Hello and welcome to the Akamai Technologies First Quarter 2025 Earnings Conference Call. All participants will be in listen-only mode. After today’s presentation, there will be an opportunity to ask questions. Please note, this event is being recorded. I would now like to turn the conference over to Mark Stoutenberg, Head of Investor Relations. Please go ahead.

Good afternoon everyone and thank you for joining Akamai's first quarter 2025 earnings call. Speaking today will be Tom Leighton, Akamai's Chief Executive Officer; and Ed McGowan, Akamai's Chief Financial Officer. Please note that today's comments include forward-looking statements, including those regarding revenue and earnings guidance. These forward-looking statements are based on current expectations, assumptions that are subject to certain risks and uncertainties and involve a number of factors that could cause actual results to differ materially from those expressed or implied. The factors include but are not limited to, any impact from macroeconomic trends, the integration of any acquisition, geopolitical developments and any other risk factors identified in our filings with the SEC. The statements included on today's call represent the company's views on May 8, 2025. And we assume no obligation to update any forward-looking statements. As a reminder, we will be referring to certain non-GAAP financial metrics during today's call. A detailed reconciliation of GAAP to non-GAAP metrics can be found under the financial portion of the Investor Relations section of akamai.com. With that, I'll now hand the call off to our CEO, Dr. Tom Leighton.

