Q2 2026 Earnings Call Transcript

Good afternoon, and welcome to the F5, Inc. Second Quarter Fiscal 2026 Financial Results Conference Call. [Operator Instructions] Also, today's conference is being recorded. If anyone has any objections, please disconnect at this time. I'll now turn the call over to Ms. Suzanne DuLong. Ma'am, you may begin.

Hello, and welcome. I'm Suzanne DuLong, F5's Vice President of Investor Relations. We are here to discuss our second quarter fiscal year 2026 financial results. Francois Locoh-Donou, F5's Chairman, President and CEO, and Cooper Werner, F5's Executive Vice President and CFO, will be making prepared remarks on today's call.

Other members of the F5 executive team are also here to answer questions during the Q&A session. Today's press release is available on our website at f5.com, where an archived version of today's audio will be available through July 27, 2026. We will post a slide deck accompanying today's webcast to our IR site following this call. To access the replay of today's webcast by phone, dial (800) 770-2030 or (609) 800-9909 and use meeting ID 6076834. The telephonic replay will be available through midnight Pacific Time, April 29, 2026. For additional information or follow-up questions, please reach out to me directly at s.dulong@f5.com.

Our discussion today will contain forward-looking statements, which include words such as believe, anticipate, expect and target. These forward-looking statements involve uncertainties and risks that may cause our actual results to differ materially from those expressed or implied by these statements. We have summarized factors that may affect our results in the press release announcing our financial results and in detail in our SEC filings. In addition, we will reference non-GAAP metrics during today's discussion.

Please see our full GAAP to non-GAAP reconciliation in today's press release and in the appendix of our earnings slide deck. Please note that F5 has no duty to update any information presented in this call. Before I pass the call to Francois, I am pleased to announce that F5 will be hosting an Analyst and Investor event in New York on Thursday, May 28, 2026. Details about the event will be provided in a press release soon. I'll now turn the call over to Francois.

