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

August 9, 2022

NASDAQ US Information Technology IT Services earnings 60 min

Earnings Call Speaker Segments

Tom Barth

executive
#1

Thanks, operator. Hello, everyone, and I appreciate you being on the call for Akamai's earnings discussion covering the second quarter of 2022. With us today are Tom Leighton, who serves as Akamai's Chief Executive Officer, and Ed McGowan, Akamai's Chief Financial Officer. I want to point out that remarks made today contain forward-looking statements, which include guidance on revenue and earnings. Such forward-looking statements carry risks and uncertainties and depend on various factors that might lead actual outcomes to differ materially from what those statements express or imply. Among these factors are any effects stemming from macroeconomic trends, the integration of any acquisitions, and any consequences of geopolitical developments. More details about these factors appear in Akamai's filings with the SEC, such as our annual report on Form 10-K and quarterly reports on Form 10-Q. The forward-looking statements made during this call reflect Akamai's view as of August 9, 2022. Akamai undertakes no obligation to update these statements to reflect new information, future events, or circumstances, except as required by law. Also, please remember that we will reference certain non-GAAP financial metrics on today's call. A detailed reconciliation of GAAP and non-GAAP metrics is available under the financial portion of the Investor Relations section of akamai.com; and now, I'll hand the call over to Tom.

F. Leighton

executive
#2

Thanks, Tom, and I appreciate everyone being here today. I'm happy to share that Akamai turned in solid performance for the second quarter, even as the global economy remained challenging and Internet traffic growth continued to slow. Revenue for Q2 came in at $903 million, which represents a 6% increase compared with the same period a year ago and a 9% rise in constant currency. The momentum behind our Security and Compute businesses fueled this outcome — combined, those two grew 30% in constant currency. Together, these 2 business lines now represent 54% of total revenue. For Q2, non-GAAP operating margin reached 29%. Non-GAAP EPS for Q2 came to $1.35 per diluted share, a 5% decline from a year earlier, though it edged up 0.5% in constant currency. Ed will get into this later, but EPS took a hit from foreign exchange rates as well as a higher effective tax rate than we saw last year. Free cash flow was quite robust, hitting $223 million in Q2, which equals 25% of revenue. We've been putting our financial strength to work through significant investments in enterprise security and cloud computing. Some of that cash has also gone toward repurchasing more stock. During the first half of the year, we used $268 million to buy back 2.6 million shares. This means we remain on track to go beyond the amount needed to counterbalance the impact of employee equity programs for the year. Next, I'd like to share some thoughts on our three primary business segments: Security, Compute and Delivery, and I'll start with Security. In Q2, our Security Solutions brought in $381 million in revenue, which represents a 17% increase compared to the same period last year, and a 21% rise when measured in constant currency. The expansion in security was driven primarily by our application and API security offerings, which include our industry-leading web app firewall, Bot Manager, Account Protector and Page Integrity Manager products. Our Zero Trust enterprise security offerings, led by Guardicore, also enjoyed a solid Q2, with numerous significant customer wins. One of the biggest suppliers of financial information worldwide, which had ransomware concerns, brought in Guardicore's segmentation offering to join the seven security items it already buys from Akamai. A sizable French insurer turned into a fresh Akamai client after choosing our Guardicore product to assist with meeting European financial rules. That deal, steered by one of our carrier allies, reflects the strong interest our partners are showing in our Zero Trust offerings; over in Australia, Telstra, the biggest telecom operator there, grew its relationship with us by bringing our Secure Web Gateway into its lineup of Akamai offerings. We heard from them that Akamai has been a crucial industry ally throughout Telstra's efforts to provide fit-for-purpose solutions, offering network-based anti-phishing malware protection and content filtering. Every single day, Telstra blocks millions of threats, and Akamai plays a key role in that defense. In total, our Zero Trust offerings generated $43 million in revenue during Q2, representing a 59% increase year-over-year in constant currency. This is a space where we are continuing to make major investments and where we expect substantial future growth. Now let's turn to Compute. I'm glad to report that our Compute product group brought in $106 million of revenue during Q2. That's up 74% year over year, and 78% higher when you strip out currency effects. As a quick refresher, the Compute product group includes Linode, plus Akamai's offerings for Edge computing, storage cloud optimization, and Edge applications. One thing I kept hearing over and over in Q2, whenever I talked with executives around the world, was how increasingly anxious they are about being locked into deals with the big cloud providers, which keep consuming bigger and faster-growing chunks of their IT budgets. They're seeking greater choice in compute and are receptive to alternative clouds such as Linode as a more efficient way to build, run and secure their applications. Because of this, numerous large enterprises have started testing the Linode platform, among them a major U.S. airline, one of the world's top gaming companies and a global provider of weather data. Major media companies especially voiced substantial concerns regarding their expanding use of the giant clouds. Beyond the fact that costs are high — partly due to fees incurred from moving data from cloud storage to a CDN for delivery — they're also worried about their growing dependence on a direct competitor. As an example, I'm pleased to share that a contract was just finalized with a major global media organization, which will operate an important new workload on Linode. That service is already running as of now, and we expect it to bring in several million dollars each year as their consumption increases down the road. What we're seeing here is only a small glimpse of the opportunity ahead for revenue expansion in Compute. The cloud computing sector is expanding quickly and is worth several hundred billion dollars. As we grow Linode into an enterprise-grade offering, with points of presence and hundreds of sites globally, we ought to be well positioned to win a portion of that market—especially among businesses that appreciate our best-in-class security and delivery offerings and prefer not to get tied down to pricier alternatives from a cloud giant that also competes with them. Revenue from our Delivery products came to $417 million in Q2, a decline of 11% compared with the same period last year and a drop of 8% on a constant currency basis. The main factors behind these numbers were a further slowdown in how fast our biggest media clients' traffic is growing, plus the reduced per-unit rates we offered when those clients renewed their contracts recently. Worth keeping in mind: when it comes to delivery, Akamai is still far and away the leader in the market. On top of that, churn among customers during the first six months of this year stayed at all-time low levels, and the yearly revenue we lost from accounts that churned in Q2 was actually smaller than what we lost in Q1. We also had a number of well-known delivery customers bring their business over to us during Q2, having tested out what competitors offer. Regardless of which rival they had been using, they picked Akamai, drawn by our better performance and reliability. As with numerous other businesses, we're navigating a period marked by considerable economic challenges and unpredictability. Inflation keeps climbing, the U.S. dollar has grown stronger, and there are mounting concerns about a global recession, worsening geopolitical tensions and conflicts, and slowing growth in Internet traffic as the world tries to get back to some semblance of normal life during a pandemic. Ed will get into this shortly, but in my view it makes sense to assume that these pressures will hold back both our growth rates and our profitability over the next few quarters — particularly in our Delivery segment, which is dealing with the same problems that a number of major media organizations have recently reported. Given how much room there is for growth in the market, we're going to continue putting our money into Zero Trust enterprise security, which Guardicore leads, and cloud computing, which Linode leads. Those investments, along with the economic headwinds I mentioned earlier, will likely keep our operating margin under 30% for now. As our Security and Compute businesses continue growing and make up a larger share of total revenue, we expect margins to improve over time. On top of that, we're kicking off a number of major efforts focused on lowering costs and improving operational efficiency, which should help get margins back to 30% or higher over the medium to longer term. Ed will get into the details, but basically we're going to use Linode to cut what we pay the hyperscalers for cloud by a lot, slow down the money we're putting into the delivery platform's capital expenditures, and make better use of the real estate we have. There's no denying this is a tough stretch, but I think Akamai is moving the right way when it comes to growing and succeeding over the long term. Our position is a good one thanks to the global-edge platform that sets us apart, the security and delivery solutions that lead the market, the new compute capabilities that give great performance at a fair price, the top-tier enterprise customers we serve, and the strong financial standing and profitable business model we have. At this point I'll turn things over to Ed, who can say more about Q2 and what we expect ahead. Ed?