Thanks, Mark. I'm pleased to report that Akamai got off to a solid start on the year. First quarter revenue grew to $1.015 billion, up 3% year-over-year reported and up 4% in constant currency. Non-GAAP operating margin came in above guidance at 30% and non-GAAP earnings per share came in well above the high end of our guidance range at $1.07, up 4% year-over-year and up 6% in constant currency. Security and compute combined to account for 69% of our total revenue in Q1, growing 10% year-over-year as reported and 11% in constant currency, underscoring Akamai's ongoing transformation from a CDN pioneer into the cybersecurity and cloud computing company that powers and protects business online. Security growth in Q1 was driven in part by continued strong demand for our market-leading Guardicore segmentation solution as more enterprises relied on Akamai to meet compliance requirements and to defend against malware and ransomware. The FBI's Internet crime report last month called ransomware the most pervasive threat to critical infrastructure and segmentation is the last and most important line of defense for an enterprise. Once an attack finds its way through the perimeter defenses, Guardicore is designed to spot it and proactively prevent it from spreading and causing serious harm. In Q1, we achieved several large competitive takeaways for our Guardicore solution, including at a major bank in the U.S. and a government revenue and customs authority in Europe. Both told us that they were not getting the help they needed from their incumbent provider. They also told us that their trust and confidence in Akamai were factors in switching from the competitor to us. We beat out another station competitor at a clinical research laboratory and at a major credit union which told us that our solution was easier for them to manage while offering unified control over their on-prem and hybrid environments. In addition, we expanded existing contracts for Guardicore at a well-known financial analytics company in the U.S., a global Fortune 500 manufacturer in Europe, and a major telco in Latin America. We also continued to see strong interest in our market-leading API security solution in Q1. Almost everything online today leverages APIs, including attackers who know that APIs usually don't come with sufficient security and controls. Akamai's Internet security report issued last month showed that API vulnerabilities have cost organizations around $87 billion a year, with shadow and zombie APIs being especially vulnerable to attack. When we conduct proof of concept for our API security solution, most CISOs and CIOs are surprised to learn just how many APIs they have exposed. For example, at one of our customer events in Q1, the CISO of a major Korean company told attendees about how our solution detected numerous undocumented APIs, along with multiple vulnerabilities revealing about 40 security issues, ranging from exposure of sensitive data through unauthenticated and publicly accessible APIs to weak password policies and unencrypted passwords. In Q1, we signed contracts for API security with numerous companies, including a leading U.S. fintech provider, one of the largest banks in Canada, one of the largest pharmacy benefits managers in the U.S., two large U.S. insurance companies, a well-known auto manufacturer in the U.K., a global online fashion retailer in China, and a multinational investment bank in Australia. Last week at the RSA Security Conference, our API security solution won a global Infosec Award from Cyber Defense Magazine. Also last week at RSA, we announced our newest security offering, firewall for AI which CISO Magazine highlighted as one of the top cybersecurity products at RSA and which CRN hailed as one of the coolest new cybersecurity products at the show. Many enterprises today are deploying LLMs to provide myriad services such as chatbots, inference engines, content generation, and cataloging. Some are beginning to go further and deploy agentic AI, software agents with the ability to gather information, connect with other tools, reason, make decisions and then act autonomously. In fact, Deloitte forecasts that one-fourth of the organizations using GenAI today will deploy AI agents this year and half will use them by 2027. As is often the case when enterprises adopt new technologies, cybercriminals quickly learn how to attack the new technology for nefarious purposes. GenAI and the use of AI apps and agents are just the latest examples, with various exploits already well publicized. Akamai's firewall for AI is designed to protect AI agents, AI-powered applications, LLMs, and AI-driven APIs from these new threats. By securing inbound AI queries and outbound AI responses, it offers multilayered protection by detecting and blocking sensitive data leaks in AI-generated responses, defending against remote code execution, model back doors, and data poisoning attacks, filtering inappropriate content, such as misinformation, hate speech, and other offensive material, and mitigating AI-driven denial of service attacks by controlling excessive carry usage and model overload. The new product is generating strong interest from customers. For example, a financial services firm told us, 'Akamai's AI firewall has given us great insight into the interactions with our AI chat agent which helps paint a picture of the threat landscape unique to LLMs. This tool will allow us to save critical systems resources for real clients rather than being consumed by bad actors.' In Q1, we also received accolades for our other security products. Akamai was named a leader in the Forrester Wave Web Application Firewall report and achieved the highest possible score for 11 criteria including detection models, Layer 7 DDoS protection, pricing transparency and flexibility, roadmap, and vision. Our WAF continues to be an industry leader and serves as the bedrock for app security used by many of the world's largest enterprises. For example, in Q1, we signed multimillion-dollar contracts with a global bank based in India and the world's third largest railway with over 60 million daily users. As is often the case, these new WAF customers are also making use of our content delivery services to provide the best possible performance for their web applications. Turning now to delivery. I'm pleased to report that we saw better-than-expected results for our delivery products in Q1. This was due in part to improved overall traffic growth, our continued excellent performance for both enterprise web