François Locoh-Donou: Thank you, Suzanne, and hello, everyone. Our team delivered another robust quarter with 11% revenue growth. Product revenue grew 22%, marking our seventh consecutive quarter of double-digit product growth. This includes strong 26% systems revenue growth and 17% software revenue growth. Hybrid multicloud has become a strategic architecture, and it is increasing demand across F5's core markets. Customers are rapidly scaling their digital infrastructures to improve resiliency, meet data sovereignty requirements and get ready for AI. Our strong Q2 performance reflects those dynamics and F5's alignment with where customers are headed. We captured robust international demand for digital sovereignty initiatives. We also converted hybrid multicloud adoption into meaningful systems and software growth. We capitalized on heightened demand for best-in-class security solutions, and we built on AI momentum with another standout quarter for AI wins. As a result of our strong growth and our proven operating model, we delivered 14% non-GAAP earnings growth and a record $348 million in free cash flow. The powerful combination of secular and cyclical demand trends is providing strong Q3 visibility and a growing pipeline. As a result, we are raising our fiscal year 2026 outlook to reflect revenue growth of 7% to 8%, up from 5% to 6% previously. Cooper will elaborate on our outlook in his remarks. Our outlook for stronger growth is reinforced by what we are seeing in the market. We see three forces significantly reshaping how our customers operate, hybrid multicloud adoption, threat landscape expansion and AI inference inflection. First, hybrid multicloud adoption. Workloads now span on-premises, private cloud and multiple public clouds. Our research shows more than 90% of enterprises run hybrid multicloud today across an average of 19 locations. Organizations need flexibility, resiliency and digital sovereignty in every environment, and they are investing to support these demands. Second, threat landscape expansion. As AI models become more capable, attackers are using them to launch attacks against production applications at higher volumes and with greater variation than traditional defenses were designed for. Our customers see this and they are responding. They are deploying more application security and prioritizing best-in-class defenses. The era of checkbox security is over. AI applications require best-in-class security to match both the volume and the sophistication of AI-driven attacks. Third, the AI inference inflection. Organizations are connecting their applications and APIs to AI models and inference calls are becoming a regular part of how applications run. Our research shows 78% of enterprises run inference themselves using more than seven models on average. Organizations are standardizing on a new architecture with models distributed across the data center, the cloud and the edge. And the next shift is already underway. AI agents are moving into production and enterprises are adapting their applications for agent interaction. This is driving more compute, more data delivery and more security to protect inference. These three market forces are driving demand across our business. Because of accelerating hybrid multicloud adoption, we are taking an already strong refresh cycle and leveraging it into significant opportunities for expansion, competitive displacement and platform consolidation. I will double-click on each of these, spotlighting customer examples from the quarter. With this refresh, we are seeing a Refresh plus dynamic that is different from prior cycles. Customers are deploying higher performance, higher capacity F5 systems as they upgrade their data centers to support modern applications, digital resilience and sovereignty and AI. And as customers refresh, we are capitalizing on that moment to attach new use cases, expanding our footprint and growing overall wallet share. For example, this quarter, a large healthcare services organization started with a life cycle refresh across hundreds of legacy systems. As the project progressed, they expanded the scope to support an AI-driven consumer engagement platform. F5 became the control point for secure, low-latency traffic and data movement across applications, storage and their GPU server environment. That gave the customer a more resilient foundation for both sensitive internal workloads and new AI interactions at scale. Our deliberate investment in hybrid multicloud solutions is translating into market share gains. We are winning customers from competitors who did not build the same breadth and depth of capabilities across on-premises, software and SaaS. In Q2, we displaced a long-standing incumbent at a Fortune 100 energy company whose environment had hit scalability limits. The customer needed a platform that could scale into cloud while maintaining strong on-premises performance. Their incumbent provider was unable to serve workloads in hybrid multicloud environments. F5 modernized traffic management and simplified operations, improving reliability and creating a clean path for long-term cloud adoption. Hybrid multicloud customers require stronger performance and security with fewer tools and simpler operations. We are replacing point products with a unified approach that improves performance and security and is easier to operate at scale. For example, during Q2, an energy and utilities provider, an existing BIG-IP customer needed to secure APIs with better visibility and automation across their data center, cloud and edge environments. They selected F5 Distributed Cloud Services to simplify their approach and standardize API protection across their full footprint with simpler management. Moving on to threat landscape expansion. The pace and scope with which the threat landscape is expanding is driving demand for best-in-class application and API security, both on-premises and across cloud environments. For example, this quarter, a software and managed service provider needed to standardize application and API security across a rapidly expanding hybrid multicloud estate built through acquisitions. They lacked a consistent way to enforce front-door and API protections across their multiple public cloud environments and on-premises. With F5, they deployed a single policy and management layer with security enforced locally in every environment, supporting strict privacy, audit and healthcare requirements. F5 enabled faster regional expansion with stronger security and improved data sovereignty alignment. Finally, the AI inference inflection is driving demand for F5. We are seeing this indirectly through hybrid multicloud adoption and the requirements that come with it. We are also seeing it directly through our three primary AI use cases. With our industry-leading traffic management, we are winning new AI insertion points, including AI data delivery and AI factory load balancing. And we are capturing AI runtime security wins, protecting AI applications, APIs and models from emerging threats such as model abuse, data leakage and prompt injection. In an AI data delivery win, a global payments company needed a more resilient way to move rapidly growing AI data between storage and compute as they scale the training and retrieval workloads. F5 improved performance and resiliency while displacing both an in-house solution and a competitor, positioning us at the center of the customers' AI infrastructure strategy. In an AI runtime security win, an industrial automation firm needed a scalable way to assess risk and govern a growing number of AI applications and models. They chose F5 based on the depth of our red teaming insights and strong integration with their existing security stack. In an AI factory load balancing win, a major manufacturer an existing F5 customer needed to support operations and established a digital twin of their manufacturing environment for simulation and optimization. They deployed BIG-IP as the production traffic layer across their GPU server environment, improving availability and offloading encryption. Taken together, these wins underscore two things. The forces reshaping our customers' environments are real and F5 is well positioned to capture them. Staying ahead of the pace of change requires relentless innovation. In Q2, we brought multiple new capabilities to market, strengthening our leadership in application delivery and security for the AI era and driving greater value for customers. We introduced AI-powered capabilities in Distributed Cloud WAF, replacing manual policy tuning with automated outcome-based threat blocking. Our F5 training model helps customers stay ahead of increasingly sophisticated AI-driven attacks that are growing in both speed and complexity. We launched Agentic Bot Defense, extending our industry-leading Bot Defense to autonomous AI agents, a new and fast-growing category of traffic. The result is that customers can confidently adopt Agentic AI while ensuring only verified trusted agents reach their applications. We released F5 AI Remediate, which closes the loop between our AI Red Team and AI Guardrails products. It collapses the path from vulnerability discovery to runtime protection from days or weeks into minutes. And finally, we launched F5 Insight for ADSP, providing deeper visibility across application estates. The result is that customers can identify and resolve issues faster with less guesswork. We are innovating so customers can run faster, stay protected and simplify their hybrid multicloud and AI environment. And we are accelerating that innovation by rapidly integrating AI into our solutions to create practical capabilities customers can deploy quickly. That innovation engine is also sharpening our view of what's next. As we look ahead, we have conviction in the power and durability of hybrid multicloud, the expanding threat landscape and inflecting AI inference as demand drivers for F5. We look forward to digging deeper into these drivers and our expectations for how they will shape F5's longer-term growth outlook at our May Analyst and Investor event. Now I will turn the call over to Cooper, who will walk through our Q2 results and our outlook. Cooper?

Thank you, Francois, and hello, everyone. I will review our Q2 results before I provide our guidance for Q3 and update our outlook for FY '26. We delivered a strong Q2, growing revenue 11% to $812 million with a mix of 51% product revenue and 49% services revenue. Product revenue totaled $411 million, increasing 22% year-over-year, while services revenue of $401 million grew 2% year-over-year. Systems revenue totaled $226 million, up 26% over Q2 FY '25. Our software revenue of $184 million grew 17% year-over-year.

Subscription-based software revenue totaled $165 million, up 20% year-on-year, representing 90% of our Q2 software revenue. Perpetual license software totaled $19 million, down 4% year-over-year. Revenue from recurring sources contributed 70% of our Q2 revenue. Our recurring revenue consists of our subscription-based revenue and the maintenance portion of our services revenue.

Shifting to revenue distribution by region. Revenue from the Americas grew 3% year-over-year, representing 50% of total revenue. Both our EMEA and APAC regions delivered very strong quarters. EMEA grew 22%, representing 32% of revenue.

APAC grew 19%, representing 18% of revenue. Looking at our major verticals, enterprise customers contributed 66% of Q2's product bookings. Government customers represented a strong 24% of product bookings, including 8% from U.S. Federal.