Edward McGowan

executive
#3

Thank you, Tom. As Tom noted, we had yet another strong quarter at Akamai in Q2. For the second quarter, revenue came in at $903 million, which represents a 6% increase from the prior year, or 9% when measured in constant currency. That top-line figure matched our guidance, with the Security and Compute segments driving the results. During our previous call, we indicated that sizable foreign exchange pressures would weigh on Q2 revenue. The considerably stronger U.S. dollar reduced our year-over-year growth rate by 3 points, translating to roughly $29 million in year-over-year revenue impact and $14 million sequentially. Taken together, the Security and Compute units accounted for 54% of total revenue, expanding 26% from a year earlier and 30% in constant currency. Security brought in $381 million, posting 17% year-over-year growth and 21% in constant currency, supported by ongoing momentum in our Zero Trust offerings, particularly Guardicore. Guardicore generated around $14 million in revenue during Q2. Security made up 42% of overall revenue in Q2, an increase of 4 points versus the same quarter last year. Compute revenue reached $106 million in Q2, up 74% from the prior year and 78% in constant currency. Linode added roughly $32 million in revenue during the second quarter. Delivery revenue totaled $417 million, declining 11% year-over-year and 8% in constant currency; it's important to note that although network traffic kept rising, the pace of that growth slowed from one quarter to the next in Q2. As Tom said, we think the prevailing macroeconomic conditions have weighed most heavily on customers within our media vertical, especially in advertising and gaming. These pressures show up most clearly in our delivery numbers; furthermore, as we discussed at our Analyst Day in May, we began adjusting our pricing approach to reflect the new traffic growth rates observed across our network over the last 2 quarters; beyond reducing discounts offered at renewal, we chose to walk away from certain business involving a very small number of customers whose traffic peaks are extreme relative to their typical daily usage. Although this means somewhat lower revenue, it allows us to bring down network CapEx considerably as we prioritize cash flow from our Delivery segment to support faster-growing areas such as Compute and Security. International sales made up 47% of total revenue in Q2, the same as Q1. International revenue rose 6% year-over-year, or 13% in constant currency. Lastly, U.S. market revenue came to $477 million, growing 6% compared with a year ago. Turning to costs and profitability. Cash gross margin stood at 75%. GAAP gross margin, which accounts for depreciation as well as stock-based compensation, came in at 62%. Non-GAAP cash operating expenses totaled $292 million. Adjusted EBITDA reached $388 million, giving us an adjusted EBITDA margin of 43%. Non-GAAP operating income was $262 million, and non-GAAP operating margin was 29%; note that unfavorable foreign exchange reduced our non-GAAP operating margin by about 1 point. Capital expenditures in Q2, excluding equity compensation and capitalized interest expense, amounted to $104 million. The results came in well ahead of what we had anticipated, in part because certain Linode-related CapEx shifted out of Q2 and into Q3, and in part because we benefited from savings tied to our approach of exercising greater selectivity regarding particular categories of customer traffic. Those savings showed up most clearly in network CapEx outside of Linode, which amounted to just about 3% of revenue. On a GAAP basis, net income for Q2 reached $120 million, equal to $0.74 per diluted share. Non-GAAP net income came to $216 million, or $1.35 per diluted share, which represents a 5% decline versus a year ago and a rise of 1/2 of 1% when measured in constant currency; worth pointing out is that foreign exchange took a toll of roughly $0.07 on our non-GAAP EPS. The tax figure baked into our non-GAAP earnings stood at $42 million, reflecting an effective Q2 tax rate near 16%. That was roughly 1.5 points above the prior year. Turning to cash and how we deploy capital. On June 30, cash, cash equivalents and marketable securities came to roughly $1.3 billion. Through Q2, we used about $165 million for share repurchases, retiring around 1.6 million shares. Thanks to our continued repurchase efforts, non-GAAP fully diluted shares outstanding have shrunk on a net basis by about 4 million shares, or approximately 2% compared with a year ago. We closed Q2 with roughly $1.5 billion left under the repurchase authorization currently in place. Our plan is to keep repurchasing shares over time so as to offset dilution stemming from employee equity programs, while also staying opportunistic when it comes to both M&A and buybacks. Before I get to our Q3 outlook and an update on 2022 guidance, there are several factors I want to call out. First, since close to 50% of revenue is generated outside the U.S., foreign exchange remains a meaningful headwind to the results we report. At current spot rates, our guidance now bakes in a negative $114 million year-over-year hit to 2022 revenue from foreign exchange. As I noted earlier, FX also weighs on our margins and earnings. We estimate that FX will reduce non-GAAP operating margin by about 1 point year-over-year, and non-GAAP earnings by roughly $0.31 across the full year 2022. Second, we have grown somewhat more cautious regarding the traffic growth outlook for Q3 and Q4. Given the trend we have seen year-to-date, along with what other large Internet companies are telling us, we expect traffic growth to be slower than is typical for the rest of 2022. Third, because of that anticipated slowdown in traffic growth and the updated pricing strategy we referenced earlier, we now expect CapEx to land significantly under our prior outlook. Lastly, as Tom noted, we believe the tougher macroeconomic backdrop will put a damper on revenue growth and margins in the near term. Combined with the fact that 8 of our top 10 customers renewed during the first half of the year, we expect Delivery revenue to fall at a somewhat steeper year-over-year rate across the next 2 quarters. Because of those dynamics, we anticipate that margins will likewise shrink in the immediate future; we have strong conviction about the possibilities that lie ahead over the long haul, particularly within Security and Compute, and so we intend to keep putting money into capturing the market chance in those segments. Having said that, as Tom pointed out, we are starting a number of significant programs aimed at cutting expenses and boosting how efficiently we operate. The possible savings tied to every one of these segments sits in the tens of millions of dollars. To be precise, we intend to carry out these 3 actions. Number one, substantially lower what we pay hyperscalers for cloud usage as we shift Akamai's internal workloads over to Linode. Number two, capture savings stemming from our approach of being pickier regarding specific categories of customer traffic, which we anticipate will produce reductions in both capital expenditures and depreciation expense; and number three, make our real estate footprint more efficient now that the hybrid work experience is becoming more lasting; moving on to what we expect for Q3. Our projection for revenue falls between $868 million and $883 million, which is an increase of 1% to 3% as reported, or 5% to 7% on a constant currency basis, compared with Q3 2021.

