customers. So it's a -- think of Web Division is a little over half, so call it, 30% roughly of our Web Division business.

Got it. But I imagine there's extreme seasonality. So 16%, that's not an annualized, that was a Q2-specific comment?

That was more of an annualized number. Yes.

Okay. Very helpful. And maybe just with seasonality in mind, you've got the spike in traffic during COVID-19. And I'm sure customers were thankful -- at least commerce customers for this shift. Can you talk to us about the impact of the pricing change since it was implemented. And I mean, you talked about it maybe flattening out some of the spikes that you would see in fighting out revenue. What's the balance of perhaps giving up some upside versus winning over customers with that predictability that you can offer them?

Yes. So I'd say so far the returns are early because our typical contract is 1 to 2 years with an average of about 18 months. Sometimes we'll go 3 years, but typically 1 to 2. So we're just starting to see the renewal cycle. The goal with the renewal cycle is to change the paradigm a bit. Typically, the conversation is, hey, I've got 2-year-old rates, so I want a lower unit rate, my traffic is up, et cetera. Here, it's -- your -- basically, your bill has been flat for 2 years or 18 months and your traffic has risen. So it's a different conversation. It's not about X percent off of last rate. It's about what is a fair renewal if your traffic has grown 30%, maybe your bill goes up by X percent rather than have that disruptive drop in rates and then grow your way into it. So we've seen pretty -- early returns have been pretty good. Customers have been happy with it. Obviously, anything in an environment where you're in cost-cutting mode and certainly a lot of our retail customers are, anything that is more predictable, is better than not. So, so far, the feedback has been very positive.

Cool. And Adam, maybe this is a good segue into media pricing. Can you help maybe give us an overview of how typical large media customers contract with Akamai? Maybe, say, your largest OTT customers, how do those contracts get structured? Is there a standard? Do you have breakpoints, overages? How should we think about that?

Yes. I mean, I think it runs the gamut, Brad. But I think generally speaking, you think about it in terms of -- and Ed kind of talked about length of contract. I think in the media space, you think about it more in the 12-month kind of range. And then you think about it in terms of more of a tax stable type race in terms of [ gradiated ] with volume, but also typically tied to geography. So I think we typically try to tie either geographical-based pricing or in many cases, and I've asked a lot today about this as well, which is tying together the delivery of traffic in easier to deliver regions like the U.S. with delivery outside the U.S. So kind of ensuring that if we get the lion's share of international delivery that we also get a piece of domestic delivery as well. So kind of all those things kind of tied together.

And maybe just in light of what we're seeing in this step function increase in volume. How should we think about pricing on renewal going into next calendar year? And I mean, as financial analysts, I know you know all the complexities and the exceptions, but we like rules of thumb. So if we think about just to benchmark to, say, media pricing on average compresses 20% year-on-year as traffic grows 30% or so plus. Do you expect to have to give a little more on price than you ordinarily would just given the step up? Or how do you see those conversations playing out between committed volumes and pricing?

Yes. I don't think the pricing dynamic has changed or is going to change as we head into next year, very similar renewals that -- in terms of discussions we have with customers, there's no difference because of the increased volume. Matter of fact, the interesting thing would be that some of these larger players are now attempting to enter emerging markets where they expect and understand that price points typically go higher. So the mix of traffic kind of ensures that the decline is what you'd expect in a standard way. So I think your benchmark that you already have, I think, is the right benchmark to continue to think about.

Got it. And is there any one competitor out there that's worth calling out that's really leading the pricing curve down at this point and perhaps being irrationally aggressive or is it -- is there no real change in what you've seen in years past?

I mean nobody has been irrationally trying to drive price down. I think a few years ago, I think we saw some irrational pricing, let's say, 3 or 4 years ago, but things have stabilized somewhat in the media marketplace in terms of the standard kind of declines that you'd see. So nobody, in particular, is attempting to drive price down in irrational way yet. We'll see what happens when some folks that maybe aren't generating, let's say, the gains that they'd like to see might feel like they have to do that. But today, we're not seeing that.