Finally, service providers contributed 9% of Q2 product bookings. Our continued financial discipline contributed to our strong Q2 operating results. GAAP gross margin was 81.4%. Non-GAAP gross margin was 83.7%.

Our GAAP operating expenses were $482 million. Our non-GAAP operating expenses were $406 million. Our GAAP operating margin was 22.1%. Our non-GAAP operating margin was 33.8%.

Our GAAP effective tax rate for the quarter was 21.9%. Our non-GAAP effective tax rate was 21.5%. Our GAAP net income for the quarter was $148 million or $2.58 per share. Our non-GAAP net income was $223 million or $3.90 per share, reflecting 14% EPS growth from the year ago period.

I will now turn to cash flow and balance sheet metrics. We generated $366 million in cash flow from operations in Q2 and free cash flow of $348 million, both records, highlighting the strength of our operating model. CapEx was $18 million. DSO for the quarter was 47 days.

Cash and investments totaled $1.46 billion at quarter end. Deferred revenue was $2.12 billion, up 10% from the year ago period. In Q2, we repurchased $100 million worth of F5 shares at an average price of $269 per share. We had $522 million remaining on our authorized share repurchase program as of the end of the quarter.

Finally, we ended the quarter with approximately 6,500 employees. I will now speak to our outlook and guidance, beginning with Q3, followed by our full year view. We expect the market trends we've outlined, hybrid multicloud adoption, threat landscape expansion and AI inference inflection to drive strong demand for our products and services in the second half of FY '26. We expect Q3 revenue in a range of $820 million to $840 million, reflecting approximately 6.5% growth at the midpoint. We expect non-GAAP gross margin in the range of 82.5% to 83.5%.

We estimate Q3 non-GAAP operating expenses of $406 million to $418 million. We expect Q3 share-based compensation expense of approximately $68 million to $70 million. We anticipate Q3 non-GAAP EPS in the range of $3.91 to $4.03 per share. Turning to our fiscal year 2026 outlook.

With continued strong close rates in Q2 and strong pipeline creation into the second half, we are raising our FY '26 outlook. We now expect FY '26 revenue growth of 7% to 8%, up from our prior outlook of 5% to 6%. We continue to expect mid-single-digit software revenue growth, double-digit systems revenue growth and low single-digit services revenue growth for the year. Our gross and operating margin outlook for FY '26 is unchanged.

We expect FY '26 non-GAAP gross margin in a range of 82.5% to 83.5%. On a modeling note, we expect higher component costs, primarily related to memory will cause gross margin to step down sequentially from Q3 into Q4. We expect non-GAAP operating margin in a range of 34% to 35%. We now expect our FY '26 non-GAAP effective tax rate will be in the range of 20% to 21%. Reflecting the strength of our second quarter and our increased revenue outlook, we now expect FY '26 non-GAAP EPS in a range of $16.25 to $16.55, up from the prior range of $15.65 to $16.05. Finally, we expect our full year share repurchase to be at least 50% of our free cash flow.

I will now pass the call back to Francois.

François Locoh-Donou: Thank you, Cooper. Looking ahead, our strengths are well matched to the secular shifts transforming IT infrastructure, hybrid multicloud adoption, threat landscape expansion and AI inference inflection. We expect these trends to support continued growth for F5 in fiscal 2026 and beyond. F5 is built for hybrid multicloud and the AI era. We deliver and secure every app and API anywhere with one unified platform across on-premises, multiple public clouds and the edge. Our application delivery and security platform reduces complexity. Customers get centralized security, high-performance delivery and consistent policy without stitching together point products. And we provide a control point for traffic, APIs and data flows as applications and AI become more distributed. Operator, please open the call to questions.

[Operator Instructions] We'll take our first question from Tim Long at Barclays.

Yes, one question and one clarification. On the software side, it looks like it was a pretty good quarter and you're keeping the mid-single-digit for the year. So I know sometimes these are on 3-year cycles given the term. So maybe just touch a little bit on why not a little bit more of a raise there after a pretty solid growth quarter.

And are you still looking at potential acceleration on that number into next year? And then after that, I'll come back with a follow-up.

Thanks, Tim. This is Cooper. Yes, I'll take that. So we did have a good growth quarter in Q2. I would say it was right where we expected it to be for the quarter.

You're right, we do caution against kind of over rotating on any individual quarter's reported revenue growth rate. The second half of the year is where we have a more balanced growth expectation for the year. And just based on where we're at with the renewal base, we continue to expect to perform as we had seen it shaping up for the year. And so that's where we're still at the mid-single-digit growth rate for the year, but all trends look very healthy.

And then, yes, as we look ahead to next year, we do expect to see an inflection in the growth rate. We're continuing to see strong trends around consumption rates across that renewal base, and we have a larger base coming up for renewal next year. And so with the expansion we would anticipate against that larger renewal base, we feel pretty confident about a higher growth rate into FY '27.

Okay. Great. And then if I could, on the AI front, a lot of different applications, a lot of activity. Maybe you could help us a little bit with some benchmarks or some metrics, how do we frame the success revenues, orders, customers? How should we look at it? Any data points you can give us as far as the scale and the traction you guys are seeing on the customer side?

