F5, Inc. (FFIV) Q2 2021 Earnings Report, Transcript and Summary

Good afternoon, and welcome to the F5 Networks Second Quarter Fiscal 2021 Financial Results Conference Call. [Operator Instructions] Also, today's conference is being recorded. If anyone has any objections, please disconnect at this time. I will 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. François Locoh-Donou, F5's President and CEO; and Frank Pelzer, 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 on hand to answer questions during the Q&A session. A copy of today's press release is available on our website at f5.com, where an archived version of today's call will be available through July 25, 2021. Today's live discussion is supported by slides, which are viewable on the webcast and will be posted to our IR site at the conclusion of today's discussion. To access the replay of today's call by phone, dial (800) 585-8367 or (416) 621-4642 and use meeting ID 3461547. The telephonic replay will be available through midnight Pacific time, April 28. 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. Factors that may affect our results are summarized in the press release announcing our financial results and described in detail in our SEC filings. Please note that F5 has no duty to update any information presented in this call. With that, I'll turn the call over to François. François Locoh-Donou: Thank you, Suzanne. And good afternoon, everyone. Thank you for joining us today. I am pleased to be with you and to share with you our strong Q2 results. With a second consecutive quarter of 10% revenue growth and double-digit non-GAAP earnings growth in Q2, we are outperforming both our revenue and EPS goals. It is important to note that we achieved Q2's strong results with a different revenue mix than we expected early in the quarter. We delivered exceptional systems growth of 17%, while software growth was more muted than our expectation at 20%. During this call, we will discuss the market dynamics behind our product revenue mix in the quarter. The transformation we have worked persistently to achieve has put us at the center of applications, both traditional and modern, with a truly multi-cloud approach. As a result, we are benefiting from strong and sustainable macro growth drivers ultimately powered by application growth. Business and consumers' increasing reliance on applications has accelerated all prior expectations about the pace of digital transformation. Our customers across the globe are scaling their digital assets faster, resulting in growing demand for F5's application security and delivery solutions. With our strong overall results, our conviction in our opportunity, both short and long term, is stronger than ever, driven fundamentally by accelerating application growth. We feel very good about our future given robust demand drivers and our strong and differentiated market position. I will turn the call over to Frank to walk you through our Q2 results and our Q3 outlook. Frank?