apps and large live events and incremental growth from new customer accounts that we acquired from Edgio in December. One of those new customer accounts led to a $16 million commitment over 3 years for Akamai to deliver 100% of their traffic along with Akamai API security, app and API Protector, professional services, and two solutions from our compute ISV partners. That's a great example of the opportunity and competitive advantage we have by providing world-class delivery, security, and compute together on a single platform, supported by the best in managed professional services. Turning now to cloud computing. In Q1, we continued our strong momentum in cloud computing, introducing new capabilities to serve customers, signing up new accounts and expanding our go-to-market reach, both in-house and with partners. In March, we introduced our new Akamai Cloud inference solution which provides what we believe is a better architecture for customers to build and run AI applications, data-intensive workloads closer to end users. AI inference and agentic AI apps often require high throughput networking to manage large volumes of data and return accurate results within milliseconds. By running these workloads at the edge, we can achieve better throughput, lower latency and significant cost savings compared to other cloud providers in the market. For example, in one proof of concept, a company's artificial intelligence application achieved 30% faster response times using our new Butte platform than they had with a hyperscaler. In another proof of concept, a publishing company saw 2.5x faster response times with 3x higher throughput and significant cost savings compared to another hyperscaler. Customers who signed deals with us for cloud computing in Q1 included a global live streaming service with hundreds of millions of users, a major cybersecurity provider in the U.S., a European industrial products company, and a global developer of immersive video games. We also saw numerous large deals in Q1 come through our growing roster of ISV partners, including one of the largest retailers in the world, one of the large pharmacy benefits managers, a major European fashion retailer, and a leading broadcasting company in Japan. As a result of our competitive advantages, Gartner named Akamai an emerging leader for GenAI specialized infrastructure. It's one of the reasons why Vast Data, the AI data platform company, has partnered with Akamai to make data-intensive AI inferencing more efficient for our joint customers. Our edge computing capabilities will be further enhanced by the introduction of our new managed Container Service, or MCS, that we announced in Q1. MCS will provide support for customer containers in any of our 4,000-plus pops around the world. That means we'll be able to run customer compute instances in over 700 cities. No other provider comes anywhere close to doing that today. Also in Q1, Akamai became the first and only provider to offer video processing units, or VPUs, in the cloud, with our new accelerated compute instances solution built on specialized processors from Net. A VPU architecture can offer up to 20x greater throughput than CPU solutions which results in greatly improved performance as well as significant savings for media companies. This gives Akamai another way to complement and cement our long-standing relationships with major media companies, which include all of the top 10 media companies in the world. It's also very synergistic with the media workflow services provided by our ISV partners on Akamai Cloud. Overall, we believe that the combination of world-class delivery, compute, and security available on our platform provides a low-cost, high-performance, and massively scalable alternative to the hyperscalers for media and entertainment companies. And unlike the hyperscalers, we do compete with our customers. At the NAB Show in Las Vegas last month and in recent bids with customers across Europe and Asia, many executives told me how they see real value in what we're doing, in part because they're tired of writing huge checks to the hyperscalers who then use the funds to compete against them. On the go-to-market front, the sales transformation efforts that we outlined during our last call are on track as planned. We're making good progress in rebalancing our sales team to provide greater focus on new customer acquisition while maintaining strong customer relationships. We've implemented a better sales play methodology to improve our installed base penetration, especially for fast-growing security and cloud infrastructure offerings. And we're seeing good early success with changes to reward sellers and customers for longer multiyear contracts. Like all of you, we're keeping a close eye on global economic and political challenges. As we noted during the call, we've taken steps to minimize the impact of tariffs on our business. And as of now, we anticipate that the direct impact to Akamai from tariffs in 2025 will be about $10 million in CapEx, which is amortized over 6 years. That said, there is concern among some of our customers regarding a possible recession which could impact later in the year. Some of our customers outside the U.S. have raised concerns about whether they should rely on American companies for critical infrastructure. Neither issue has impacted our business to date. We're working to reassure customers that Akamai will continue to serve them as we always have. We're also staying very engaged with our U.S. federal sector customers. Overall, we derive less than 5% of our revenue from the U.S. public sector, including state and federal. Based on our current line of sight, we had to lose a few million dollars of revenue in the back half of 2025 related to federal cutbacks. There's also a possibility of increasing revenue in situations where our solutions can generate significant savings for federal agencies. Lastly, we were very pleased to see Akamai recognized last quarter as one of the most trustworthy companies in America by Newsweek Magazine which partnered with a market research firm to analyze more than 100,000 evaluations generated by customers, employees, and investors. Trust and integrity are core values at Akamai and they really matter to customers who rely on us to be there for them. I want to express thanks to our employees who work so hard to help Akamai earn the trust of our customers and shareholders. Now, I'll turn the call over to Ed to say more on our Q1 results and our outlook for the rest of the year.