François Locoh-Donou: Yes, Tim, it's Francois here. The -- so what we're seeing in AI, Tim, is that enterprises are now putting AI into production. And what the term we use is inferencing. And that's creating substantial opportunity for F5. We've talked about three big areas where we see opportunity. The first one is in hardening data pipelines between data stores and AI models, a use case we call data delivery, and we're seeing growing demand for F5 in these use cases. We're also seeing growing demand in securing AI in runtime. So both AI applications and AI models increasingly require security that is tailored for AI models that traditional security solutions do not address. And we also address load balancing, AI factory load balancing, which is a third area where we're starting to see growing demand. If you look at all that, we -- if you look at the first half of the year, -- we had approximately $50 million in sales in the first half of the year on these use cases. That's up more than 200% year-on-year. And we're now approaching about 100 customers that are using F5 for their AI use cases. That's why we have a bit of a conservative estimate because those are our customers from whom we absolutely know that they are using F5 for these AI use cases. We believe there are other parts of the business where we're getting indirect benefits from customers getting ready for their AI infrastructure, but those are harder to quantify, harder to count. So the ones I'm sharing with you are ones where we actually have the data and can attribute it directly to these use cases. So enterprise AI is one of the big trends that's fueling some tailwinds in our business. And hybrid multicloud and an expanding threat landscape are the other two very significant trends we're seeing.

We'll move next to Samik Chatterjee at JPMorgan.

Francois, pretty strong quarter. You're raising the guide for the year as well and getting ready, it seems to give us a more longer-term view of the business. Just trying to get sort of how you're thinking about sustainability of the high single-digit growth as you look forward given that you did sort of have a softer year in software this year, but you also have the hardware sort of tailwinds in relation to end of support for some of your products. Like how should we think about sustainability of these growth rates as you look forward beyond this year?

How are you thinking about that, if you can help us? And then I have a follow-up.

François Locoh-Donou: Yes, Samik. The -- I mean, as it relates specifically to software, I think Cooper touched on it, where we expect stronger -- even stronger software growth next year than this year. But let me step back a little bit and talk about the overall business, Samik. We are seeing a couple of things. One, of course, is we are seeing a very strong refresh cycle and the refresh cycle, by definition, is cyclical. But we are also seeing three secular trends that we think are very durable and that are just growing and accelerating our business. The first one is hybrid multicloud. We went -- we've been talking at F5 about hybrid multicloud for several years. If you look at the past few years, hybrid multicloud was by default. Customers needed the flexibility to put their application in different environments. But now we're seeing it being more of a strategic architecture that is by design and customers are implementing that for digital sovereignty reasons to be able to rely not just on big public clouds, but local cloud alternatives or on-premise environments. And they're also implementing digital hybrid multicloud architectures for resilience reasons. And increasingly, AI is also pushing customers towards these hybrid multicloud architectures. That is a secular trend, Samik, that is there for the long term, and that is providing substantial tailwinds for the business that we believe are durable. And then the other trend that we're seeing is the threat landscape is expanding. And so what we're seeing is customers are having more frequent attacks that are more sophisticated attacks because of AI. There was a report published recently that showed the increase in web attack year-on-year was up 77%. The increase in bot attacks were up 150% year-on-year. And all of that means that our customers have more apps to protect because they're apps, they're APIs, they're now AI models, both on-premise and in the cloud. And with the frequency and the sophistication of attacks increasing, there is a need for best-in-class application security solutions. And that is right where F5 has been focused, and we are seeing that demand in our business. To give you a couple of data points, in our distributed cloud services platform, for example, we saw this quarter, the number of customers choosing F5 for web application firewalls are up 62% year-on-year. The number of customers choosing F5 for API security is up 54% year-on-year. And for bot defense, it's up 33% year-on-year. So you can see these trends of increasing attacks, our customers responding, needing more application security solution that are best-in-class and coming to F5. So these are important trends. We think they are durable, Samik. And therefore, we think the inflection we're seeing in our business is likely to continue.

Got it. Got it. And Francois, maybe I'll follow up on that aspect itself on sort of the attacks that customers have to be ready for. Have you seen any change in engagement or even a step-up in engagement following all the discussion that enterprises have to deal with in relation to Anthropic's Mythos model and sort of the vulnerabilities that they've highlighted.

Are you seeing any step change in your engagement with customers on the security front? How are you sort of looking for -- looking to your customers trying to address some of those issues?

François Locoh-Donou: Thank you, Samik, for the second one -- the second question. Yes, we are seeing a step change, Samik. We've had a number of conversations over the last several weeks with customers. If you think about it, we are now in an era where the window of time for an enterprise to patch their applications has closed as we have AI models that are very powerful and can now find and exploit vulnerabilities in any application almost in real time. And so there are a couple of implications for that. The first is given if you don't have a significant window of time to patch your applications, you are going to rely more on runtime security and specifically runtime security that is protecting the front-door of your applications. That's precisely where F5 has focused. And we're having conversations with customers who are sharing with us that they're going to have to rely on us even more than they had in the past. The second implication is that we believe that all security is going to be AI-powered. Static security, static signatures are really not going to be able to cope with the power and the speed that these new models have in terms of creating exploits. And so this is a shift that we saw coming. We have been investing in AI-powered security for a while now. Just this quarter, you may have seen this, we released our AI-powered web application firewall. We also released our Agentic Bot Defense solution. And so over time, our entire portfolio is going to be AI-powered. But we are basically already fighting AI with AI, and that, we think, is a significant shift to our customers. And probably the other step change for our customers, it's a trend that has been happening, but I think the new era really accelerates this is the consolidation towards platforms. If you're a customer that's operating in multiple environments and 95% of our customers are operating into hybrid and multicloud environment, the era of having a point product solution in any one of these environments really just creates complexity that you don't want to have to deal with if you have to try and really patch your systems very quickly. And so I think we're going to see more customers move towards platform and the breadth of our portfolio can really help them simplify their operations. So those are three of the implications that we see with this change, and we're seeing that in our conversations with customers already over the last several weeks.