Thank you, François. And good afternoon, everyone. As François just outlined, our team delivered another very strong quarter. Second quarter revenue of $645 million was up 10% year-over-year and at the top end of our guidance range. Please note, as I review our revenue mix, I will be referring to non-GAAP revenue measures for the year ago period. Q2 product revenue of $309 million is up 18% year-over-year, representing a significant acceleration from 10% in the same period last year and even more so from flat growth in the second quarter of fiscal 2019. Product revenue accounted for approximately 48% of total revenue, up from 45% in the year ago period. The progress we are making driving double-digit product revenue growth and the increasing mix of product revenue as a percentage of our total revenue are both strong indicators of our transformation momentum and the long-term health of our business model. As François noted, we are seeing stronger-than-anticipated demand across the board. Short term, more of that demand is coming from systems, leading to the quarter's product revenue mix. Systems revenue of $201 million is up 17% compared to last year, when systems were down 11%. Systems demand was higher than anticipated in the quarter, largely from broad-based increase in application usage and the corresponding increase in application traffic, continued growth of systems-based security use cases as well as the emergence of 5G-driven service provider demand. Against a particularly tough 96% growth comparison in the prior year period, Q2 software revenue of $108 million is up 20% year-over-year, representing 35% of product revenue. We continue to drive our transition to a subscription-based model, delivering record subscription volume in Q2, with subscriptions representing 79% of software revenue in the quarter. This is up from 73% in the year ago period. Finally, our global services revenue of $336 million is up 4% compared to last year, representing 52% of revenue. Revenue from recurring sources, which includes term subscriptions, as-a-service and utility-based revenue as well as the maintenance portion of our services revenue, totaled 64% of revenue in the quarter. On a regional basis in Q2, Americas delivered 6% revenue growth year-over-year, representing 54% of total revenue. Our EMEA and APAC teams drove strong growth in their regions, with EMEA delivering 16% growth, representing 27% of revenue; and APAC delivering 15% growth, accounting for 20% of revenue. The quarter's strength spanned customer verticals, with especially strong demand from enterprise and telco. Enterprise customers represented 68% of product bookings. Service providers and government customers each represented 16% of product bookings, including 6% from U.S. federal from within the government vertical. Let me now share our Q2 operating results. GAAP gross margin in Q2 was 80.1%. Non-GAAP gross margin was 83.4%, reflecting higher systems revenue as well as higher levels of managed service solutions and our usual Q2 seasonal decline in global services margins. GAAP operating expenses were $463 million. Non-GAAP operating expenses were $342 million, reflecting our usual Q2 operating expense seasonality and the addition of 2 months of Volterra-related operating expenses. Our GAAP operating margin in Q2 was 8.3%, and our non-GAAP operating margin was 30.3%. Our GAAP effective tax rate for the quarter was 17%. Our non-GAAP effective tax rate was 20.2%. GAAP net income for the quarter was $43 million or $0.70 per share. Non-GAAP net income was $155 million or $2.50 per share. I will now turn to the balance sheet. We generated $128.5 million in cash flow from operations in Q2. Cash and investments totaled approximately $662 million at quarter end, reflecting both the cash used for the Volterra acquisition and the initiation of a $500 million accelerated share repurchase program. As a result of the ASR, we retired approximately $400 million of shares in Q2, reflecting 2.1 million shares purchased at an average price of $194.91 per share. We expect the remaining $100 million of ASR-related shares to be retired early in Q3. DSO was 52 days. And capital expenditures for the quarter were $9 million. Deferred revenue increased 7% year-over-year to $1.4 billion. We ended the quarter with approximately 6,360 employees, up approximately 200 from Q1, in part as a result of the Volterra acquisition. Now let me share our guidance for our fiscal third quarter. Unless otherwise stated, please note that my guidance comments reference non-GAAP metrics. Near term, we expect customers will continue to invest to support application growth and the modernization of their application infrastructures. We also anticipate continued focus on and investment in application security. It will come as no surprise that, like others in the industry, we are seeing some tightening in our supply chain. Thus far, our team has navigated it well, particularly given stronger-than-anticipated systems demand. Obviously this is an industry-wide challenge, and like others, we have mitigation efforts in place and we'll be watching it closely. With that as context, we are targeting Q3 fiscal year 2021 revenue in the range of $620 million to $650 million. We have accounted for the reduced supply chain visibility with a wider revenue range for our Q3 outlook, lowering the bottom end of our range by $10 million. While we do not expect to routinely provide product revenue mix guidance, we expect software growth for fiscal year 2021 will be at or around 35%, implying software growth in the back half of '21 exceeding what we delivered in Q2. Near term, we expect continued systems strength, with a slower growth rate likely in the fourth quarter. We expect Q3 '21 gross margins of 84% to 84.5%, and we estimate operating expenses of $338 million to $352 million. We also expect to achieve our fiscal year 2021 non-GAAP operating margin target of 31% to 32%. We anticipate our effective tax rate for the year will be approximately 21%. Our Q3 earnings target is $2.36 to $2.54 per share. We expect Q3 share-based compensation expense of approximately $63 million to $65 million. With that, I will turn the call back over to François. François? François Locoh-Donou: Thank you, Frank. The big takeaway from our Q2 results is that, across the globe, our customers are experiencing a pronounced growth in application traffic, which in turn is driving increased opportunity for F5. Customers across geographies and verticals are experiencing application demand well ahead of initial expectations and time lines. This escalating demand is coming on the heels of a prolonged period of sweating assets in anticipation of cloud modernization efforts and more recently pandemic-induced investment reprioritization. With no signs that application usage will slow, customers are urgently working to ensure they are able to support application traffic growth. The result for F5 is strong and sustainable overall demand for application security and delivery as well as a temporary change in customer buying behavior, evident in our Q2 product revenue mix. I will stress that, while we believe the opportunity around application demand is a long-term one for F5, we expect the current trend favoring hardware-based delivery models is short term. There remain a durable preference for software- and SaaS-based application security and delivery that will once again be evident in our results over the next several quarters. So let us dig into the quarter's demand drivers. I will frame my discussion with the 3 growth drivers we discussed at our November 2020 Analyst Day: one, ongoing software and subscription momentum; two, systems-based demand; and three, growing demand for application security in both software and systems form factors. We continue to see rising demand for application security, and increasingly, F5 is seen by customers as an application security leader. In fact, Q2 was our highest security quarter yet, with strength spanning both systems and software form factors across both traditional and modern applications. Where application security is [ threaded through ] virtually all of our customer interactions, both software and systems, for the purposes of discussion today, I've highlighted security use cases throughout both the software and systems form factor discussions. I will start our discussion with our software and subscription momentum anchored on trends we are seeing with application-driven demand. Our growth in modern applications continues to accelerate, driven by NGINX container and cloud-native deployments. We are seeing several top use cases emerge for NGINX, including API gateway, Kubernetes synchronized controller and software-based load balancing. Customers' modernization efforts and the availability of NGINX Controller and enterprise-level app security with NGINX App Protect are also driving larger NGINX deal sizes as we anticipated. In one example, our sales team successfully layered NGINX Controller with NGINX App Protect to enable one of the largest digital product companies in APAC to modernize more than 40 digital properties, replacing multiple competitors with NGINX Plus with app protect and controller. As a result, the customer got a much better ROI in a long-term subscription framework. Customers are also increasingly aware and appreciative of NGINX' extreme versatility. For instance, during the quarter, we secured an NGINX win with a regulatory body in APAC. The customer was facing 2 distinct challenges. First, they needed to refresh their electronic payment systems. And second, they needed to deploy microservices in a VMware Tanzu Kubernetes environment. NGINX proved the ideal solution for both. In the first instance, the customer moved from NGINX Open Source to NGINX Plus for the additional functionality and the added benefit of world-class global services support. In the second, they opted for a flexible subscription agreement which gives them the ability to scale NGINX