F5, Inc. (FFIV) Earnings Call Transcript & Summary

Welcome back, everyone, for the last session of the day. We have the pleasure of closing out with F5 Networks, and we're joined specifically by really the big one-two combo CEO, Francois Locoh-Donou; and CFO, Frank Pelzer. Francois, I apologize for always mispronouncing your name, but we were joking around there that it's pretty much happy hour on the East Coast. And the next time we're going to have to do happy hour here. But welcome, guys. Thanks for doing this with us.

Thanks, Jim.

Thank you for having us, Jim.

Maybe just to start it off with the most obvious part for F5 in my view, is the transition towards a more software-defined and subscription model, that's really been a multiyear journey. You're now up to a point of seeing that subscription element, renewals really start to flow in. Are the software-only solutions leading to new customer wins or more expansionary and stickiness with existing customers at this point?

Jim, I'll start with that and bear with me for a second because before I respond, I need to get our safe harbor on record. Please note that our discussion today may contain forward-looking statements, which involve uncertainties and risks. Our actual results may differ materially from those expressed or implied by these statements, please see our SEC filings for more information on those risk factors. With that, Jim, let me just answer on your question on software and whether it's cross-selling, expanding in new customers or -- sorry, in existing customers or new. It really is both. And let me start with new customers. We have -- what we're seeing is more and more -- one of the things that has happened with the pandemic is the acceleration of digital transformation, the acceleration of people using digital channels to communicate, to work, to bank, to shop, to live, et cetera. And so we're seeing a lot of new customers that are putting new applications into production and that come to F5 to either deliver these applications and secure them. And I would say the new logo acquisition with new customers is across the board. It's with BIG-IP, it's with NGINX for those customers that have started with open source and want to move into a commercial relationship, and it would shape security to mitigate fraud. So new customer acquisition is an important vector of growth for us. The cross-selling is pretty substantial, because one of the transitions that we are making as a company is moving from a single domain company, which was the BIG-IP platform historically, to becoming a multi-domain platform with NGINX and Shape and then Volterra to come. And so what we're saying is for large enterprises, all of these new value propositions are relevant. And so more and more of the deals we're doing in software, especially these multiyear subscription agreements, have multiple products in them. And probably the best example of that is BIG-IP and NGINX. So last quarter, almost half of the multiyear subscription agreements that we did included NGINX and BIG-IP. And that points to a reality about F5 and about where customers are going that I think is often misunderstood, which is that for the next 10 to 20 years, our customers are going to live in a hybrid world. They're going to live in 2 houses. They have their kind of traditional house that they know and love and has all kinds of legacy and furniture and it's working, and it's driving existing revenue. And then there's a new house, which is all these modern applications typically built in microservices, in a lot of cases, built in public clouds. And the misperception about that is often that people think that customers are going to leave their old house and get a go and move into a new house. And the reality of large enterprises is this is not the case. They're going to live in this hybrid world for a very long time. And our role is to help them find the right balance between these 2 worlds and make the migrations when they need to. And we now have the solutions to enable them to do that. And that's why you're seeing so many customers increasingly consume both our products that support traditional applications and those that support modern applications. And that's part of the cross-selling and the expansion that we're seeing in existing accounts.

Yes, there's a point I want to go into there maybe a little bit later for now. But in terms of the go-to-market on software only, are you guys actually incentivizing the sales teams to sell software only? Or are the commissions roughly the same between traditional F5 systems and the software-only business?

So the answer is yes and no. So the no part is that, for the most part, if you're selling perpetual software licenses, your commissions are basically the same as if you were selling perpetual hardware licenses. So that's -- there's not a specific incentive there. We have, however, put in place incentives for our sellers for software subscriptions because, of course, if you are going to sell a multiyear subscription say, a 3-year subscription for $3 million for the full term, but the annual value of that subscription is $1 million, if we didn't compensate the sales teams such that they be compensated for the full $3 million that they've sold, they would always favor selling a perpetual license that has a bigger upfront component to it. So we have done things to normalize that. There's also -- so incentives and accelerators around certain propositions, new software propositions for F5. But we place a lot of value and importance in customers being able to make their choice around which form factor makes the most sense for them. And so we haven't tilted the incentives, such that when a customer has a pure hardware versus software choice are -- especially for perpetual licenses that our sellers are incented to do one versus the other.

Got it. And especially heading into this big 5G core cycle upgrade cycle and really the government budget focus time frame here, how should we think about the adoption of software versus systems across some of these, what I call, Tier 1 verticals like government, like service provider, like financial services, where really performance matters and on-premises, where mission-critical apps still exist today?