Well, it's good to hear. And it's obviously good in the context of a strong demand environment. Just maybe turning to a comment that Tom had pointed out on your last earnings call. I think he talked about Akamai's advantage in terms of cost of delivery. How important and maybe differentiated are your strategic relationships with carriers in achieving that type of cost advantage? Or is it simply that you have enough scale where peering relationships are a natural function of traffic flows?

Yes. I always like to think about it that we have an interesting relationship with those carriers. We have a 3-pronged relationship, right? We sell things to them in terms of security services that they can then white label and generate profit and revenue and services around. We buy from them -- the supply you're asking about in this question. And then we sell-through them, may become resellers of other solutions like our standard delivery solutions. And so we have a very strategic relationship with many carriers all over the globe and it provides us unique access to real estate that others don't get access to it. So I think that's one aspect. And then the scale component gives us that pricing access to them as well. So I think you kind of get both dynamics. There's a much more strategic relationship than just let's buy some capacity. It's a broader business relationship we have with those carriers. And I think that gives us both pricing and scale inside of that real estate.

There's a lot you can do with scale, too, in terms of -- in some cases, we don't need transit providers. We have our own backbone connecting major cities because you have enough traffic to make that economically viable. The amount that we pour into research and development to get our machines to be more efficient is another advantage that we have that we can always get more and more out of our machines every year. So there's a lot of different things, but the relationships with carriers are critical.

Excellent. Maybe, Adam, if I could move on to large streaming events. As we think about these events, let's take Warner Bros' recent announcement that they're going to release Wonder Woman '84 on HBO Max, the same day that it's released in theaters, for example. What are Akamai's competitive advantages in helping to serve that traffic? And what percentage of traffic flows would you expect to get on average for an event like that?

Yes. I think strategic advantage for us is always kind of goes to the core principles there, which will be that you're providing the best performance and we have the scale available for them to be able to burst into. And of course, with many of these customers, we have long-standing relationships and years of delivering for them over and over again. So they lead into us for critical events that they might have like their launch, in particular. In terms of something like Wonder Woman '84, a VOD event tends to not be quite as big as a live event because people don't have to tune it at the same time to watch it simultaneously, they'll watch it over many months. So you'd see a increase, I think, but not a single pivot or inflection like over a period of 2 hours where millions watch. You might see millions watched over a week or 2 weeks in this particular case, potentially. So I don't think you're going to see this exist as an event online, like you kind of couched the question.

There is a fear factor, though. So like we saw very recently with a live event with a number of splitters, they were anticipating big demand. So there's a fear of demand. Oftentimes, they'll shift the traffic to us. So depending on what the marketing department inside of HBO is thinking and how big the event will be to the extent that it's something that they believe will be much larger than normal. Our phone rings and they -- we tend to get a much bigger share because the other folks don't have the capacity. And you don't want to have a bad experience and have to try to switch your way out of it before the event.

Right. Got it. And Ed, I know in the past, you've said one event doesn't really move the needle for Akamai. But -- and I think there's all sorts of conventional wisdom in terms of looking at even years versus odd years, Olympics, elections. How should investors think about the potential impact of the model on these types of blockbuster announcements?

Yes, I think there's 2 ways to think about it. One is the longer the duration, the better. So a 1-hour event, you're just not going to generate much revenue because you're billing by the bits that go across the network. If it's several months and it drives demand or more subscriptions, new users coming online, that can start to move the needle a bit. You can get a couple of million bucks in a quarter if it's a prolonged event and you're getting greater share. I think the thing to watch, though, with that particular event is this notion of releasing same day. And does that generate a model that others are going to follow? And if that's the case, then that would have a much better impact on the business because it would become more of a standard, and you'd see maybe a half a dozen releases for major studio every year, and that would be a decent amount of traffic. And again, it's that fear factor of what if I get a massive crowd, I can't risk my brand. I'm going to go more with Akamai because they can guarantee me the traffic. In that particular case, sometimes, we can even get reservation fees and say, we'll guarantee you with a certain amount of traffic that no matter what, come hell or high water, that's yours. So you want 10 terabits, you got it. And for that, there's a fee to do it.