Thank you, Tom. Today, I have to review our Q1 results and then provide some color on our Q2 expectations and our updated full year 2025 guidance. As Tom mentioned, we got off to a solid start to the year with total Q1 revenue of $1.015 billion, which was up 3% year-over-year as reported and 4% in constant currency. We continue to see strong performance within our compute and security portfolios during the first quarter. Compute revenue grew to $165 million, a 14% year-over-year increase as reported and 15% in constant currency. Security revenue was $531 million, growing 8% year-over-year reported and 10% in constant currency. During Q1, we had approximately $6 million of one-time license revenue compared to $4 million in Q1 of last year and $12 million in Q4 2024. For the first quarter combined revenue from compute and security increased by 10% or 11% in constant currency, accounting for 69% of total revenue. Our delivery revenue was $319 million, down 9% as reported and down 1% in constant currency. Delivery revenue was stronger than expected in Q1. And while it's too early to call a bottom in delivery, we are encouraged by some improving trends in the delivery business to start 2025. International revenue was $486 million, up 2% year-over-year or 5% constant currency representing 48% of total revenue in Q1. Foreign exchange fluctuations had a negative impact on revenue of $5 million on a sequential basis and a negative $14 million impact on a year-over-year basis. Finally, it's worth noting that GEO contributed approximately $23 million of revenue in the quarter which was in line with our expectations. Moving to profitability. In Q1, we generated non-GAAP net income of $256 million or $1.70 of earnings per diluted share, up 4% year-over-year as reported, up 6% in constant currency and well above the high end of our guidance range. Our EPS outperformance was driven by better-than-expected Q1 revenue, lower-than-expected transition services or TSA costs related to the traction, better-than-expected bandwidth costs, lower payroll taxes primarily related to stock-based compensation as a result of a lower stock price, and lower employee medical claims related to our self-insured medical plan. Finally, our Q1 CapEx was $226 million or 22% of revenue. Our first quarter CapEx came in slightly lower than our guidance which was mostly due to timing as some CapEx has been pushed from Q1 into Q2. Moving to our capital allocation strategy. During the first quarter, we spent approximately $500 million to buy back approximately 6.2 million shares. We ended the first quarter with approximately $1.5 billion remaining on our current repurchase authorization. Our intention remains the same — to continue buying back shares to offset dilution from employee equity programs over time and to be opportunistic in both M&A and share repurchases. As of March 31, we had approximately $1.3 billion in cash and cash equivalent securities. It's worth noting that subsequent to the end of the first quarter and prior to today's earnings announcement, we used cash on hand and funds available under our revolving credit facility to fully repay $1.15 billion of our outstanding convertible notes that matured on May 1 of this year. Before I provide our Q2 and full year 2025 guidance, I want to mention a few items. First, we have completed our migration of GEO customers to the platform. As a result, we will not have any additional TSA costs moving forward and our revenue expectations from the transaction remain the same, approximately $85 million to $105 million of GEO revenue contributions for 2025. Second, we expect an increase in operating expenses for the second quarter, partly due to a weaker U.S. dollar as well as higher marketing expenditures for Q2 events, our annual sales President's Club trip, and the impact of our annual employee merit cycle which takes effect on April 1. Third, we continue to expect our CapEx to be heavily front-end loaded with significantly higher expenditures in the first half of the year compared to the second half of the year. This includes approximately $10 million of CapEx pulled forward to the first half of the year to help mitigate potential tariff risks. Fourth, we expect interest income to decline in 2025 due primarily to lower cash balances resulting from stock repurchases, recent acquisitions, and the retirement of our $1.5 billion convertible debt. Additionally, we anticipate lower investment yields as interest rates are projected to come down throughout the year. As a result, we expect net interest income to decrease by approximately $5 million per month starting in May of 2025. Finally, we are maintaining our forecast ranges to effectively navigate increased volatility within the current economic environment, the volatility in the foreign exchange markets, and the potential impact of the pending TikTok ban. So with those factors in mind, I'll now move to our Q2 guidance. For Q2, we are projecting revenue in the range of $1.012 billion to $1.032 billion or up 3% to 5% as reported and in constant currency over Q2 2024. At current specs, foreign exchange fluctuations are expected to have a positive $15 million impact on Q2 revenue compared to Q1 levels and a positive $7 million impact year-over-year. At these revenue levels, we expect cash gross margins of approximately 72%. Q2 non-GAAP operating expenses are projected to be $315 million to $320 million. This is slightly above Q1 levels due to the items I just mentioned. We expect Q2 EBITDA margin of approximately 41% to 42%. We expect non-option expense to be between $135 million and $107 million and we expect non-GAAP operating margin of approximately 28% for Q2. Moving on to CapEx. We expect to spend approximately $226 million to $236 million. This represents approximately 22% to 23% of our projected total revenue. Based on our expectations for revenue and costs, we expect Q2 non-GAAP EPS in the range of $1.52 to $1.58. This EPS guidance assumes taxes of $54 million to $57 million based on an estimated quarterly non-GAAP tax rate of approximately 19% to 20%. It also reflects a fully diluted share count of approximately 148 million shares. Looking ahead to the full year for 2025, we expect revenue of $4.050 billion to $4.2 billion, which is up 1% to 5% as reported and in constant currency. As a reminder, we would expect to come in at the higher end of that range if we see continued weakening of the U.S. dollar, traffic growth materially exceeding levels, and there is no ban in the U.S. for our largest customer. We would expect to come in at the mid to lower end of that range if we see significant strengthening of the U.S. dollar, a significant downturn in the economy in the back half of the year, or traffic growth materially slowing from current levels. Moving on to security. We continue to expect security revenue growth of approximately 10% in constant currency for 2025. We also continue to expect the combined ARR from our Zero Trust rise and API security solutions to increase by 30% to 35% year-over-year in constant currency for 2025. For compute, we continue to expect growth to be approximately 15% in constant currency. As a reminder, included within our compute, we expect cloud infrastructure ARR year-over-year growth in the range of 30% to 45% in constant currency for 2025. At current spot rates, our guidance assumes foreign exchange will have a positive $8 million impact on revenue in 2025 on a year-over-year basis. Moving on to operating margins. For 2025, we are estimating non-GAAP operating margin of approximately 8% as measured in today's FX rates. We anticipate that our full-year capital expenditures will be approximately 19% to 20% of total revenue. Moving to EPS. For the full year '25, we expect non-GAAP earnings per diluted share in the range of $6.10 to $6.40. This non-GAAP earnings guidance is based on a non-GAAP effective tax rate of approximately 19% to 20% and a fully diluted share count of approximately 150 million shares. In summary, although we anticipate heightened economic volatility, we believe that Akamai is in a strong position to continue delivering revenue growth and maintaining healthy margins. This is driven by our newer security and cloud computing products, a moderation in our content delivery revenue declines, and our commitment to disciplined cost management. With that, I'll wrap things up and Tom and I are happy to take your questions.