We'll go next to Simon Leopold at Raymond James.

I wanted to ask about, I guess, a phenomenon that may be occurring. And what we've heard is that some customers may be showing a preference for your hardware solutions based on the performance, the relative performance that perhaps the total cost of ownership of implementing software is actually more expensive than the relative hardware. I'm wondering if you're seeing this shift and that might explain some of the relative growth between your hardware and software.

François Locoh-Donou: Well -- Simon, there is -- first of all, we are seeing, in fact, a number of customers that are recommitting to hardware. I wouldn't say that it's just about performance. Performance is a factor. There are a number of reasons for customers to want to be doing that. I think one of those reasons is a lot of customers are modernizing their data centers and wanting to have strong on-prem infrastructure with strong performance in their data centers. And we have seen over the -- in the first half, just to give you a data point, we generated about $60 million in sales from customers who had previously kind of stopped buying hardware and recommitted to hardware. So we are seeing this phenomenon of customers recommitting to hardware. But if you expand from that, what we -- because we delivered 22% growth on hardware this quarter and 17% growth on software. The broader trend we're seeing, Simon, is that the hybrid multicloud is really what's driving customers to both modernize their data center and continue to invest in software to have the flexibility to be able to deploy the same solution, the same software stack from F5, either on-prem or in public clouds. And so yes, at this moment, there is a very strong momentum on hardware, but we continue to see customers wanting to have the flexibility of software or subscription-based software to be able to deploy license across their environment.

And just as a quick follow-up, please. Could you update us on any progress around the engagement and discussions you've had with NVIDIA? You've talked about that on earlier calls. I'm not sure that you've updated us on the prepared remarks. So any updates you can offer?

François Locoh-Donou: Yes, of course. So yes, we have -- as you know, we have developed an integration with NVIDIA where we have been able to basically refactor our software to work in ARM architectures and specifically work on NVIDIA BlueField technology. We've done a lot of work with NVIDIA over the last 18 months. As of December, we have now been formally put into NVIDIA's reference architecture. Since then, there have been a number of tests, including third-party tests to test the efficiency gains from this integration. Those tests have validated that basically the integration of F5 software on these NVIDIA DPUs helps AI factories generate 30% to 40% more token for a certain amount of GPUs. And we are now taking that value proposition to market, and we are involved in a number of proof of concepts and trials around this technology and this integration. I would say that what we are seeing is that a number of customers who are building AI factories are early in terms of sophistication in that their first priority is to get these AI factories, these GPU farms up and running, get them running -- get them working, get these Kubernetes clusters to work. That takes quite a bit of technical sophistication and customers are really focused on that. And for those who are really providing GPUs as a service, really, the goal initially is to get these GPUs to work and to be able to provide that to their customers. I think the issue of making those GPUs more efficient is the issue that comes next. And I think as more and more customers go to inferencing, we think that this value proposition is going to resonate.

We'll go next to Matt Hedberg at RBC Capital Markets.

Based on a lot of our conversations with partners and customers, we think F5 sits at really a critical junction in really this hybrid cloud infrastructure build-out and increasing AI app traffic. In your prepared remarks, you talked about sort of your role in this evolving threat landscape. And I'm curious, you have a lot of security solutions now, but are you hearing customers pull you into additional use cases? Or you're in such a unique spot of the traffic flow with the lens that you see.

Are there other opportunities for you to add either further security capabilities in this kind of this new AI era?

François Locoh-Donou: Well, absolutely. We -- so a couple of things. I shared earlier that in this new era, runtime security and specifically securing the front-door of applications is going to be even more important than it was in the past and especially for folks who have invested like us in best-in-class application and API security. So the first thing we're seeing is really strong growth in web application security, in API security and in bot security. We're also seeing API discovery and whether on-prem or in the cloud being a growing use case with more and more customers really now worried about knowing where all their APIs are and being able to protect them. Now when you go to AI, we also have now a new attack surface, which is these AI models and these agents both of which will be using more APIs and our customers, of course, will need help discovering and securing them. But we've also introduced in the last few months, AI Guardrails, which is AI Red Team and AI Guardrails. So technologies that help our customers both detect vulnerabilities in their AI models and mitigate these vulnerabilities. And we have introduced a product called AI Remediate that automates the process of creating mitigation for these vulnerabilities. All of these are new use cases in security that are going to grow as our customers deploy more AI models in production. So we are seeing new use cases and new opportunities to insert F5. Security, I think, is a very significant opportunity. But as I said earlier, we're also seeing that opportunity in delivery, specifically in data delivery for AI.

That's great. That's great. And then Francois, the other thing you touched on in your prepared remarks was you're starting to see AI inferencing inflect with your customer base, which is -- it makes sense given some of the AI models, the innovation that we're seeing. And I guess it feels to me like the broader sort of non-AI native cohort of customers are becoming increasingly AI-leaning.

Is there a way to talk about how early we are in that? And is this part of a multiyear really inflection? Could we be talking about this inferencing inflection 2 years from now, for instance?