both as a software ADC and as a Kubernetes ingress controller as they grow their microservices. And they are using controller for visibility and manageability across both use cases. Of note, NGINX was the only true multi-cloud solution they found able to work in both a VMware environment and the cloud. With pronounced application growth and an ever-expanding threat landscape, we also see continued demand for application security in cloud environments and rising demand for fraud and bot defense. With our Shape anti-fraud and anti-bot solutions, we are learning our win rate is best when there is significant automated traffic that can often circumvent traditional WAF protection. This is Shape's sweet spot. As an example, during Q2, a large credit union faced a massive credential stuffing attack, which their existing WAF could not defend. Shape could and did. While momentum in NGINX and modern application security software use cases increased in Q2, we experienced a notable shift in customers' delivery preferences for application security and delivery for traditional workloads served by BIG-IP. We continue to see growing demand for BIG-IP software in multi-cloud environments and expect the resumption of large-scale modernization efforts will remain a powerful growth driver for BIG-IP over time. Short term, however, we saw a mitigating factor to BIG-IP software demand which paradoxically was driven by continued application growth. Facing escalating application traffic, several customers opted to refresh and augment their existing infrastructure instead of transitioning to software or a cloud environment. When asked, they explain that the systems form factor offered an expedient way to get the urgent capacity their applications and users needed in a well-operationalized deployment motion. While cloud modernization and expansion are most certainly part of their future plans, they chose to deploy systems now. We expect this pattern with BIG-IP software is temporary and will moderate in the fourth quarter as customers resume BIG-IP software purchases in support of longer-term strategic projects and modernization efforts that COVID working conditions have made challenging. Stepping back. We continue to drive very positive software trends in the business. Our software subscription momentum continues, with 79% of Q2 software revenue coming from subscription-based sales, up from 73% in the year ago quarter. In terms of volume, the number of multiyear subscription agreements were up 32% quarter-to-quarter and more than 200% compared to last year. In addition, while it is early still, we are beginning to hit some multiyear subscription renewals, and we are also seeing positive signs there. During Q2, we achieved a 100% renewal rates on the long-term subscription agreements that expired in the quarter. In addition, 75% of those renewals grew over prior levels. While we've yet to hit the inflection point for renewals, we are very encouraged by these early data points. In addition, we continue to see utilization improvements. Our Q2 customer success metrics show long-term subscription customers are achieving 100% utilization sooner than ever before. Turning to systems. Part of F5's value proposition is our ability to offer our customers enterprise-grade application security and delivery solutions in multiple form factors, systems; software; and increasingly, going forward, SaaS. As I just discussed, sudden and rapidly accelerating growth in application usage led to accelerated customer demand for systems in Q2. Last quarter, we articulated several systems demand drivers. Another quarter in, we believe we have additional clarity around the drivers of our systems growth and we can break them into 2 broad categories: one, accelerating application traffic; and two, security, including emerging 5G-driven demand. To a large extent, I have already spoken to the accelerating application traffic. With surging application consumption, customers have urgent capacity demands to fulfill. As a result, several customers are opting to refresh and augment their infrastructure at a faster rate than we anticipated. Of note, they are doing so without a new systems platform from us. In some cases, the deployments coincide with the "end of software development" date we discussed last quarter, but broadly speaking, the underlying driver is the need for application security and delivery capabilities to deal with escalating application usage. We are seeing demand span geographies and customer verticals, from financial services to technology, to global SaaS providers. As a case in point, one of our biggest systems deployments this quarter was with one of the planet's largest cloud-based software companies looking to build a scale model for organic growth and the scale within their existing systems architecture. Similarly, last quarter, we had a sizable system deployment with one of the largest tech giants in support of their global web-based collaboration platform, and they scaled with their existing systems architecture. These are cutting-edge players, and in all likelihood, they will be among the first to move to software and cloud-based infrastructures in the future. However, today, because of the confluence of demand, the complexities of deploying and scaling in the cloud and the challenges of taking on transformative projects in a COVID-influenced environment, they are solving their application and delivery challenges with systems from F5. Wrapping up our growth drivers discussion, let's talk about systems-based application security, including a new driver, emerging 5G demand. As we have said for several quarters now, our systems business is benefiting from increasing demand for security use cases. Cross-sell of application security is driving systems growth as more customers look to consolidate vendors and combine ADC and application security functionality. Among enterprise and government customers, we also are seeing strong demand for web application firewall and continued strength in identity and SSL orchestration. The new driver that developed this quarter is emerging service provider 5G demand. We expected that, as 5G traffic began to flow from the radio edge into service providers' 4G core networks, we will benefit from increased demand given our strong position in 4G Gi LAN infrastructures. We initially and conservatively estimated 5G-related demand would begin to materialize in late 2021 or early 2022. In Q2, we secured several Gi LAN expansion projects, including a large Gi LAN firewall expansion with a North American carrier. These wins and other active opportunities suggest that we are in fact beginning to see the emergence of 4G core expansion driven by 5G demand. So what does all of this mean for our business and growth going forward? Fundamentally, we expect customers to transition to more software- and SaaS-based solutions over time. We are confident that the investments we have made, both organic and inorganic, and forward momentum our teams are driving position F5 as a significant beneficiary of that transition. This is true in the short term. As Frank said, we expect our software growth in fiscal year 2021 to be at or about 35%. And it is also true in the long term as software- and SaaS-based revenue account for a more sizable portion of our total revenue. We expect demand for application security will continue to grow as application demand grows and customers scale and modernize their applications. We believe that we are exceptionally well placed with the right perspective and tool set to solve our customers' most pressing application security challenges. Our opportunity in application security is even more exciting with the ongoing integration of F5 and Volterra, which will bring enterprise-grade F5 application security to the edge in an easily deployable SaaS model. We expect the recent high growth rates we have seen from BIG-IP systems will begin to moderate in the second half of this year. And we expect service providers' 5G-related demand will likely begin to migrate from systems to software in 2022 as their 5G cores start to hit production. In the meantime, we see demand for systems-based application security persisting over multiple quarters. Before I close our prepared remarks, I will say a few brief words about Volterra and our integration process. We launched our integration and value creation efforts immediately following the acquisition close on January 22. We are thrilled to have Ankur Singla and the Volterra team as part of our security organization, led by Haiyan Song. While we have work ahead, we are very pleased with the initial positive customer response. Early indicators show our vision of F5's Edge 2.0 is resonating with customers. We have started a pilot program concentrating on specific use cases and focused on bringing F5 security to the edge, with the goal of leveraging Volterra's organic momentum and early customer interest. Through this pilot, we will develop new business as well as derive customer behavior insights. As the first step in our long-term integration of F5's solutions and Volterra's platform, we have begun the process of strategically and methodically combining our best in web application and API protection security offerings and Volterra's innovative platform. We are also formulating our go-to-market approach and exploring ways to maximize the benefits with our channel partners. Only a few months into this integration, we are even more excited about the potential of this combination as we move into FY '22. We will continue to share our progress with you in the coming quarters. I will wrap up today's prepared remarks by thanking the entire F5 team again as well as our customers and partners. In particular, our thoughts are with the F5 team in India, their families and loved ones as they endure the extreme health risk and loss of life occurring there as a result of the current COVID-19 outbreak. With that, operator, we will now open the call to Q&A.