Edward McGowan

executive
#4

On the FX side, currency movements are dragging Q3 revenue down by about $11 million relative to Q2, and they're costing us $36 million versus the same period last year. With revenue at those levels, cash gross margins ought to land near 74%. In the near term, margin is getting squeezed by what we keep putting into Guardicore, the staff we've added for Linode, and higher bills from third-party cloud vendors. Even so, we see this squeeze as short-lived. Over the long run, we expect margin to move up toward the high 70% target model, powered by continued strong expansion at both Guardicore and Linode as third-party cloud expenses come down over time. Looking ahead to the third quarter, our non-GAAP operating expenses are projected to fall somewhere in the range of $283 million to $291 million. We anticipate EBITDA margins around 41%. Non-GAAP depreciation expense is expected to be between $125 million and $128 million, and for Q3 we're targeting a non-GAAP operating margin of approximately 27%. As mentioned before, the short-term decline in operating margin is driven by a few factors: softer traffic and revenue growth, the annual merit-based wage increase that kicks in on July 1, unfavorable foreign exchange effects, continued investments related to Guardicore and Linode, and rising costs from third-party cloud services. With our Security and Compute segments continuing to expand, and with the cost-reduction measures I mentioned earlier beginning to pay off, we have confidence that operating margins will return to 30% and improve beyond that level. Now, on capital expenditures: for the third quarter, we anticipate spending somewhere in the range of $109 million to $119 million, leaving out equity compensation and capitalized interest. That comes to roughly 13% of the total revenue we're projecting. Based on the revenue and spending figures I just walked through, our Q3 non-GAAP EPS outlook is $1.21 to $1.26. That guidance bakes in $37 million to $38 million in taxes, reflecting an estimated quarterly non-GAAP effective tax rate of roughly 16%, and also assumes a fully diluted share count of about 160 million shares. Turning to the full year, our revenue expectation is now $3.57 billion to $3.61 billion, representing 3% to 4% growth year-over-year as reported, or 6% to 8% growth in constant currency. For full year 2022, we still expect Security growth of approximately 20% in constant currency. Our updated projection for non-GAAP operating margins is roughly 28% to 29%, with non-GAAP diluted earnings per share expected between $5.19 and $5.37. This outlook assumes an effective non-GAAP tax rate near 16% and about 160 million fully diluted shares. For the entire year, we expect capital expenditures to come in around 12% to 13% of revenue. To wrap up, even though the macroeconomic environment has grown less certain, we're confident we're competing in the right sectors with distinctive offerings that our customers continue to prize highly. Thanks so much. Tom and I are ready to answer your questions. Operator?

James Breen

analyst
#5

Could you discuss the fact that the revenue guidance for Q3 is roughly $10 million below consensus? Could you also address the balance between margins and pursuing some of this delivery traffic overall, since EBITDA margins are roughly unchanged, yet clearly top-line growth is decelerating?