Got it. If I could pivot to a different topic of DIY and maybe somewhat direct here, Adam, our industry conversations suggest that Disney has been building out their own CDN. What advantages in delivery does Akamai bring to the table that would preclude Disney for moving the majority of their traffic, at least, maybe domestically to their own network. And with that, Ed, your business is pretty well diversified at this point. Is there any 1 customer that would have a discernible impact if they were to move off Akamai similar to the DIY cycle that we went through several years ago.

So I guess I'll start, Ed, and I'll pass it over to you. But I think in terms of any 1 customer looking at DIY, so any large provider, I think they -- somebody like a Disney will clearly look at DIY for easy delivery like domestic delivery. And they'll do parts of it for that. And I think the advantage is that a third-party CDN brings to the party like an Akamai -- particularly Akamai, would be the scale that they're not going to want to build to, first of all, to take all the traffic that they might have. They want to build a portion of it, especially a nontechnology provider in contrast to something like an Apple or Microsoft who might build very be focused and have investments specifically to build technology. A content provider may not want to do that. I think the second part is investment in engineering to keep up with industry standards in terms of new delivery protocols that might be required or they might want to provide on the Internet, and then security services that you need to provide around that delivery. So they would look at it probably in the case of a Disney or any other large OTT platform as -- think about it as an augment to their total delivery platform solution. So I would look at it as a third-party CDN competitor being introduced to the mix that we would compete against with features, price and scale and performance. And they would run a hybrid solution, and it would be a component of their delivery, but not the vast majority as kind of you couched in the question. And Ed, I'll pass it over to you kind of what the rest of the -- other color you want to add there?

No. So I mean, I think just look at what's happened with the big Internet platforms over the years, stabilization and now growth. And as Adam described that we're part of or an extension of what they're doing themselves. I'd expect anyone that's going to do it, especially in the broadcast field where redundancy is a key tenant of what you do, having multiple pass, places that are harder to reach. And that can even be domestically, you can have some carriers that are holding the line on price or you might not be able to get as good access to their end users that you still would want to have another CDN in the mix. But in terms of diversification, you're right, Brad, our revenue diversification is fantastic. For our business, a 1% to 2% customer is a really good-sized customer. So to think if anyone were to go DIY, if you lose 1%, 2%, something like that, not the end of the world, certainly not pleasant, but just not going to put a huge impact on the business, given our diversification.

Got it. Maybe, Ed, if we could shift to the supply side of the equation. How should we generally think about the supply side? Capital intensity has been running in the mid-20s range this year. And across the industry, it seems like supply was ramped despite some supply chain issues. How should we think about, A, the level of utilization of your network? And B, the needed capacity additions into next year?

Yes. So I would say, if you asked me this question back in April, I'd say the utilization of the network was probably higher than we would have liked. And the decision that we made myself and Adam and Tom is we said, "Okay, we don't know what the pandemic is going to bring." So let's lean in and aggressively go after a build-out and see what happens. If it -- if we see another second wave, we'll be ready for it. If we don't, we'll grow into it. And based on what we're seeing now and what I've said on prior calls, I think next year, we'll probably be back in more of a normal, what I'd call, network CapEx, which is in the a 7% to 10% range. We're running about 15% this year. So you've got another 7% roughly of cap software, another 1%-ish of your kind of, call it, your IT or internal office kind of nonnetwork type CapEx. So you're back in sort of that 16% to 18% range, down from the mid-20s. That's my expectation.

Okay. Cool. With -- I think we've only got like 5 or so minutes left. So I've got to move on to security. And apologies, we haven't gotten there sooner. But maybe to start with Adam. If we could talk about security within the carrier side of the business. Historically, I think as of your last Analyst Day, at least, security was a small part of the Media and Carrier division. And more recently, it seems like more and more security deals are coming through this segment of your customers. Is that fair? And have you made any changes to how you bundle media services with security? I know there were changes back when Ed was in the role, but is there anything that's happened different since then?

I think when Ed and I came into the group to kind of redesign it back in 2017, Ed really deserves all the credit for this, really altering the way the sales force went to market. And that's really been the same kind of direction in the last few years as we ramp up the training on the sales force, including enterprise security as well as web security. But we didn't change any of the bundle specifically. We really just focus in on driving compensation models around the sales force as well as training around the sales force to kind of identify opportunities and bring them down quickly inside of our OTT gaming and gambling and publishing customer base.