[Operator Instructions] Your first question comes from James Fish from Piper Sandler.

François, you really talked a lot about the near-term benefit from systems and the expectation for it to get software bookings later on in the year. Are there deals already engaged for that cross-sell opportunity of, say, NGINX and other parts of software that you can point to that's already happening? And maybe, Frank, is there any way to quantify? Last quarter, we heard about roughly $10 million impact from the end of development. Was there another kind of $10 million this quarter of kind of pulling of demand or no? François Locoh-Donou: Jim, thanks for the question. The short answer is yes. There is -- we have a strong pipeline of software deals both for NGINX and for BIG-IP that will materialize in the second half of the year. What we did see this quarter -- so if you look at the software growth drivers, Jim, we talked about 3 growth drivers on software. And in that order, one is modern applications. Second one is security, and the third one is BIG-IP deployments in multi cloud. The first growth driver, around modern applications, is going very well, largely driven by NGINX exceeding our expectations and driven by the addition of the controller and security on NGINX and more and more customers deploying modern applications in these cloud-native and container-native environments. The second driver of security, also going to plan with more and more customers deploying security either in these modern application environments or on top of traditional applications both in systems but also in software form factors. The third driver of software growth is really the deployments of BIG-IP into multi-cloud environments and specifically some customers migrating from systems to software-first or cloud-first environments. And whilst we see continued growth on those BIG-IP deployments, what we did see this quarter is a bit of a moderation as -- on that specific driver as, customers that really need to increase their capacity to support application usage and application growth, a lot of them chose the most expedient way to deploy that capacity, which is in hardware where they're operationalized; and don't have the -- either the time or the mechanism right now, because of the COVID environment, to go and think about re-architecting things and moving to software. We do think that, that will resume in fairly short order, so we're very confident about software growth both for the second half of the year and for 2022.

And Jim, specific to your second question. I think, the end of that software -- the end of software development cycle that we saw in the beginning of April, that was one trigger, but it's really hard to split out exactly dollar-wise what was the drivers for everything. I think the others that we have to include in there are the capacity expansion that François spent a lot of time talking about in the prepared remarks, customers' depreciation schedules and then just their inability to continue to sweat the assets that they have probably for a couple of years now, some reprioritized budgets and a few other factors. If I had to quantify it, I think it's probably about the same as it was last quarter, but there are a lot of other factors that went into the strength of hardware.

Yes, totally understand and appreciate that. Just one more follow-up for me, if that's all right: Obviously a lot of major breaches occurred before you reported last time, and really that pipeline hadn't had a chance to necessarily build too much. I guess, what are you guys seeing? Obviously you're talking some really good security results on -- including the carrier side, not just the enterprise side, but I guess, anything to call out regarding that security pipeline for WAF given what happened with those breaches? And are there areas of security you guys might look to expand into to kind of bolster the portfolio overall? François Locoh-Donou: Jim, is your question specific to WAF?

Yes. François Locoh-Donou: Okay. So on and specifically on web application firewalls, Jim, we continue to see good traction there. It's one of the most in-demand use cases from our customers. We have seen that traction this quarter happen both in hardware form factors and either in stand-alone form factor or with customers sometimes bundling that with ADC. And we also see demand for web application firewalls in these modern application environments. And we're really pleased to see the traction with NGINX App Protect and which as you know is that effect we're seeing of having made the decision to port our security capabilities onto NGINX shortly after the acquisition. That's gaining a lot of momentum now. So we're getting embedded for security in these modern apps environment. And we also see that with BIG-IP. In terms of the portfolio, Jim, we really are focused on the web application firewalls, application security in general, API protection, DoS protection and bot management and mitigation. Those are really the areas of application security where we are placing focus and significant investments for the near future.