François Locoh-Donou: Yes. So Matt, I think the customers who are today really focused on -- have already started worrying about AI security and protecting AI models and AI applications that have new types of vulnerabilities like prompt injections, model abuse, et cetera. Those customers are a small minority, typically the largest customers in any vertical, the customers perhaps have a lot of sophistication in security, financial services companies, very large technology companies. But today, it's a small minority of the universe of customers we serve. And I think that number of customers is only going to grow over the next couple of years as more and more customers actually implement AI in inference. So I think we are just at the very start of this trend and the number of models for inference and agents will dramatically increase over the next couple of years.

We'll go next to George Notter at Wolfe Research.

If I look back, you guys have been raising prices pretty conservatively, I think, once per year. Obviously, there's some more memory costs here. You mentioned in the context of gross margins. But any thoughts about raising prices a bit more aggressively or a bit more frequently?

And I think if I look back historically, you guys also talked about kind of balancing price increases with the opportunity to gain share. And I'm just curious like on the share side of things, are you making progress there? Are there any metrics you can give us in terms of logos or incremental revenue or share that you can point to that kind of reinforce the idea that you guys are winning share?

Yes, George, thanks. This is Cooper. I'll start on the pricing. So we do have kind of an annual pricing review that we do.

Typically, it's in our Q2 where we make price adjustments to factor in the innovation we've been bringing to market, and that's part of our ongoing playbook. We've also been closely monitoring what's been going on with memory and SSD pricing, which has just been accelerating through the year and really kind of had a big step-up in Q2. And that's something that we continue to look at price adjustments to pass through some of that impact through to offset the impact on our gross profits. It's a combination of price adjustments and discount discipline. And that's something that we have to stay really agile with, and we'll continue to kind of monitor that and make those adjustments on more of a one-off basis tied specifically to the rising cost of memory.

But then long term, as we think about share, what we've seen, particularly recently is our competitive takeout rate has gone up pretty materially. And I think that really speaks to the hybrid multicloud adoption that our customers are seeing where we're really the only vendor in the space that can support the customers' applications in any environment. And that's really been resonating, particularly recently with the evolving threat landscape as customers are looking for a platform approach to resolve a number of complexities in their environment, and they've been coming to F5. And so we've been seeing a lot of share gain in that regard.

Our next question comes from James Fish at Piper Sandler.

Great quarter. Maybe to give Francois a bit of a break on especially the AI side, Cooper for you. I'm going to get asked this tomorrow. So on the 2-point raise to guide here for the year, it looks about 1 point is from this past quarter's upside.

Are you actually passing through memory much at this point? And what are you guys assuming from memory prices kind of in the back half of the year? Do you have enough supply still lined up given the outperformance of hardware? And how far along with, you are on migrating the DDR5 from DDR4 in particular?

Yes. Okay. I'll try to make sure I hit all three. But if I forget, please let me know.

So in terms of our revenue guide for the year, that doesn't really contemplate new pricing adjustments. I just referenced the work that we're doing around that. But any pricing adjustments that we did are more likely going to flow through into FY '27 just based on where we are with the cycle. So it is something that we continue to look at, but it's not really a significant component to our back half revenue guide for the year.

In terms of supply availability, yes, we feel pretty good about our near-term visibility. We've really been out in front of this, and I'm really proud of our manufacturing team for identifying this as an issue going back to kind of mid-FY '25, where we increased our build forecast. We extended the length of our build forecast, and we took on additional supply on components that we thought might have more constraints. And so that's allowed us to secure the memory that we need not just for the revenue outlook that we had at the time, but for the upside we've been delivering over the last 6 quarters or so.

And so we feel pretty good, at least for the near term. Now you get it longer term into FY '27, the build forecast we have out there are within our needs for what we would expect to do on the high side for our systems business. Obviously, the visibility 4 or 5 quarters out is not as strong as it is in the near term. But right now, we feel pretty good with where we sit.

And then the last question -- DDR4. So yes, so the current appliance lineup that we have leverages DDR4. Future appliance cycles will be on newer technology. We haven't discussed the timing of those, the next generation of appliances.

Fair enough. If I could follow up, just because if I look at your billings, you had a really strong deferred here, especially on the current side. What are you guys seeing with any net pull-in of demand or buildup of product backlog here as a lot of us here will kind of be reminiscent of the supply chain crisis just a few years ago and that -- this would be about the time you guys would start to see a buildup in product backlog.

Yes. So just to be clear, backlog does not show up in our deferred revenue. So our deferred revenue strength is almost entirely tied to our services business where we have maintenance renewals. And we saw the strength both on short-term and long-term deferred maintenance is actually a little bit higher on the long term.

And we did see some customers that were doing multi-year renewals. I'm certain that some of them are getting in front of perceived risk around price increases as they're working with other vendors. And so that is playing out to an extent, I would imagine, on the maintenance side. But the growth is not tied to product orders.

Next, we'll move to Meta Marshall at Morgan Stanley.

A couple of questions for me. One, just on the continued strength that you're seeing in EMEA and particularly around data sovereignty, just how much further or kind of are there initiatives that you guys are taking to kind of capitalize on that opportunity? And then maybe second, a very clean competitive landscape kind of on the ADC front, just as a lot of those vendors have kind of fallen by the wayside. But just as you move more on to the security space, just what are you seeing in terms of the competitive landscape there or the chance to gain mind share there?