You know Jim, that's true about soliciting those verticals. That's true about government. It's true for financial services, for service providers. And I would also add technology, which may be counter-intuitive, but I would add technology companies to the list of segments where that's true. So for -- in all these segments, there is actually -- what we're seeing is that traditional applications, which can be supported with hardware and software, but traditional applications are growing faster than we had anticipated. And so if you think back, Jim, at the bare thesis of F5, the bare thesis was traditional applications, they're not going to grow. And by the way, they will all move to public cloud and being refactored. And so what we're seeing is, number one, the migrations to public cloud are happening but not at the pace that anybody thought. The -- most of those applications are not being refactored, so that when they move to public cloud, they are moving with F5, the vast, vast majority of them. But most importantly, those traditional applications in those 4 verticals are actually growing faster than anybody thought because of the changes, which we think are secular changes brought in by the pandemic, where we're all using digital channels for everything. And so the result of that is we serve those traditional applications and their growth. They need more traffic and they need more security. And we serve that with our big IT platform, both in hardware and in software. So the reason you're seeing this our hardware growing is because of this secular growth in application traffic and application security for traditional applications in the big verticals that we've mentioned. I'll say one last word about the technology vertical, which I added because you would think that that's the vertical where you've got the most leading-edge practices around building and securing applications. And we're seeing that. By the way, we're seeing enormous growth in the -- with technology companies, with things like NGINX and Shape Security in securing modern applications. But we're also seeing in those companies, a lot of growth in hardware and in software, in our BIG-IP platforms because of all of the applications that are already built that are attracting a lot more traffic than they anticipated.

Makes sense. And NGINX Sprint was fairly recently with really that modern architecture that you really envision. Now was it 3 years ago almost roughly or so?

2.5 years.

Coming into the full -- yes. I mean what has made NGINX so successful inside of F5? How big of a business do you actually envision within F5 that NGINX could be one day? Understanding there's actually like kind of come together in the model here. And really, what was the most important announcement from Sprint 2.0 in your view?

So if I go back to -- if you go back, Jim, to what has made BIG-IP successful as a platform, and it's that BIG-IP started with load balancing as a functionality. But because of its position in front of the application servers of all of our customers, it became an insertion point where you could consolidate a lot of functionality, like DNS or authentication or SSL encryption or web application firewalls, et cetera. And it became -- for our customers, it was a lot simpler to concentrate and consolidate all that functionality into one platform rather than have 15 different vendors inserting and looking in the traffic when you could have it done by F5's proxy. That created the success of the BIG-IP franchise. The reason NGINX is being successful inside of F5 is because we know that playbook, and we're doing the same playbook with NGINX in modern application environment. And so on the NGINX platform, of course, there was a load balancing capability. But we've added to that security. We're adding API gateway. We're adding a service match. It's already a web server and a map server. And so the consolidation of all these functionalities in a very scalable, modern container native piece of software, makes NGINX the -- essentially the software of choice, the platform of choice to take all of these Kubernetes environments to scale, right? And so what we're finding is customers test these microservices apps, but when they want to go in production and they want to scale, they realize that they need this tool and that tool, and these things don't work together and they need this and that. And NGINX comes in and with the investments we made 2 years ago, we can consolidate a lot of that functionality and it just makes it a lot simpler to scale Kubernetes. And so we are winning all over the place with Kubernetes. And we're -- even in the early innings of that. So how big the business will be, I think it's all up in the year, but it is going to be a multibillion dollar market, this segment of modern applications. And we intend to play a substantial role in that market over time. And the early sort of trends and what we're seeing are very encouraging. And in terms of the conference, Sprint 2.0, I think a couple of the big announcement was the commitments we made to the open source community. When we acquired NGINX, there was perhaps a fear that the NGINX -- the open source community would start to disassociate with NGINX because F5 is a big commercial entity. But to the contrary, we've seen the adoption of NGINX open source continue to grow. We have over $0.5 billion websites that are supported by NGINX in the adoption of a web server, our app server, we've released now our service mesh. All of that continues to grow. And we wanted to set a strong signal that our commitment to put more capabilities in open source is strong, and we're putting control plane and management plane capabilities in the open source domain for the open source community. I think that was the bigger -- the biggest part of the announcement at Sprint 2.0.

Wow, that's a lot there. Speaking of modern, you recently bought this company called Volterra, with that, to me, helped move F5 in the F5-as-a-Service direction at an accelerated pace. I guess what's left to integrate with Volterra? And how should we think about what solutions will come first on top of Volterra in terms of F5 software?