Your next question comes from Meta Marshall from Morgan Stanley.

Great. Maybe a couple for me. Just on the first, is there an organic software growth number that we have for year-over-year, I guess, excluding kind of the 2 months of Volterra that would have been included? And then -- and just maybe on kind of extrapolating on the last question, but just as you're deepening question -- sorry, deepening conversations and sales to encompass kind of more than one product, just sort of what are you kind of seeing, as far as sales cycles, for cross-sell opportunity versus maybe a point product?

Yes. So Meta, let me take the first one, and then I'll let François take the second. I think we talked about we were not going to split out Volterra because it is rather immaterial. And I mean super immaterial especially this quarter with only 2 months. And so I -- it's not enough to move a growth point, and so I will just say it's rather immaterial to the results of the operation. It becomes a critical asset for us, particularly in the out-years. And we're really, really excited by the excitement that we've seen within the customer base and our sales force already but, in terms of actual revenue results, really nothing to speak of from Volterra, but I'll let François address the other. François Locoh-Donou: Yes. Meta, in terms of cross-sell opportunities, a couple of things. Where we are really pleased is we are seeing more and more customers of F5 adopt multiple elements of our portfolio. And so more and more of the multiyear term subscription agreements that we are selling have multiple products from F5. And it's allowing us to elevate the conversation with our customers, really continue to elevate the position of F5 in a lot of these large accounts. And we're really pleased with the traction of these agreements. In fact, as a data point: The software multiyear subscription agreements were up 200% year-on-year in terms of the volume of transactions that we completed this quarter relative to Q2 of 2020. And this points to the number of our customers that are moving to these term subscription agreements where they can easily consume multiple of our products. In terms specifically of our progress on the new point solutions, I think Volterra is too -- it's really too early for us to speak to that because we're basically still through the integration, porting our capabilities onto the Volterra platform, including our security, so it's more over the next few quarters that we'll be able to speak to deals that are coming through on Volterra. And then on Shape specifically, where -- we are pleased with the value proposition. I mean the customers that use Shape are extremely pleased with the efficacy of the solution. It is best in class. A lot of times, we see customers who are -- actually already have solutions perhaps from another vendor, like a CDN vendor, but they're still under attack. And we come in with Shape and have a higher efficacy solution and can solve those attacks, as a point solution. Where we are seeing things take longer than we would have liked is in the COVID environment. When you have to go through a proof of concept and -- mechanically it's actually difficult to go through with. And we're finding that, in this environment, being incumbent and operationalized is a big advantage, but when you're trying to introduce a new solution that requires a proof of concept, the sales cycle is a bit elongated and it takes a little bit of time. And I will say that overall is the picture on cross-selling the various elements of the portfolio.

Your next question comes from Sami Badri from Crédit Suisse.

Congrats on solid results. The -- François, in your prepared remarks you talked about escalating demand. You talked about some of these other tailwinds, including 5G security and just service providers, going forward, but I guess I think the big question that we have is, as you see these incremental big drivers start to come in and how that actually impacts your full year guide, is there a reason why you guys aren't at a position yet to increase the Horizon 2 guide points at this point? Or is there a reason why you're being a little bit more conservative about the growth outlook and why not to increase the outlook? François Locoh-Donou: Yes, Sami, I think there are a couple of things. We are -- Horizon 2 is, as you know, an 8-quarter period. We are 2 quarters into that, so [ generally ] we feel it would be really early to make any change to our Horizon 2 guidance, but I think the what we can say is a couple of things. I think we are definitely seeing those trends of application usage go up significantly as a result of most of us, frankly, consuming more things online. And so more and more of our customers see more usage on their application, whether they are traditional enterprises or even cloud providers. As you know, we are embedded in SaaS provider stack or even cloud providers. I think you've heard me say before that somebody's cloud is somebody else's data center, and we are in those stacks. And a number of our cloud provider customers are seeing escalating demand for their solutions and they have to grow their capacity as a result, and we're benefiting from that directly. So we're definitely seeing that. As a result of that, we can definitely say that our hardware for 2021 will be in positive growth territory, for sure, given what we've seen. And even for Horizon 2, the performance of our systems will be materially better than what we thought it would be at the beginning of the Horizon. And I think those are the things we can say. There's a little bit of uncertainty we're dealing with, with the supply chain challenges that all vendors, I think, are seeing at the moment, so we're a little cautious about that. And we shared in the prepared remarks, I think Frank shared, that we took that into account a little bit in our guidance range, but in terms of the underlying demand, it is very strong.

Got it. And then I just wanted to just visit U.S. federal demand. The presidential administration has been very active to raise and secure funds and modernize a lot of infrastructure. Are you -- and you guys came in, I think, relatively in line on your number for U.S. federal in fiscal 2Q, but when we think about the rest of the year and how things are going with U.S. federal, can you just give us a bit of an update in terms of how that is looking, just we can get an idea? François Locoh-Donou: I think, Sami, we don't see a fundamental change in the demand trends. The Fed business has been pretty strong, and we expect this to follow kind of normal patterns and given what we see in the pipeline today.