François Locoh-Donou: Thank you, Meta. On EMEA we think the trend that we're seeing there is durable. In fact, we saw an acceleration in that trend this quarter. A lot of the customers, whether it's government agencies, the defense sector, of course, all regulated industries, including financial services, all have a strong push for digital sovereignty. That implies in a lot of cases, modernization, reinvestment in data centers and also creating consistency of security and delivery across their hybrid multicloud environments. We're seeing an interesting trend there where when customers went to the public cloud, they created a separate team between public cloud and on-premise environment. And now that they're kind of coming back and creating true hybrid multicloud architectures, they are merging those teams together, and it's creating more opportunities for the provider that can cover their needs across on-prem and public cloud with a single platform. That trend, we think, is going to continue in EMEA. We're leveraging it more. We have increased our coverage -- our field coverage in EMEA, and we'll probably continue to do that in the future and probably accentuate our focus there on the defense sector because we're seeing significant spend in defense in EMEA. As it relates to the landscape -- the competitive landscape in security, we are focused, as you know, on runtime application and API security. And in that space, we are seeing substantial growth, both for on-premises requirements and cloud requirements. Our differentiation is really the ability to serve both environments with an extensive security portfolio that includes application firewall, securing APIs, securing against bot, securing against DoS attacks. And frankly, none of our competitors, whether it's for application security or AI security are really hybrid multicloud. And so the more we see customers embracing these architectures and needing a solution for both on-prem and public cloud, we are alone in that category and have a very, very strong value proposition. There are a few examples that I mentioned in my prepared remarks where customers needed to secure APIs or they need to secure their applications for a solution that worked both on-prem and in the cloud, and they came to F5. That consistency is more important than ever, and that's where we are focused. And we're going to continue to invest there. One of the highlights of the quarter for me, and I'm really proud of our product team for the work that we did this quarter was incredible innovation in security. We released our AI-powered WAF. We have already a significant interest for that, a new solution for Agentic Bot Defense, which is really important now to understand which agents are authorized to access a model and which agents are not. We innovated on our AI security solutions with F5 AI Remediate, a new solution. We introduced new solution that is AI-powered F5 Insights. We brought API Discovery on-premise with our BIG-IP solution. So a lot of innovation that is accelerating, is in part -- by the way, because we're also leveraging AI to do that. But I'm excited about the place we're at as the company that invested in hybrid multicloud architectures, I think, ahead of everybody else and is now starting to reap the reward of that and now doubling down on our innovation, accelerating the pace of our innovation to capture a growing landscape of opportunities in front of us.

We'll move next to Jeffrey Hopson at Needham.

I just wanted to follow up on the memory situation and the gross margin implications. You gave guidance for the last quarter to have a step down from 3Q to 4Q. Just curious if there's any more color on the magnitude of that step down. I think I had like around 150 basis points.

And is this just a function of memory bought today takes about 2 quarters to flow through, and that's kind of dynamics at play?

Yes. Thank you. So yes, that's the dynamic. So we've -- as I referenced earlier, we had taken a pretty extensive position early.

And so we've been able to kind of mitigate any impact up through the first half of this year, but we're now starting to see some of the later purchases that we have been doing at higher price points is -- are going to start to flow through into the model. And it will start to flow in, in Q3, but it will be kind of more at full run rate in Q4. It's an incredibly dynamic situation with memory pricing. It's -- we're trying to get the signals on what it could look like in the next few quarters. Our expectation is that there will be relief several quarters out.

But right now, for at least through the better part of FY '27, we would expect memory prices to stay elevated.

Got it. And maybe just on the U.S. Federal side. It's been a couple of really nice quarters. Maybe just any additional information on the dynamics that are going on in U.S. Fed?

François Locoh-Donou: Yes. Generally, the dynamics are strong. We had a strong Fed quarter. And I would actually expand that beyond U.S. Fed to the global government sector in the first half was really strong. I think you're seeing that. That's not, I would say, just an F5 trend. I think you're seeing that generally defense spending across the globe has been growing, and we are a beneficiary of that trend, in part because generally, defense customers are investing more in security, in part also because those customers are very hybrid multicloud. We have a number of customers in the defense sectors that want air-gapped environment. Sometimes they want to leverage cloud as well, but a lot of them want air-gapped environment in their own data centers. We have been making investments for that opportunity, and we're seeing the benefit of that today. So I think the Fed has been strong for us. But globally, government spending has been strong, and I think will continue to be for the next several quarters.

We'll move next to Amit Daryanani at Evercore ISI.

This is Ketan on for Amit. I guess services growth at 2% was fairly muted. Can you maybe just touch on what's happening there and maybe your updated thoughts on how to think about it in the long term?

Yes. I'll start. I would say, ironically, I think this is tied to a good news story, which is the strength that we're seeing with the refresh, and this is kind of the dynamic that we've seen with past refresh cycles. When you see a strong refresh in the very near term, it has a little bit of a headwind to the services business.

And part of that has to do with you're replacing legacy appliances that have been carrying service for a number of years. And as those come out of the system, and then you backfill with the new appliances, there's a little bit of a lag on the maintenance revenue stream. Conversely, when we've had periods where customers are sweating assets, that's where you saw some strength in the maintenance revenue. So the longer-term picture is that the refresh has been very strong, and it's a refresh plus expansion story.

And what we're seeing is that we're getting better retention of that footprint than we had in prior cycles. Ultimately, that's going to be a great story for services because with the larger footprint that you get maintenance revenue against, you're going to see a better revenue outcome. But in the very immediate term, as customers are making the transition, it's a bit of a headwind on the maintenance revenue.