Edward McGowan

executive
#6

Jim, it's Ed here; regarding the guide, there are really two components. The first is that FX has clearly worsened incrementally; that accounts for roughly half the delta. The remaining portion is the slower traffic we mentioned. I believe your question focused on the trade-off we're making. The initial direct effect of that trade-off from accepting less of that peak traffic is that you save on CapEx; so that flows through the P&L via reduced depreciation and clearly straight away via cash flow; thus the choice we've made is, as we said in May, that as traffic levels have declined, the discounts offered at renewal are smaller; and in certain instances, we've observed a small number of customers that have the traffic. Their peaks continue to grow at fairly substantial rates, but their everyday traffic does not. So there's a formula we follow in a trade-off we made that I believe makes a great deal of sense for us, particularly when I consider the kinds of returns available from faster-growing areas such as security and compute.

James Breen

analyst
#7

So the overall move is just toward a business that's a bit more asset-light, and putting more attention on a few of these other areas?

Edward McGowan

executive
#8

Yeah. What I'd say is, clearly, compute isn't something I'd describe as asset-heavy, but it does require a decent amount of CapEx; we mentioned that for Linode our CapEx comes to roughly 2% of revenue overall. And clearly, it will be somewhat weighted toward the front end. We actually seized the opportunity to spend a bit more on Linode. Given some of the discussions we're having, we're beginning to see solid demand; and clearly, Tom mentioned that we signed a fairly major customer during the quarter. So what we are doing, though, with Akamai — which of course I refer to as the legacy Akamai delivery business — is that we're being somewhat more selective when it comes to cutting CapEx. I noted that our network CapEx, leaving out Linode, was 3%. Normally that figure is around, let's say, 8% to 10%, at least based on how it's been trending over the past couple of years. So I think it simply comes down to being more selective about particular kinds of traffic. It's really only a small number of customers that generate that kind of traffic pattern; it tends to be event-driven. And what we're finding is that redirecting some of that investment elsewhere just delivers a better return.

James Breen

analyst
#9

James, thanks. And Tom, maybe one thing on Linode—could you share a bit about how things are going when you sell that offering into the customers you already have?

F. Leighton

executive
#10

Sure. Clearly, it's still quite early. There's a great deal left for us to do when it comes to Linode, like we've discussed; even so, it's exciting to bring aboard our first really big client, running a mission-critical application. In fact, it involves transcoding, putting user content onto the Linode platform. Like I said, that single app by itself, once it grows, brings in several million dollars annually in revenue for us. And that's only the beginning, since this client spends many hundreds of millions of dollars each year on cloud computing. And that's only 1 client. So plenty of strong opportunity lies ahead, and our Linode road map is coming along nicely.

James Fish

analyst
#11

We've noticed a few security vendors putting up numbers here, and Security is actually still holding up as a solid area even with the macro backdrop; and looking at your normalized figure of roughly 17%, that comes in a little softer than what we saw in Q1 across the last couple of years really. What's happening there, particularly in the core WAF and DDoS plus Bot management when it comes to sizing and growth versus some of the newer offerings such as EAA and PIM here? Could you help us think about how to balance net new business against the expansion you're seeing from existing customers and security, especially if you're all seeing higher overall net churn?

F. Leighton

executive
#12

I'll start with that, and then Ed might add something. Security is performing very well; our WAF leads the market by a substantial gap. Actually, when you check the most recent Magic Quadrants and analyst ratings, our advantage over rivals has widened. Bot Manager is also unmistakably the top player in its market; with Account Protector now added on top, that's expanding quite well. EAA remains small at this point, but it's a solid growth opportunity, especially as it follows Guardicore. Guardicore is a really exciting one. When you look at the security space for enterprises, segmentation tended to get ignored in the past, largely since it wasn't executed particularly well. Still, looking ahead, it may be the single most critical action a large organization can take. A business could purchase every security offering available on the market right now. Even so, attackers will still manage to get inside somehow. What truly matters is spotting the moment malicious software has breached the perimeter, pinpointing its location, and stopping it from moving any further. So Guardicore stands out as the strongest shield we have against both malware and ransomware. Analysts rank their offering at the top, and while the numbers remain fairly modest, the momentum we're seeing is genuinely encouraging. Then there's Page Integrity Manager — a newer entry with a lot of promise ahead. It's early days, so revenue is still limited, yet it gives businesses a way to meet the latest PCI mandate, which is tough for them to handle alone. Under the updated rules, an organization must put protections in place to stop malicious software that infiltrates the digital supply chain and ends up on the browsers of its users. That's an extremely difficult thing for any enterprise to accomplish. However, when content is being delivered by Akamai, our scripts can be placed on the page, which lets us guarantee a safe experience for the user and ensures that if malware has been picked up by that user at some point in the supply chain, we're able to detect and stop it, keeping the user from being compromised. I'd say our security lineup is performing quite strongly, and for several of our most promising offerings, we're still at the beginning stages with plenty of room to grow ahead.

James Fish

analyst
#13

Understood. And perhaps for Ed, regarding the delivery side, Ed, you mentioned a small number of clients that you basically would prefer not to serve moving forward, specifically concerning the sort of traffic being delivered for them, obviously the peaks compared to the average. Looking at the bigger picture, what observations do you have concerning this fresh pricing approach relative to churn, plus how should we consider this dropping of clients in the second half guidance, and does it truly affect the Security business down the road?