Your next question comes from Tim Long from Barclays.

Two questions, if I could. First, could you talk a little bit about the cloud vertical, which you broke out last fiscal year, I think, as $100 million? Can you just give us an update how growth has trended in that vertical piece of the business? And then second, on the software ramp for the second half of the year, it sounds like visibility is pretty good. Could you just give us a little more color? It's -- after a few quarters of kind of flat, it looks like a pretty big dollar growth. Are we to expect some ELAs or anything like that in the numbers to help the sequential revenue increases given that most of it's going to be on subscription? I would assume, if not, you'd need kind of a big scale of subscription deals, so any color you can provide there would be helpful. François Locoh-Donou: Thank you, Tim. So let me just start with the second part of your question, on software growth in the second half, and then I'll come back to the first part. On the -- so in terms of the second half of the year in software, Tim, there are a number of catalysts and drivers that have continued to gain strength. I talked about NGINX a little earlier, but these modern apps environment, we see them continue to grow. We have in fact a number of subscription agreements in the pipeline, but I would say the predominant factor is that the volume of software multiyear subscription agreements that we are signing is just growing extremely rapidly. They grew -- sequentially from Q1 to Q2, they grew roughly 30%, but if you look at it on a year-on-year basis, the volume of subscription agreements was up 200%. And it just continues in that direction. So we're getting visibility to this. We're -- at some point, we're going to hit also the inflection point on renewals of these subscription agreements. We're not quite there. We think that's more of a 2022 effect, but even what we're seeing today from the renewal rates of the ones that are coming to term and the true forward on the ones that have more than a year in maturity, those metrics are just excellent at the moment. So these catalysts are important. Security continues to be a growth driver and a catalyst for growth in software and then the cloud. And I want to come back to the first part of your question. The cloud continues to be a growth driver for our software. However, this quarter, it was a little less in terms of the growth rate. And the reason for that is because we think that there was a moderation in the number of customers migrating applications to the public cloud and specifically migrating more of these traditional applications to the cloud. And we think this is back to the effect we're seeing on our hardware business, where customers really sweated their assets for a long period of time, saying, "Hey, we don't want to deploy any more hardware because we're going to migrate to the cloud". And what we're finding is a lot of them, after having sweated their assets for a long time, are caught with increase in capacity needs that they have to serve very quickly. And the -- operationally, the best and the easiest way to do that is to actually go with hardware. And so that, we think -- and when you factor on top of that the fact that we're in a COVID environment, that moving from hardware to software or moving from hardware to cloud requires either re-architecture or working in collaboration across multiple teams, which is much more difficult to effect in this environment, then we're seeing that less customers are able to do that at the moment. We think, over time, the trend will continue towards these cloud migrations, but this quarter, it was a little more muted.

Your next question comes from Samik Chatterjee from JPMorgan.

Great. François, I just wanted to first start off with the implications of the reprioritization of spending that we're talking about towards a bit more systems, but you also talked about the 5G spend. Now as you put both of those together and think about longer term, is the outlook still for systems to decline in that mid-single-digit range? Or does -- some of this reprioritization, the increase in 5G, does that change overall how you think longer term about systems? And I have follow-up. François Locoh-Donou: Samik, depends on what we mean by longer term. So I'm going to speak within Horizon 2, Samik. As I said before, I think, 2021, we'll see for the full year our systems business will grow. And for the combination of 2021 and '22, I think performance for systems will be better than what we had anticipated in the beginning of Horizon, which was high to mid-single-digit decline. I think we'll be materially better than that. In terms of what this means longer term, frankly, it's too early to predict beyond Horizon 2. I would say, in terms of the hardware ADC space, we think, hardware ADCs per se, those are in secular decline and should decline and continue to decline over time beyond Horizon 2. And we're not seeing that in hardware security today. We'll continue to see growth in hardware security, and that represents more of a mix for us. So well, too early to make a determination in terms of what's happening beyond 2022. And specifically when it comes to service providers, what we are seeing, Samik, is right now a lot of carriers have put in place their 5G radio -- as you know, the first smartphones with 5G have come out in the fourth quarter of 2020. A lot of the spectrum auctions have happened. And so we're now starting to see utilization of 5G from the devices into the radio, and that is coming into the carrier core networks. Now most of these carrier core networks are still 4G and we have a very strong presence in a lot of these infrastructure. And so the capacity requirements are fulfilled by 4G capacity upgrades, and that's providing tailwind to our hardware business. And I expect that to continue for the next couple of quarters, probably a little longer. Now as you go into 2022, a lot of these carriers will start to put in production their 5G core. And those are virtualized and so that opportunity will turn into a software opportunity for F5. And we have already won significant design wins for these 5G cores. And we're in the process of doing first office applications and pilots and things like that before going into production in the next 6 to 12 months. So it's a hardware expansion opportunity today, and over the next 12 months, it will move to be a software opportunity in 5G cores for F5.