Next, we'll move to Tal Liani at Bank of America.

I think everyone is trying to basically get to the same question, whether this is finally a sign that AI is showing its impact on the company's growth or whether this is just a refresh story that is temporary. And the question I have is why are we seeing -- I think you touched on some of it, but why are we seeing the growth only outside of the U.S. or less in the U.S.? Meaning U.S. is leading AI. Out of $80 million growth year-over-year, U.S. was only $11 million growth.

And last year out of $56 million, it was $7 million. So the majority of the growth is outside of the U.S. And what I'm trying to understand is to link the story of AI uplift to the fact that the growth is coming only from outside of the U.S. Why don't we see more U.S.?

That's number one. And number two, why do we see a lag between system growth that is consistently growing every quarter. You went from $160 million to $226 million in 5 quarters, but software is back to Q1 level of '25, so $160 million, give or take, $164 million. So why do we see a lag between software?

And at the time of refresh, don't companies upgrade their software package as well and then we should see growth in software?

François Locoh-Donou: Okay. Tal, thank you. I will start and then Cooper may complement me on a couple of aspects you've raised. So let me start with the U.S. First of all, the trends of our business in the U.S. are very healthy. I would not read too much into a given quarter's performance of this or that region. Some of it is the timing of what was able to ship to which customers in the quarter. But generally, the trends that we're seeing around the expanding threat landscape that's creating more opportunity. We had a very strong security quarter, as I shared. That trend around expanding threat landscape, driving more security opportunity for F5 is a global trend. The trend of AI that -- and I shared some numbers earlier, I shared we did -- we are approaching 100 customers in AI. We did about $50 million in sales in the first half of the year in AI. That is a global trend that obviously includes the U.S. and the U.S. is actually pretty strong in that trend. The hybrid multicloud trend is also global and including, of course, in the U.S., where we are seeing more customers want resiliency. But that particular trend is, in fact, very pronounced in Europe, Middle East and Africa because of digital sovereignty requirements there, and we are seeing extra growth coming from there. So I would say when you're trying to dissect, you said you're trying to dissect what's the refresh versus what are secular trends. The three trends that I've mentioned are secular and they are global. In addition, of course, we have a strong refresh cycle. Cooper mentioned the attributes of the refresh. It is stronger than usual because we have an even higher retention rate than we've had in the past, and we have more customers expanding at the time of refresh. That is also a global trend. But what I would take away is that the three big trends that I've talked about are cyclical -- sorry, they are secular and they are global, and they are at play in the U.S. as well.

Yes. And then just to touch on the software and systems dynamic, just a couple of dynamics that I would point you to is, one, the software business is largely a subscription business. We said this quarter, 90% of our software business was subscription. And of that subscription business, the majority does come through in a renewal motion.

And so we are seeing strong attach of software at the time of refresh, but it's still a relatively small component of the overall software number. The majority of the software number is this base that we continue to expand over time. And we referenced this year that because the renewal cycle is coming off of our flat software year from FY '23 that there would be a bit of a slower growth rate this year, followed by a much stronger growth rate next year. So don't mistake that the slower growth rate this year is having to do with expansion in attach rates at that time of refresh because those trends are actually pretty healthy.

And next, we'll move to Michael Ng at Goldman Sachs.

I just have two. First, this is just on systems revenue growth in fiscal '27. Obviously, you guys have had two strong back-to-back years in systems revenue. Could you just talk about your early expectations around whether fiscal '27 systems can grow just given the strong refresh that we've had in the last couple of years? And then a related question, it's been about, I think, 4, 5 years since the launch of rSeries and BIG-IP VELOS.

Are you expecting a new kind of ADC form factor system to drive another refresh cycle, particularly given all the incremental demand from AI? Just wondering how you guys think about new products on the ADC side.

Yes. Okay. So I'll start with the growth question. It's a little bit early to be guiding for next year.

But yes, we do expect there to be a growth opportunity for the systems business, just where we're at with the refresh cycle right now and the strong trends we've been seeing both in expansion at the time of refresh, but also new use cases. And we haven't spent as much time on that, but we really have been seeing new growth, pretty healthy growth outside of the refresh. So some of it's the AI use cases that we've talked about. We've been seeing higher takeout rates from competitors.

Some of the data sovereignty -- digital sovereignty dynamics are coming through as new business in addition to expansion at the time of refresh. So all of that's kind of giving us pretty good visibility 2 quarters out into next year, and we feel pretty good about the growth opportunity in that regard. As far as the next range of appliances and systems offerings. We wouldn't get into specifics at this point.

But yes, of course, we are well down the path of planning. We think there are some pretty interesting growth opportunities further downstream as we start thinking about things like PQC. And so continuous investment and innovation on our systems as well as our software has been something that's been critically important. And I think we're really kind of the only player in the space that has stayed steadfast in investing in systems.

And I think that's really paying off right now. We've always felt like customers are going to need choice and that their environments are dynamic and how they architect can change over time. And so giving that flexibility for customers to deploy how they need to is going to be important, and that's really coming through right now with the business that we're seeing.

And that concludes our Q&A session. I will now turn the conference back over to Suzanne for closing remarks.

Thank you, Audra. We look forward to seeing many of you during the quarter and especially at our Analyst and Investor Day in May. Watch for more details in the press release about the event coming soon. And thank you all for joining us.

And this concludes today's conference call. Thank you for your participation. You may now disconnect.