Today's first question comes from Amit Daryanani with Evercore.

I guess maybe to just start on the delivery side. Can you just talk about what really drove the upside for you folks in the quarter? Was it more pricing or traffic driven? And then what's reframing saying this can sustain into the out quarters? Just in what was the upside and why the hesitation about sustaining?

Amit, this is Ed. Thanks for the question. Yes, so I'd say what drove it was really traffic. Pricing typically, unless you have large renewals, doesn't have a huge impact in any given quarter and we didn't have any significant concentration on renewal. So it was all about traffic growth. And it was pretty strong pretty much every sub-vertical. So we saw strong video traffic, strong gaming traffic, strong software downloads, and even across commerce and some of the other sub-verticals. So it's just a better improved environment. And just being cautious in terms of calling a bottom. This is the third quarter in a row here where revenue has kind of been flattish sequentially which is better than what it's been which has been declining. But we're seeing decent trends here in April and we're just being a bit more cautious and don't call the bottom quite yet.

Got it. That's still fair. And if I just follow up on the security side, I'd love to hear how you would characterize your performance in the March. It looks a little bit lower, I think, like not $15 million from what the Street was modeling perhaps. I'd love to understand how that ended up versus your internal expectations. And then as you think about this acceleration from 8% to call it 10% in security, what enables us for the year?

Yes. Good question. So I would say it came in, in line with what we had expected for the quarter. And as you heard a few minutes ago, I reiterated our guidance for 10% constant currency and we did deliver 10% constant currency this quarter. There's a little bit of noise in there with license revenue and I sort of called that out for you. We had about $12 million of license revenue last quarter, about $6 million less this quarter. So that can sometimes folks don't quite get the model right, going from Q4 to Q1. And also with some of the bundles, Q4 tends to have a little bit of volumetric revenue associated with some of our commerce customers and things like that, that you don't see repeat at such a level in Q2. So I would say it came in, in line with what we expected. Big drivers for growth is exactly what we had thought in terms of the bookings. We're seeing a very strong API security growth and very strong Guardicore growth as we expected. And things are playing out exactly how we had thought.

The next question is from Jonathan Ho with William Blair.

Congrats on the strong quarter. Can you provide a little bit more detail on the role you can play with agentic AI with your AI firewall? And perhaps help us understand how you could potentially benefit as AI utilization in the enterprise really begins to expand?

Yes. There's a lot of attacks now that have been published against AI apps of all kinds, including agentic AI. Attacks where the adversary is able to trick the agent, the AI agent into giving up sensitive data, code, personally identifiable information, trick the agent into saying things that could be offensive content, could be deals, prices for things that aren't right. You've got to protect the app, particularly if it's an agent making decisions and doing things. You have to protect it from the attackers that cause it to do bad things. You've got to protect it from ingesting bad information. A lot of these apps are learning as they go along. The adversaries have figured out how to submit information to it that causes it to learn bad things which can cause problems in the future. AI is a brand-new attack surface and is being rapidly deployed. In fact, many enterprises are forming rules inside the enterprise regarding what they can and can't do with AI. But on the other hand, they can't even keep track of all the AI applications that they are using and supposed to be utilizing. Another aspect of what we're doing, same as we do for API security, is helping our customers know what AI apps they have exposed, and then ensuring that they're protected. I imagine this will be really a war of regulations. The first attacks are being seen out there. We have the first line of defenses now with the OAS top 10 and additional capabilities tailored to particular customer use cases. I think there will be some back and forth where more attacks will be figured out. Our job is to stay ahead of that and keep our customers safe.

Excellent. And you mentioned some customers expressing worry over whether they could rely on U.S.-based services. Is there potential for diversification or multi-sourcing pressure from some of your customers internationally? Just wondering.

Today, I don't think there's good alternatives, especially when it comes to security. Pretty much all the world's major banks and most of the major commerce companies rely on Akamai for their security because we're really the best in terms of protecting them. There are not good alternatives for them today. I think over the long haul, our job is to make sure that they understand that they can continue to rely on Akamai. No matter what happens with the geopolitics and the rhetoric, but it's a topic that's come up in some accounts. We're being careful that they understand that Akamai is here for them over the long haul. That we're not going away and that they can continue to trust us.

The next question comes from Will Power with Baird.

Okay, great. I come back to the Security segment. You called out a number of competitive takeaways for the segmentation products. I guess it would just be helpful if you can kind of just help distill down what's really setting you apart in that market, allowing you to take share? And then secondly, it'd be great to get any updates on the progress, trends within the kind of the rest of the broader Zero Trust portfolio and where we are with go-to-market since there?

Yes. I think scale sets us apart. The larger enterprises could have hundreds of thousands of applications and devices that need to be protected, and we uniquely can do that. Ease of use has come up in several accounts, particularly with our new AI-enabled interface that can make the integration be a lot faster and simpler. Our new engine suggests the initial firewall rules that should be set up for all the various devices in the enterprise. It actually tells our customer and their operator what the various devices are, notifies them if firewall rules are out of date or not the most recent set of rules that they need to have. So ease of use is really important because if you think about installing hundreds of thousands of agents, you need really good automation to do that effectively. And trust is really important. We've been told that by our customers because these agents are running inside the company, and it's vitally important that they be reliable. Akamai has a great track record there. We're not the lowest cost provider; there are others out there with lower-cost solutions. But when it comes to protecting critical infrastructure from ransomware devastation or data exfiltration with software that's running really in the inside important applications, Akamai is pretty unique in being able to do that and being trusted for that. In terms of Zero Trust, Guardicore is our flagship solution, and we do have other capabilities. We've combined North, South, East, West protections built around Guardicore and our Enterprise Application Access solution so that we can provide a really comprehensive set of solutions to protect the enterprise applications and data.