Okay, got it. François, just as a follow-up, looking at the growth trends across the different geographies, I'm just trying to get a better sense of the variability we are seeing. Like Americas was a strong-growth -- strongest-growth geography for you last quarter. It's moderated significantly in terms of the growth rate while still growing, whereas EMEA and APAC are driving the growth this quarter. Is it like something on the ground that's impacting it in terms of like sales force, et cetera? Like what's really driving that amount of variability in the performance by geography? François Locoh-Donou: There isn't really anything of note. I would say, the service provider vertical, we have some important expansions in EMEA and APAC that help drive the growth there, but I wouldn't put too much into the quarter-to-quarter variability into the geographies. Overall, Samik, we're -- the strength in demand that we are seeing both in hardware and in the software subscriptions that I've talked about, that is across the board and across geographies.

Your next question comes from Alex Henderson from Needham.

So I was hoping you could talk a little bit about the transition to hardware in the context of the launch of the integration of the Beacon technology; and a platform last year, in the spring, which does tie in the Kubernetes NGINX-oriented products back to the BIG-IP; and to what extent you may be seeing some refresh there in those systems because the admin, IT administrator, the CIO is able to bring some control back into the DevOps process and use that integration between the BIG-IP to the NGINX product to put guardrails in front of the DevOps people. Is that part of what's going on here as well? Or is it just simply that, last year, you couldn't do installs because things were so locked down that the installed base is aged and therefore it needed the upgrade? Is it -- could you give us a little bit more clarity around that? And then the other piece of that is with Kubernetes adoption, according to a recent [ Flexera ] number, is like 2x what it was a year ago; and 96% of companies saying they're going to go with multi cloud. Can you talk a little bit about what kind of integration you're doing with Hashi and partnering -- how your partnership with HashiCorp is [ setting up ]?

Alex, it's Kara. I'll start on those. On the transition to hardware, I mean, we talked through a number of factors that are driving the strength and the robust demand that we're seeing in hardware today. One of those factors, I do believe, is around the strength of the portfolio is coming together. And customers are seeing F5's relevance in modern applications and strong relevance in security use cases. And certainly that's playing into some of what we're seeing, but I'd have to -- I really do think, if you go back to the script and what François has already addressed in the questions, the robust demand that we're seeing is really more of the latter that -- latter factor that you described. So that's at least the answer to your first question. On the second question, in terms of what we're doing with Hashi. We have a partnership with Hashi. We have several integrations in the BIG-IP space around various elements, including we make strong use of Terraform integrations to enable automation for our customers and to simplify their deployments into public clouds. And in addition, we're also looking at how -- we also have integrations in place to tie more into some of their orchestration tools like Consul to automate deployment and provisioning of BIG-IP offerings.

Your next question comes from Jeff Kvaal from Wolfe Research.

Frank, can we start with you with a question on the guidance, please? Could you help us parse kind of what you're thinking at the high end and the low end and what the impact of the systems mix would be in there? I mean typically, when companies widen a range, they often widen it on both ends. You widened it on the lower end, and I'm just wondering what kind of expectations and variables are built into that range.

Sure, absolutely, Jeff. So when we're looking at all the factors coming into setting guidance expectations in a normal cycle, if there weren't any supply chain concerns, we would have said $630 million to $650 million, with $640 million as the midpoint. And taking into account what I talked about in regards to just some less visibility than we would normally like on some of the supply chain components, we wanted to make sure that we gave the additional room of the $10 million on the low end. And so that's a result of the range. In a perfect world, it would still be that $20 million range, but we just wanted to be cautious with our outlook.

Are you more concerned that you might not get what would have been at the midpoint, the $640 million number? Or are you more concerned that the mix might shift more heavily to systems than you thought?

No. I think, when we take a look out at the mix that we expect and the way that we talked about being at or about that 35% for software for the year, we're very comfortable with that outlook on the software side. The services, we didn't give exact numbers for that. And what's remaining is the systems side. And so when we take a look at that $10 million of risk, we see that $10 million of risk really in -- on the supply chain, certainly not on the demand side. As we've talked about, on multiple factors, the demand is quite strong.

And then I guess, lastly, I think we would all look forward to a little bit more smoother trajectory in the software side of business. I think you had suggested that this March quarter was the last really, really tough comp period. How close are we to having a more volatile -- or a less-volatile software number from quarter to quarter?

Yes. I -- again I can't say with 100% clarity, Jeff, but I think 2 factors that I say in the back half of Horizon 2. And so when we're taking a look at the fullness of FY '21 and FY '22, we knew this was a 96% comp quarter for us going in and that there was going to be some variability. Obviously, in the last quarter, we had 70% with Shape, 35% without. And so the -- that range was right there where we thought it was going to be, and we think it's going to be a bit smoother from here. There will continue to be variability. Where we start to see a lot more clarity is when we start to lap some of the multiyear subscription agreements, and the bulk of that is going to happen in the back half of FY '22. And so we do have 4 more quarters of what I would say variability. I don't think the choppiness will be what you saw between Q1 and Q2 of this year, though we will continue to have some variability in that but starting in that lapping period. And then frankly, the denominator just gets so much bigger, so you don't have things that swing. We will start to see that as well. With the SaaS piece of the subscription continuing to grow as well, this is more -- when we take a look even beyond Horizon 2, that will become a lot more ratable in that factor and smooth things out. And so we do have some more choppiness ahead of us, but I think things really do start to smooth out when we take a look at the back half of '22 and beyond.