Got it. And how important is the carrier channel to the security strategy? I mean, how many deals or what percentage of revenue does the channel influence at this point?

And so our channel business is a little over 1/3 in total, and the carriers are the biggest partners that we have. I think we've got a pretty exciting model right now for SMBs with the carriers for what we call our SPS or personalization services for -- think about that as safe surfing at home or SMB, providing a level of basic security and essentially controlling what people can do when you're on the local Wi-Fi. That's a business that I love the margins there. We basically provide the carriers with the capabilities, they take it to market, white label it, sell it down market. We don't go after that market. It's really not profitable for a company of our size to go after that market. And that's been scaling very nicely. So huge margins, value-add to the carrier. And then Asavie, really fantastic acquisition there. Our first kind of dipping the toe in the water there with 5G security and also device management. So seeing some pretty good early traction there. Again, it's a fantastic model carrier, takes it to market. They're bundling it in as they're selling connectivity packages, think of, say, the CARES Act, getting computers in the hands of kids that don't have Wi-Fi access. You now have access over the 5G, 4G network. But with that, you need to be able to control that device, lock that device, so it doesn't -- you don't go serve anywhere you want, you need to protect it. We can provide that service. It's a per seat model, per user, per dollar, per month. So it's a great model. Value-add sell on for the carriers, opens up the opportunity for some interesting edge computing opportunities. Think about the billions of connected devices that are out there, carriers driving interest for doing, say, polling sensors and figuring out what's going on with all those different devices, they all need to be secured, et cetera. So the carriers are really critical to us. They've shown they've done a really nice job selling other enterprise security products, getting them ramped up, selling our other enterprise security products. So I'd say it's probably the most critical channel that we have.

Very insightful. And Ed, maybe just to put a bow around it around security just because we're running out of time. I mean, the growth has been extremely durable and very impressive. And some of that comes from M&A, but the organic execution has been outstanding. How should investors think about the organic growth of this business going forward and your appetite to do more deals to add to the security portfolio?

Yes. So the appetite is definitely there. I mean, we're looking constantly. Valuations are very stretched. It's hard to make things work. I mean, just given the size that we are, you buy a company that might be growing at 20%, 30% that's under $100 million in revenue, that growth sort of gets lost in a $3 billion, $3.5 billion company. So it's really -- you got to look at the longer-term and say, things like Page Integrity, for example, new product, it's a recurring revenue stream. It'll ramp quickly like Bot Manager does, doesn't happen overnight. I think investors probably aren't giving us enough credit for the potential with enterprise security. There's a much, much bigger market. This whole SASE framework that people talk about. I think Gartner says by 2024, over 40% of IT professionals will have some SASE plan, where today it's single-digit percentages. Zscaler has done a really nice job of sort of showing kind of a path of what we're doing and the way where the market is going, that I think if we can get some traction there over the next several years, that could be a significant growth engine for us. And hopefully, in 5 or so years, the company is over 50% security. Certainly, I can -- I don't know what year it will be. But looking into the future, it's very possible that we're sitting down here and doing a fireside chat, talking about over half of our business being coming from security.

Got it. Listen, we're out of time. But what we've accomplished here, I think, is golden, and I truly appreciate everything that you've shared with us. Maybe just to wrap it up. Either Ed or Adam, is there anything you want to leave us with as we close out?

Ed, maybe I'll let you do that.

Look, I think I'm -- I would just say I'm really proud of how the company rallied around our goal to get to 30% operating margins. We delivered 31% this year, able to get top line growth back to 10%. Phenomenal year in media, phenomenal year in security, overcame some challenges. So really proud of the way the company kind of rallied around this and the employees has done a fantastic job. Innovations, the way we're measuring just productive hours of engineering and coding has gone up, which is great, and our employees have responded very favorably to surveys. So I think coming through this pandemic as well as we have, I'm very happy. I think we're positioned very well for the future.

Awesome. Well, with that, gentlemen, thank you so much again. Great to see you, hopefully, soon, we get to see you in person, but that's a wrap and really appreciate your time today.

Thanks, Brad. Really appreciate it.

Thanks, Brad.

Yes. Cheers.