Okay, great. No, that's helpful. And maybe just to add quickly for you, nice to see the better margins in the quarter. I mean, it sounds like you expect to dip back down in Q2. I'm just trying to sort out what proportion of that upside in the quarter could be more sustainable versus some conservatism going forward?

Yes. I think for this quarter, in particular, I'd say you'd have to see some revenue upside to drive higher margins. We obviously have very good leverage. If you think about, as I broke out in the prepared remarks, the items that drove the upside both in Q1 and then in Q2, what's driving the higher cost. Some of it is event driven. But there is the — we did pull our merit cycle in a little bit earlier this year. We typically do it in the middle of the year, so we just brought it to April. So that has a recurring cost. We are doing some hiring, so that has an impact as well. But we think over time, as the business grows, we will improve our margins. Maybe in the back half, there’s potential to expand margins a bit as well.

The question comes from Frank Louthan with Raymond James.

Great. This is Rob on for Frank. So can you talk about the recent sales changes? Where are you as far as the new folks you need to hire, moving people around within different roles between the hunters and farmers? And then what are some other drivers of improvement that you see taking shape the rest of this year and into next year?

I'll start with that and then hand it off to Ed on the margin question. I'd say we're about one-third of the way there. We've made a lot of changes, seeing some very good progress. We're shifting for the force towards hunting now that we have products that a lot of new to Akamai customers need, for example, API security, segmentation, cloud computing, and they're going to need firewall for AI as well. We're also hiring more specialists that have a lot of expertise that can help facilitate those sales. We've made shifts to incent both our sellers and customers to longer-term contracts starting to see some impact there which is good. Overall, over time, we're growing our overall presence in the marketplace. So I'd say we're about a third of the way there and making good progress.

Yes, sure. I'd say it's probably three main drivers. The first one would be just as we scale up our compute business. Right now, we are subscale. If we look at some of the facilities as we start to sell those out and get to a bit of scale inside of some of the different areas, we are seeing very attractive margins in line with what we talked about. We do have a long way to go in terms of reaching scale. As we start to grow and compute, I do expect our margins to expand, obviously. There. And then as security continues to grow, it's a higher-margin product for us. API security is a very high-margin product for us, much lower CapEx associated with your securities. So that's going to help. The stock being such a big drain also helps as well. So if we get back to low single-digit or even flat CDM growth, that obviously has a big help on margins. As far as other operating areas, we're always looking into efficiencies; we've run out a lot of costs. There's a little bit in the room potentially in real estate, and if we can find a lay on some of our unused space. With compute in terms of our own use of compute, we've migrated a lot of our stuff to the node, and we'll look to do some more. But it's really going to come from scaling up the business and as the mix shifts over time.

The next question comes from Jeffrey Van Rhee with Craig-Hallum.

Just a couple maybe on the security front. Just high level, if I look at the organics and try to back out the acquisitions, it looks like it's fallen off fairly quickly from 15 to 10 to maybe now mid-single digits going sequentially last quarter was 10% year-over-year and this quarter looks like mid-single digits. Just talk to me a little bit. I know you touched on it last quarter but I want to make sure I fully understand what's played out there on the security. And I think you had pointed to WAF, but maybe if you could just start there.

Yes, sure. So, yes, there’s sort of inorganic in the number, but it's not a significant portion of revenue for us. In terms of if you break out the business into the different subsegments, we are seeing WAF slowdown naturally a little bit. It's still growing, but it's not growing nearly as fast as it was; very high penetration, which is expected. Obviously, API security is growing incredibly fast and we think can be a very significant grower for us over time. Infrastructure security business is a low single-digit growing business and is kind of in that range. Prolexic is growing in kind of in its normal range. We haven't had a lot of big attacks or anything like that, that tends to be more, call it, episodic that drives spikes in demand there. So we haven't seen that in a while. That's also slowing a bit revenue has been consistent in that sort of high single-digit, low single-digit, or sorry, low double-digit growth area. It's really just a mix shift right now as we talked about exactly what we're expecting with some of the products that have been in the market for a while, just naturally slowing down. And then the newer products like Guardicore and API taking off but just not at the scale yet to offset the slower growth in the other products.

Got it. That's helpful. In regard to compute, you mentioned last quarter about exiting some legacy revenue streams related to video, such as transcoding and image management. How significant was the impact from that this quarter in compute compared to your expectations when you provided the guidance?