Edward McGowan

executive
#14

Yeah. Excellent question, Jim. Let me make clear that we aren't working with those clients, just sort of pulling back, as I mentioned, it's a small number of clients. And usually, what you observe is these are multi-CDN clients that utilize numerous different providers and generally have significant peaks for events. And how you view those clients is, there's a specific amount of everyday traffic you receive, a specific price point you're prepared to sell at, and there's a specific peak to average ratio you're examining. So as we proceed to renew some of those clients, we're being far more disciplined concerning the — adopting a different strategy regarding the level of discount we're prepared to offer; thus, the client must decide about continuing to use us, but how will they manage those peaks? then going to resolve them over a longer time frame, attempt to locate extra peak in the market, etc.; so they remain very large clients of ours, but we are resisting somewhat regarding how much peak we're prepared to accept on the network. And consequently, we might lose a bit of that everyday traffic, but we remain a major provider. Regarding the effect on security and other products, we haven't observed any effect on that. We're not losing those clients. They're not departing the platform. It's just — we are holding firm more on price, and we might lose a bit of the delivery as a result. But we're also losing plenty of that large peak and that ratio is becoming more aligned with what makes greater economic sense.

Rishi Jaluria

analyst
#15

Great. First off, I'd like to discuss Security. You've clearly spent a lot of time on the Delivery segment, yet this quarter we observed a fairly notable deceleration in Security compared with the prior quarter, even when measured in constant currency. And in constant currency terms, sequentially between Q1 and Q2 you added just under $6 million, during what ought to be a fairly robust environment for security spending, judging at least by what other companies in this area are reporting. Could you help us grasp what's happening within the security business and how we ought to view it moving forward? Then I have a follow-up question.

Edward McGowan

executive
#16

Sure, Rishi, I'll handle that. Bear in mind, as you may recall from the prior quarter, we mentioned that Guardicore revenue included roughly $7 million in onetime license revenue, which pulled forward approximately 1% or so of growth into that quarter. So that had an effect too. Another point to bear in mind as well, if you recall, is that with our < > business across 2020 and 2021, we experienced a fairly significant increase from the education vertical. At that time, kids were working from — sorry, learning from home, attending school from home. And certain government funding allowed people to work or attend school remotely, and that funding no longer exists. So we've noticed a slight decrease in that area. I believe it's those 2 factors combined that produced somewhat softer sequential growth quarter-over-quarter; however, regarding other aspects we observe beneath the surface of our business, demand is very strong for Bot Manager. Account Protector is a newer edition of that product, or an add-on; it's performing quite well, though it remains fairly early. It takes time for that to — it's good for contributing to growth. But those are a few factors to bear in mind when considering the deceleration in growth there.

Rishi Jaluria

analyst
#17

Understood. That is quite useful; and next, I'd like to make the macro side of your guidance crystal clear. Regarding the outlook you provided, you mentioned every variable at play involving customers, the turning < > aspect of the operation, and so forth. However, I want to be direct: what macro assumptions underlie it? Do you expect conditions to remain basically unchanged from where they stand now? Or is some degree of macro deterioration being factored in? And to fully flesh that out — if I look back at the last genuine downturn you experienced, meaning '08 and '09, traffic still expanded at a solid pace across those years. Yet price erosion caused a substantial slowdown. I realize this company today is clearly nothing like what it was 15 years ago. Still, I'd simply like to grasp your macro premises, and perhaps connected to those premises, what you're presuming about the pricing backdrop. Once more, in line with what you're observing — worse, better? Any detail on that would be appreciated.

Edward McGowan

executive
#18

Certainly. Yes. So, let me think — I'll begin with pricing; so one thing — the place where pricing most heavily affects this business is delivery; we aren't seeing the usual price reductions that one would anticipate within the Security business. That business simply doesn't operate that way. And my expectation isn't that this will become a major issue for us. Clearly, we expect that traffic won't bounce back to the sort of growth rates witnessed previously. We stated as much during our prepared remarks. So a somewhat muted — a more muted Q4 compared with what we normally get. Q4 is usually a robust quarter for us. Our expectation is that the macroeconomic picture won't improve at all. A particular area I'm monitoring closely is Europe. We're observing plenty of difficulties across Europe, observing energy prices, for instance, being extremely, extremely elevated in Europe. We're heading into the winter period. So there's the possibility of reduced graphic tied to subscribers canceling or doing similar things, less online purchasing; so that's among the matters worrying me. And beyond that, the recession and the effect it may have on our customers possibly pushing back purchase decisions and similar things. Still, clearly, we're generating an enormous amount of cash. The business is in excellent condition. It's far better diversified compared with before. So I believe we'll ride out the storm quite well. But I do believe, as both Tom and I noted during our prepared remarks, that it will somewhat restrain our revenue growth here and likewise our margins for a brief stretch.

Josh Baer

analyst
#19

Josh Baer here, filling in for Keith. I'd like to ask about Linode. Are the enterprises you're working with just trialing the platform? Do they view Linode as their main cloud provider? And assuming a multi-cloud approach is the plan, which use cases are enterprises testing on Linode? Which workloads or offerings do you think will catch on with enterprises?

F. Leighton

executive
#20

Honestly, Linode can handle just about any cloud compute application. We clearly offer containers as a service and VM as a service, so whatever you run in the cloud, you could run on Linode. My sense is that at first, the use cases and the way customers see it involve trying it with some new applications, and in that one media company case we discussed, it's a critical application; they run their new transcoding app for user-generated content on us. You can't get much more critical than that. And as I said, that could grow to be worth millions of dollars as it scales, but this company spends many hundreds of millions of dollars in the cloud. So we don't think you're simply going to move all of that onto Akamai immediately. But it's a business I believe we can genuinely grow, and we're positioned extremely well to capture a very large cloud computing market. That market is $200 billion and expanding very quickly. And Akamai is a trusted partner for many major enterprises seeking some diversification, wanting a dependable partner with great performance—which we deliver through our distributor platform—and knowing how to handle scale at an affordable price point, which we're in an excellent position to provide. So I'm really optimistic about the future of our Compute business.