Due to time constraints, we will take our last question from Paul Silverstein from Cowen.

Awesome. That means I could ask 5 questions, but I'll only ask 3 -- no. First off -- and François, I'll apologize because I know you've addressed this in other responses. And if you've fully addressed it, I do apologize, but I've got to ask you, first off, to the extent that it's probably not a bold statement to say you're not going to get credit from the investment community for the hardware strength. So I just want to make sure that I heard you correctly. You're asserting that the software weakness, relative to your original expectations, is entirely a function of a number of customers unexpectedly opting for traditional hardware systems due to urgency and addressing strong application growth and security needs. Is that what you're saying, or is there something else in particular? Is there any underlying weakness in software [ demand ]? François Locoh-Donou: No. The -- so if you look back, there are 2 factors for -- if you're specifically, Paul, referring to the software growth rate at 20% versus the 35%, 40% for Horizon 2, there are 2 factors. One, of course, is that, as Frank just alluded to, it was a high comp relative to last year, but yes, the other primary factor is customers that paused or moderated on migrations to software due to operational constraints. And that is reflected in the higher hardware number and a lower software growth number, and that is very specific to BIG-IP. It's not specific -- so in the other aspects of our software portfolio around NGINX and security, we had very strong growth. In fact, NGINX grew faster than overall product revenue again, but specifically with BIG-IP, the softness, if you will, in software there is directly correlated to customers choosing, unexpectedly to us -- and it was a little bit of a surprise to us, choosing to go to a hardware form factor instead of a software form factor in a number of cases. I would also add, Paul, that -- when you step back and you look at that, the overall demand for F5 technology is very strong. And as these customers continue to deploy BIG-IP, it increases our footprint. And when they do migrate to software, they will migrate with us. We've already proven that for the last 3 years. They will migrate on a BIG-IP software form factor. So for those customers that did that, it is a temporary delay in them moving to software, but there's absolutely there a substitutive effect of customers pausing on software, going more on hardware because they were a little caught by surprise by the surge in application demand. And given the COVID environment and the inability to re-architect and work across teams to go to software, they chose to continue and deploy hardware form factors.

Understood. So that's a form factor choice [ in this ] you're indicating. I suspect those customers choosing to remain on the hardware will eventually migrate to software, but it's not indicative of weakness in software per se. It's a form factor choice. Let me move on: I appreciate you mentioning that NGINX grew faster than overall growth. Any metrics you can cite with respect to Shape? Or anything above what you just said about NGINX in terms of quantifying what you're seeing from those 2 acquisitions? I recognize that Volterra is still at a very nascent stage, at least it sounds like it is, but anything you could add in terms of revenues, order book or other metrics that you can offer providing more color as to the progress on Shape and NGINX? François Locoh-Donou: No, I mean we shared, Paul, as you probably remember, at AIM some important metrics about the growth in ARR for Shape since the acquisition and the growth in the number of customers. We have continued to see growth in the number of customers. I would say the growth -- specifically the growth in ARR with Shape this quarter wasn't as strong as we wanted. And that's largely because we -- I was saying earlier that, for customers for which we need to do a proof of concept, in the current environment, for the same reasons that in some ways are keeping customers on hardware on the BIG-IP, effecting a proof of concept when people are not in the office and have to work across teams and collaborate and connect different systems together for proof of concepts to happen, it's operationally a little more complicated. It takes more time and it takes more motivation for customers to go and do that. And so when customers are in significant pain and they are under attack, it moves extremely quickly. When they're not, it doesn't move as fast as we want to. And that's part of what we saw this quarter with Shape, but overall, when you look at the efficacy of the solution and the satisfaction of customers and the way in which it's blocking an increasingly high number of automated attacks, we're very, very happy. And we think this thing about the difficulty of operationalizing proof of concepts and getting to a sales cycle is really a temporary thing with the COVID that will go away down the road.

Finally, François, just a clarification to -- in response to a previous question that you had. With respect to macroeconomic recovery and what you're seeing on a regional basis, I recognize it's still early, but given that the U.S. and U.K. in particular appear to be well ahead of Europe and perhaps other countries in terms of vaccination rates, any data points in terms of order book in the U.S. and U.K. or other metrics that would suggest what you should expect from other countries and regions as we [ eventually return to the 21st century ]? François Locoh-Donou: I would say generally, Paul, no. I don't think -- when I look at the patterns of spend for us: This quarter, it was the -- I want to say same last quarter. So you'll remember, Paul, I want to say, 3 years ago, we felt -- in Europe in particular we had a situation where the uncertainty of Brexit was weighing down on our numbers. And we could feel the spending patterns there being really different. This quarter and the last quarter, I would say, we don't feel that kind of difference. What we are feeling across the board is this -- everybody has gone digital. They're using their digital channels more and more, and it's causing the applications traffic to continue to grow steadily. And that's -- we're seeing the same in Latin America, Europe, North America, Asia. We don't see this difference related to COVID and where people are at in vaccination rates or coming back to the office. It's not really a first-order effect in our business at the moment.

I will now turn the call back over to the presenters.

Thank you all for joining today. We appreciate it. As we said, the call is recorded and the replay will be available on our website. Have a great day.

This concludes today's conference call. Thank you for participating. You may now disconnect.