Yes, I'd say it's playing out exactly how we expected. We're still seeing very strong growth in our CIS compute infrastructure services business, and I reiterated our guidance there for very strong 40% to 45% growth in ARR. So playing out exactly as we expected. We said it would take a while for some of this to play out, probably about two years or something like that, but it came in exactly as we had thought.

Perfect. Okay. And then just one last question to revisit the previous discussion on sales and the changes you're implementing, it seems like there will be a greater emphasis on hunters. In terms of the metrics we will observe, where do you anticipate the impact of these sales adjustments will be most evident initially? If you're concentrating on hunters, perhaps the new customer metric might be a key indicator. Please explain how we will be able to see this unfold in terms of the results and metrics we should look for.

Yes. We haven't really provided any metrics in terms of sales productivity but it's something that we'll consider providing you some metrics in terms of how the investments are going, what stage we are in, in terms of adding the capacity. It's both with hunters as well as some specialists for both compute and security. But I think you will just see it manifest itself in stronger revenue growth, but we will consider breaking out some metrics that will be helpful. Just sometimes a booking metric doesn't always translate to revenue cleanly, so that's not the most helpful thing to give. But we'll give us some consideration to try to provide you some metrics that might be helpful.

The next question comes from Patrick Colville with Scotiabank.

This is Sagar on for Patrick Colville. I wanted to better understand the non-delivery opportunity to cross-sell Edgio into the existing base and what you're seeing within that segment specifically?

This is Ed again. You added several hundred customers from Edgio. There's a good opportunity there where Edgio did not have very many security products at all actually, and they didn't offer compute, certainly not the robust compute solutions that we have. Right now, there's not a lot to report in terms of upsell. The first thing you want to do is just make sure the customers are migrated over. They understand who their rep is, get familiar with the services, and work with Akamai, and then we'll start to build campaigns. I'd expect that to take your normal six to nine months to start to build a pipeline, and then we should start to see some potential upsell into that base of additional customers over time.

The next question comes from Madeline Brooks with Bank of America.

This is Kevin on from Madeline Brooks for Bank of America. I want to talk to you about the compute trends. You've noted over time that you expect compute to be kind of a long-term 20% growth driver or grower. What is contributing to this current growth rate that we saw this quarter? And what needs to happen to bring it back up into the 20s?

Yes. So the real driver there is our cloud infrastructure services. As we've talked about and as Ed said, we expect the ARR there to grow 40% to 45% this year. As that number gets a lot bigger, that drives the overall compute number. So that's the number to watch. We'll highlight that as we— that will be driving the bigger compute number.

It sounds good. As a follow-up on security, over the past three years, you've experienced annual growth in the 15% to 16% range for this segment, yet you've provided guidance suggesting a 10% growth for this year. You've already explained the reasoning behind that. I have two questions: What gives you confidence in achieving this 10% growth? And do you foresee any potential for exceeding this 10%? What factors could contribute to that increase?

Yes, I'll start with that. The confidence is driven by the large interest in a big potential market and very strong growth for our segmentation in API security solutions. We believe the ARR of those solutions this year will grow 30% to 35%. A lot of demand for that. We have the market-leading solutions, and now we are adding in firewall for AI — just starting but a lot of customer interest there. I think that has a long runway ahead, even though there's no revenue to speak of yet just signing the first customers now. Having leading solutions in large potential markets that today represent fairly small revenue but are growing at a fast clip gives us confidence that we can be growing the larger security number in the 10% range. Upside would be in part based on products, how fast they get going; and if we can beat the 30% to 35% growth in these products. Ed, do you want to add anything to that?

Yes. The other thing I would say is the business is very highly recurring, mostly subscription-based business. I do break out from time to time if we have some license revenue. You have pretty good visibility in terms of the base; there's not much pricing in your security business, a very sticky product. You don't tend to have a lot of churn. You have a lot of confidence in the underlying business that you have, and it really just comes down to the pace at which you're adding existing customers buying more product and acquiring new logos. If you look at the size of the sales force, what we've produced over the years, we have the capacity to add the type of revenue that we need. As we add get more reps up to speed, we'll have more capacity to sell even more.

The next question comes from Rudy Kessinger with D.A. Davidson.

Ed, could you just on the compute side, could you maybe quantify just what would compute growth have been if not for some of the legacy compute stuff you shifted to partners relative to the 25% in constant currency in Q4?