Josh Baer

analyst
#21

Great. And it sounds like much of the emphasis is on hyperscalers and the enterprise chance. I also wanted to check whether you're noticing any shifts in the competitive picture among the other alternative cloud providers that serve SMB, specifically regarding pricing moves or other developments in the market.

F. Leighton

executive
#22

Well, as you're aware, a competitor of ours in that space increased prices; and so far, I'd say the competitive landscape in the SMB market hasn't changed much. We do plan to provide additional offerings, like security, to that customer base. But I believe our true priority is taking Linode and upscaling it so it becomes genuinely enterprise — major enterprise grade — and expanding it out and making it far more distributed, which clearly will boost performance; and pursuing big enterprises, since many of our customers spend 10 or more times as much with the major clouds as they do with us for delivery and security. So that represents a fantastic opportunity for Akamai, being able now to give them compute too, so they can build their apps on Akamai, run them, and obviously deliver and secure them on Akamai. Truly an amazing opportunity.

Frank Louthan

analyst
#23

Could you clarify a bit further — you brought up the media companies and the problems tied to certain traffic decelerating. Is your sense that this is mostly a short-term matter, or is it seasonal? Also, what sort of timeline do you expect before that traffic starts recovering somewhat?

F. Leighton

executive
#24

Sure. Like we said, multiple different elements are weighing on the media companies, and therefore on traffic. You have, in a sense, a year — I mean COVID is still with us, but think of this as a non-COVID year measured against a COVID year. Consequently, folks are moving around outside more, so growth in traffic is smaller because of that. Then there are the effects of what, in certain markets, amounts to a recession. And a number of the media companies have reported that they face their own difficulties around usage. That has a tendency to bring traffic down. Clearly, foreign exchange headwinds are present too, and those affect the revenue we earn from media companies abroad. So every one of those things is putting pressure on traffic within Akamai's media business. My view is that none of this is permanent. Over the long run, we do anticipate traffic coming back to growth rates that are more typical, and that revenue in the Delivery business levels off. It could take some time, several quarters at this point, yet I don't view it as a long-term phenomenon.

Frank Louthan

analyst
#25

All right, that's helpful. And is there a way to put a number on the yearly CapEx savings that come from moving some of the larger-volume customers elsewhere?

Edward McGowan

executive
#26

Sure. The clearest way to frame it is this: as you'll recall, when we entered the year, the guidance we gave was in the 15% to 16% band. Our statement then was that Linode would account for a couple of points; today we sit at 12% to 13%. So something in the 3% to 4% band is likely the figure worth using. I also shared another data point: for our network-only legacy Akamai Delivery operation, or Akamai network CapEx, the level is roughly 3%, whereas historically it had been running near 8%; clearly, investment in the network is still necessary, and we anticipate traffic growth, and so on. Still, my sense is we can stay below what the long-term trend would suggest for some time. Naturally, should traffic rise substantially, delivery results ought to improve and we would handle that when it arises. But as I look ahead across the coming several quarters and into next year, my expectation is CapEx will be at the sort of levels you're seeing now.

Fatima Boolani

analyst
#27

Ed, this one was directed at you, and it concerns all the granular remarks you made about what you expect from the delivery franchise in the back half. There are quite a few moving parts in play. What I'd like to zero in on, though, is perhaps circling back to certain things you said during the prior earnings call regarding what you noticed about gaming and about traffic decelerating. My read, then, is that the majority of the difficulty you're describing on the delivery front stems entirely from OTT, yet I would really appreciate having that confirmed, and perhaps a deeper look under the hood at which traffic categories are showing the greatest elasticity or volatility, as I understand it. Also, how are you viewing those patterns with finer granularity across the back half, particularly now that the renewals sit behind you?

Edward McGowan

executive
#28

Yes, fair question. Let me attempt to respond in the following manner. Clearly, video traffic represents our biggest source of traffic. And what we're observing with video traffic is that its growth rate has fallen short of our forecasts. It's still expanding, yet not as fast as we anticipated. The more obvious spots would likely be gaming, software, and then things tied to advertising, meaning certain portals, news portals, and so forth, the kinds of things affected by advertising; that's where it shows up. However, when video decelerates, given that it accounts for such a substantial share of traffic, that factor can often end up being the largest contributor. Now, will things bounce back? What we generally witness in Q4 is somewhat of a device cycle, where fresh devices appear online, and frequently fresh content arrives during Q4. Back-to-school happens. The fall sports season begins, and typically we observe a substantial traffic bump then. This year, our stance is one of greater caution. We still anticipate an increase similar to what Q4 usually brings, only not to the degree witnessed previously, simply given the market trends we're currently observing, what we hear from other companies, and what our customers are telling us. That's likely the most accurate way to frame it.

Fatima Boolani

analyst
#29

Thanks for that. And very briefly, regarding the macro remarks you provided, particularly a few of the main points about Europe. I'm wondering whether you have any worries or things you've noticed when it comes to sales cycles getting longer, decisions on deals — postponements in making deal decisions and completing deals from a procurement angle, anything concrete you could tell us about those dynamics, just as the wider economy has somewhat weakened.

Edward McGowan

executive
#30

That's a fair question. I'd say we're still in the fairly early stages, though it's definitely something we're watching. Security is usually one of those areas where you wouldn't put off a purchase, particularly when you're being attacked. If no attack is happening, at times you may stretch that out a bit; however, we have not yet observed anything that deserves mention regarding sales cycles getting longer. What worries me more is what will happen as we head into the winter, and there's discussion about reducing natural gas usage and the effect on GDP, which clearly would affect not only Akamai but a great many other companies. So that's something to watch for; and beyond that, once you find yourself in a situation like that, paying close attention to receivables and ensuring that you don't — that your customers are financially healthy. Up to now, no problems there, but once again that's something we would be monitoring.