Yes, sure. Good question. If you remember last time we broke out the sub-number. The cloud infrastructure service and our other compute — our legacy compute, there are like four or five different things that were in there. Some legacy storage, some of the first customers that we had done some custom builds for, some video optimization and optimization, that sort of stuff. That business has grown about 20% last year, and the cloud infrastructure service grew at over 30%. I'd say we're seeing very similar growth rates in the cloud infrastructure service. Where the slowdown is coming, as expected, is in that other legacy stuff. We had a combination of some of the custom deals we did; we did a couple of big custom deals that have largely run their course and they're not growing anymore. Any new applications would go into cloud infrastructure service. Our legacy storage business is starting to decline. We've migrated a little bit of business to partners, so that's playing out exactly as we expected. So really, the decline from 20% to 15% or 25% to 15% is really all in that bucket as we had called out to you. But the cellco infrastructure service is growing really, really fast.

Okay. And on the delivery front, what are you seeing from a pricing standpoint? And in particular, since the Edgio bankruptcy, have you started to see pricing improve on renewals? Do you think 6 or 12 months from now, we might get to a point where price compression could be better than it has been historically, just given us that several low-cost providers in the space for the last few years?

Yes. Good question. We're seeing kind of a mixed bag, as you would expect. I'd say we're seeing some somewhat encouraging signs from some of our larger media customers where the price declines aren't nearly what they used to be, but you don't also have the same type of volume growth that we used to, so that sort of makes sense. There is also, like you pointed out, fewer options there across some of the other — the rest of the base for delivery. It's a mixed bag. It depends; sometimes you have a little bit more price pressure in commerce if there's pressure with the economy and that sort of thing. But yes, I think over time it's possible we'll see a moderation. We are seeing it — it's definitely improved a bit over the last year or so. There’s still room to go there. But this business has always had some element of unit economics of higher volumes at lower unit prices. I expect that to continue but that's certainly one of the drivers to getting this business back to flat and to mid- to low single-digit growth or decline over time.

The next question comes from John DiFucci with Guggenheim Securities.

I guess I think this question is for Ed. As you focus on the faster-growing parts of security, specifically API Security with NeoSec and Guardicore, which I would think are largely channel-focused but different channels than are the traditional Akamai channels? Probably sold through the channel like other traditional security products are sold. Are you taking steps to really ramp that up to drive growth with new channel partners? And if so, because we do a lot of checks and trying to hear more — we'd like to hear more about Akamai in those checks. Can this new path to market more focused path, if that's the case, also be used to sell more of your infrastructure and app security products?

Yes, I'll start... You're absolutely right. They are different than the traditional Akamai channel partners. In fact, most of our large major article sales, API security sales, and new customers to Akamai come through those channels. Examples of partners would be Presidio, WWT, Optiv, GuidePoint, Defense X, IBM, and Deloitte Macnica, OCD, Accenture, Carso. So a variety of partners that are different than the carrier partners we had in the CDN days before. Ed, do you want to add to that? Yes. I would say, John, in talking with PJ, 80% to 90% of our new logo acquisition does come through the channel. I would expect with some of our installed base where we have a strong direct relationship, I would imagine a lot of the API security sales and Guardicore sales will be led by our traditional reps. About one-third — a little over one-third of our business in general comes through the channel, and a little over half of security comes through the channel. I expect that to increase over time. As we acquired some of these companies, both NeoSec and Guardicore, they did come with some new channel relationships that we've obviously continued to grow and build on as a result. So I do expect a lot of our growth going forward to come through the channel.

Ed and Tom, sorry for my voice. Is this an opportunity to address these new channel partners to also sell like your slower-growing parts of security, but they're still needed like DDoS protection and WAF, because others — other partners or other of your competitors that sell those products are making a big push in the channel. So they're getting sort of used to selling that or at least they're starting to talk about it.

Yes, and we do that. Often, it will be the new customer sales would be led with Guardicore or API security, but you're absolutely right. The whole platform, the whole security platform can be then included with that or grown to that. And those partners are very good at it. So I think you're absolutely right and we're seeing that.

We have time for one last question. It will come from James Fish with Piper Sandler.

This is Kate on for Fish. So just one on the security weakness, was that more on the new side of the — was that more on the new side or the expansion side of existing customers?

Yes. I wouldn't refer to it as weakness. It came in as we expected it. So we're not viewing it as weakness. It was pretty much right in line. In terms of the impact within a quarter, you got to remember a lot of the business is recurring. The bulk of the revenue is already spoken for with existing commerce. Most of the growth, as I said, comes from the — into newer sales coming from API security and Guardicore as we expected. That doesn't have a huge revenue contribution in the quarter in which you sell it because it does take maybe a month or so to get it up and running and revenue generating. Like I said, I think things turned out just as we expected, and the normal mix of new customer versus upsell into the base was pretty similar to what we see in a typical quarter.

Thank you. That concludes our question-and-answer session. I would now like to turn the call back over to Mark Stoutenberg for closing remarks.

MJ, you can please end the call now.

All right. This concludes our presentation. Thank you for attending today's conference. You may now disconnect your lines.