Tim Horan

analyst
#31

I've got a couple of things on Linode. First, what kind of reduction in compute expenses do you believe enterprises can achieve by using Linode? And second, does this represent a solid run rate for the Compute segment, or could you give any detail about the growth rates you're targeting there?

F. Leighton

executive
#32

Sure, the savings can be quite meaningful. Our published rates for Linode are much lower than what you'd find from the hyperscalers — that information is publicly available — and for large enterprises, naturally, some discount off that is possible. What's really encouraging is that we think we can deliver this while maintaining strong margins; so as the business expands, in fact, it will lift our margins overall; and Ed, would you like to address the second question?

Edward McGowan

executive
#33

Okay, regarding the run rate and our growth expectations, I'd say that's probably a reasonable place to peg it for the short term. Tom and Adam talked in May at our Analyst Day about how we're making some major improvements to Linode's capabilities, and those things don't wrap up overnight. So to be honest, we'll start to see — the genuine enterprise ramp is going to arrive in '23, in the back half of '23, and in '24, and beyond. Even so, we do expect pretty substantial growth here, and obviously Compute rose over 74%, but for now it's a fine figure to use, even though expanding both capacity and the extra features is going to take some time. Like Tom said, we're progressing at a good pace, but the moment when you really begin to see that exciting growth comes once all that capability is up and running, and that shift really took shape across 2008 and 2009.

Rudy Kessinger

analyst
#34

Okay, so looking at what's left of this year, you've lowered the revenue guidance by $55 million. You mentioned that FX headwinds now stand at $114 million for the year, compared with $100 million before. So once you strip out that FX drag, the reduction is $41 million. I'd like to know whether you can break that $41 million decrease into pieces — how much comes from a weaker traffic growth outlook or macro pressures, versus perhaps revenue you're giving up because you're accepting less traffic from certain customers with high peaks, or other places where you're cutting the guidance.

Edward McGowan

executive
#35

Sure. Most of it, I'd say, is attributable to the slower traffic. That represents the largest piece of it. The effect of some of the traffic we're declining to take isn't particularly large, though it does play a role. I can't give you a precise percentage, but it isn't a major portion. After that, the macroeconomic piece would account for what's left. So if you ranked them, it would be slower traffic first, macroeconomic second, and then the effect of turning away some of that traffic we discussed when it comes to peaky traffic.

Rudy Kessinger

analyst
#36

All right. Turning to Linode, I'm interested in knowing what your overall spending on hyperscalers amounts to internally at this point? How much of that do you figure could shift over to Linode eventually? And what sort of expense reduction do you expect that would deliver?

F. Leighton

executive
#37

Sure. Right now we're paying roughly $100 million each year to the cloud providers, and across the coming 1 or 2 years, essentially all of it — nearly the whole amount — would be moved over to Linode. The savings would be considerable. That's the reason this ranks among a handful of efforts we've launched to lift margins; beyond that, it's wonderful to be consuming our own offerings, and as we continue working with Linode to prepare it for enterprise use, it will benefit us to run it across the full range of applications here at Akamai. In many ways we resemble our own customers, seeking out ways to cut costs while achieving strong performance from our cloud compute platform.

Jeff Van Rhee

analyst
#38

Just a few things from my end. Regarding the Security segment, I'd like some clarification. I realize you mentioned feeling at ease with 24% revenue growth. To be clear, is that as reported or constant currency, and then what are the various puts and takes around being able to maintain that 20% in the outyear, considering the deceleration we're seeing right now?

Edward McGowan

executive
#39

Sure. I'll handle the first part, Tom, you can address the second part regarding the future; that figure is in constant currency. And keep in mind, Jeff, that the anniversary of the Guardicore acquisition falls in the fourth quarter.

F. Leighton

executive
#40

Absolutely. Looking further out, we're clearly experiencing strong expansion from our main offerings, Web App Firewall and Bot Manager. Those bring in the biggest share of revenue. Growth there is quite solid, though they sit on smaller bases, naturally, when it comes to Zero Trust enterprise computing. Over a longer horizon, that's the piece I believe fuels much of our security momentum. The infrastructure segment is expanding at a slower pace right now, and it also rests on a mid-sized figure. Still, my view is that eventually you'll witness growth in the enterprise Zero Trust security group, with Guardicore leading as the market frontrunner in halting ransomware.

Jeff Van Rhee

analyst
#41

So I suppose, taken together, there isn't any obvious item you'd point to that ought to push things faster or slower — maybe another angle on it.

F. Leighton

executive
#42

Well, we currently have those 3 categories. App and API security is the largest, and it's growing quite strongly. My expectation is that robust expansion continues there. There's plenty of innovation happening around Account Protector, and Page Integrity Manager, which we described as being quite significant for PCI compliance; plus we've added a new audience hijacking prevention feature that seems genuinely impressive. Infrastructure is a medium-sized business at present, and its growth isn't as rapid. That covers your DDoS protection along with your D&S defenses; after that comes the area expanding fastest of all, Zero Trust enterprise security, which Guardicore leads, though among the 3 it's the smallest. So in effect you've got 3 separate comps, each with somewhat distinct dynamics.

Jeff Van Rhee

analyst
#43

All right. If I may, one final question regarding delivery. Clearly, adjustments are being made by you concerning that peak traffic. But what about pricing rules for traffic more broadly? My recollection is that during the prior quarter you mentioned adopting a far more conservative stance on pricing across the board. Setting aside that peak period, how has your method shifted in this quarter, and how is it apt to shift for the balance of the year?

Edward McGowan

executive
#44

Right. So for the past several months, we have been putting that into practice. And as Tom mentioned, churn has not been significant for us. On the renewals we have completed, we have done reasonably well. Discounts are still being extended by us to those customers experiencing traffic growth; the difference is simply that they are not at the levels we offered in the past.

Charlie Erlikh

analyst
#45

This is Charlie Erlikh, sitting in for Will. I've got a single question regarding pricing as well as contract renegotiations, sort of piggybacking on what you just said. Are there any significant agreements up for renewal during the back half that we ought to know about? Then, regarding those 8 you referenced from the first half, could you share more detail about the general tone and the discussions — things like pricing pressure, which you touched on briefly — but also the financial health of the customers involved, plus anything further along those lines would be appreciated.

Edward McGowan

executive
#46

Yeah, absolutely; nothing major is on the horizon for the back half. As I mentioned, whenever a large cluster appears in the — we'll flag it for everyone. I believe offering some commentary and making you all aware of the situation is simply useful. When it comes to the dynamics, aside from some resistance around the peak-to-average issue, which forms a standard part of our dialogue, I'd say things largely matched our expectations. We did attempt to follow our approach of exercising greater discipline. I'm not certain that's the best way to phrase it, but being more responsive to volume trends and discussing anticipated volume increases with customers, and so forth. So we're happy with the outcome. And clearly, renegotiating with major customers in the — and the delivery division is never enjoyable. But it turned out roughly as anticipated; now, your remaining question concerned — could you remind me once more?

Charlie Erlikh

analyst
#47

I believe you covered every point, actually — yes, just the general feel of those discussions. But yes, in my view you gave a solid answer.

Edward McGowan

executive
#48

Right. So on the whole, I'd say our customers remain in excellent condition. A number of them are quite big brands and continue to be very healthy. We've brought services to nearly all of them. That means we aren't merely handling delivery — on top of that, we're layering in extra capabilities too.

Alex Henderson

analyst
#49

Since neither Guardicore nor Linode has been part of the company for a full 12 months, I wanted to ask whether you could assess them on an apples-to-apples basis — had they been included for the whole year, what would their internal growth rates be — so we can get some sense of how fast those 2 acquisitions are growing.

Edward McGowan

executive
#50

Yes, that's a fair question, Alex. I'll handle this one, Tom. So Linode — Linode will wind up being with us for the majority of the year. But right before we acquired them, as we noted in our commentary, they were expanding at roughly 15%. Clearly, we'll speed that up once we begin bringing on enterprise customers. So I would anticipate that figure rising. On Guardicore, if you recall our first call, we're raising our guidance considerably, but looking at where our internal — where our run rate stands today, we're seeing more than $60 million. That's nearly twice what they had when we bought them, even a touch above that. So Guardicore is growing very rapidly; on a standalone basis, I'd say it would be greater than a double; and Linode, we just brought it in and it's beginning to add that functionality, and we expect that growth rate to accelerate, but they were at about 15% before the acquisition.

Alex Henderson

analyst
#51

Earlier on this call, someone said that once you strip out acquisitions, the growth rate underneath came to 17%. Can you verify whether that calculation is correct?

Edward McGowan

executive
#52

Yeah, the math there is basically right. On a constant currency basis, it was 21%. Guardicore contributed roughly $14 million. That works out to around 4%. So yes, that figure is essentially accurate.

Alex Henderson

analyst
#53

Great. Now, when we turn to overseas markets, what's your approach to the currency fluctuations happening right now? Regarding pricing, are you compensating for the effective dollar-denominated price cut that results, whether you're billing in the local currency or not — are you lowering prices to ease that strain? How do you deal with the pressure this puts on customers?

Edward McGowan

executive
#54

Sure. The majority of our clientele beyond the U.S. remit payment to us in their own currency. Billing is therefore done in that local currency. A handful, in specific nations, for instance, will instead be in U.S. dollars. Thus far, customers haven't resisted much, so it hasn't become a major problem. We engage in a bit of hedging on the balance sheet, though not operationally — no cash flow hedging. As for pricing, that's something we factor in when working with clients located outside the U.S. Beyond that, generally speaking, it also depends on where their traffic originates. That plays a role too. Should we be serving nations where costs run higher, we account for that as well.

Tom Barth

executive
#55

Alright. Alex, thanks, and thanks to all of you. So to wrap up, over the remainder of the third quarter we'll be taking part in quite a few investor conferences and events. You can find the specifics for those in the Investor Relations portion of akamai.com. We appreciate you being with us, and everyone here at Akamai sends wishes for continued good health to you and your loved ones. Enjoy your evening.

Operator

operator
#56

That brings today's presentation to an end. We appreciate everyone joining us. You're welcome to leave the call now.

Read the full transcript via the API

You're viewing the first half of this call. Get the complete Akamai Technologies, Inc. transcript — plus 133,000+ transcripts from 7,000+ companies, speaker segments, AI summaries and full-text search — through the EarningsCalls.dev API.

Get the API View API docs →

This call discussed

For developers and AI pipelines

Programmatic access to Akamai Technologies, Inc. earnings transcripts and 133,000+ others is available through the EarningsCalls.dev REST API. Plans from $39.99/month — full transcripts, speaker segments, full-text search, and the recently-added /api/v1/transcripts/recent polling endpoint for ETL pipelines.

More tools for stock research

OptionsBell: an unusual options activity scanner that tracks options flow across all US stocks and emails you when large trades appear in Akamai Technologies, Inc. or any ticker on your watchlist.

StockMarketScan: a stock screener with alerts: 24 pre-built screens refreshed every 5 minutes, watchlists and email notifications. Both offer API and MCP access and fair monthly pricing.