It's a great question. Security will come first, Jim. And so we bought Volterra because it's a universal edge platform, which allows us to insert critical application services at the edge and allow our customers to consume these services in a SaaS format. And that's a total net new opportunity for F5. It's a growth opportunity that we think is going to be substantial. And so what we're doing, the first order of the day in terms of integration is taking -- we have one of the best, if not the best security -- software security stack for applications in the industry, including, of course, our web application firewall, our DDoS capabilities, our API security capabilities, our bot capabilities from Shape, and we're taking that entire security stack and integrating it natively into the Volterra platform. And so that will be the first offering to come out of the combination. It's an edge security as a service proposition. And we'll talk more about that later on in the year. We've said the integration would take about 12 to 18 months, and we feel we're on track to stick with those time frames.

Yes. It sounds like it changes the competitive dynamics a little bit. We might be able to get into that a little bit here. But speaking of security, app sec is really in vogue, whether it's about protecting my applications from ransomware or as I move to public cloud WAF and identity really becoming important in those public cloud environments. I guess how should we think about the demand on the security side of the business, in particular. And how should we think about the software-only exposure for security compared to the software exposure for the core load balancing business that we've touched upon already a little bit?

Well, I think just on the last point of your question, Jim, the -- our mix in software security is higher than in load balancing because we have such a strong installed base of hardware in load balancing. So software securities and security in general, but software security, in particular, is a very strong growth vector for us. And it's driven by where we have chosen to play in security. And our focus is on applications. If you just look at -- if you step back and you look at what have we learned over the last 18 months, we've learned that the way we bank and work and shop and live and exercise is increasingly becoming more digital and will continue to be. That's a secular force. So applications are going to be the way we do a bunch of things. And if you look at everything that's happened in the last 9 months about ransomware and the importance of securing applications, security is kind of a big other secular force out there. So -- and we've positioned F5, we believe, to be at the vortex of these 2 things, applications and security. And so our experience of that in the marketplace is we are seeing more demand for fraud, because fraud attacks for a number of B2B and B2C companies continue to increase. We are seeing more demand to stop automated attacks because increasingly, those attacks are very sophisticated and they are automated. And increasingly, they are at the application layer, not just to the network layer, but the application layer. And so that creates demand for Shape in its purest form, but it also creates demand for security on BIG-IP and NGINX. And that's the accelerated growth we're seeing in security.

All right, Frank, we'll let Francois take a little bit of a break. I did want to shift over to Horizon 2 quickly. Horizon 2 called for kind of this mid- to high single-digit systems decline, which as we think about for next fiscal year, would imply a mid-teens decline for systems. I know you're not going to change any guide here on this call, but how should we think about really the sustainability of growth at the levels you're seeing today, especially with what we really just talked about with those Tier 1 verticals including 5G and carriers? And really, Francois actually on that last note, how is F5's architecture that's enabled to service a carrier like AT&T moving part of its core to the cloud as well as keeping some of it on-prem?

Yes. Let me start with the first part of that question, Jim, and then I'll let Francois touch on the specific architecture for carriers. So we -- even when we were declining double digits in those years, we talked about that there was always going to be a core demand for our systems, just based off of a number of sustainable systems drivers, and that remains the case today. If you take a look at the first 3 quarters of FY '21, we have seen a number of factors that's actually led to growth, well ahead of what we are expecting when we sat down with investors in November of 2020. And this combination of some sustainable and some transitory drivers led to 17% growth in Q2 and 13% growth in the most recent quarter that we reported. And while we don't expect to maintain this level of growth in the systems business over a longer period of time, I think it's very clear that we're going to outperform that initial expectation of what we said in that high to mid-single-digit decline in the Horizon 2 time frame. And we'll have more to say on that in a month or so during our call. But I think it's important that we touch on just some of the factors that are driving that overall system demand. So macro, Francois touched on this, but just digital acceleration in general that I think, frankly, COVID has led people to even more rapid digital transformation adoption. We certainly see that in our enterprise customer base and even in our government customers. We think that there are some transitory macro things that will help in terms of resurgence spending post-COVID as well as economic stimulus that's gone in more broadly. But that digital acceleration is going to be a long-term sort of sustaining factor in our systems business. And then on the customer side, in particular, that additional capacity for expansion in your applications, particularly in your digital and SaaS companies as well as service providers, and then customers embracing that hybrid multi-cloud future, is also going to be sustaining over a longer period of time. And then finally, the imperative over an update or refresh in the security stance is going to be a long-term factor of growing the system side of the business. And then finally, for us, specifically, we've just seen really strong in-quarter close and pipeline. We think that the software positioning of our business has actually enhanced the brand equity and seeing given legs to our systems business. And so the fact that people can buy that next generation of systems when it comes out or refresh existing systems platform now, knowing that they'll be able to migrate over to a software solution with their current vendor, does give them more confidence in that next systems purchase. So we are -- we think all of those are longer-term factors that have led to growth and will continue to see growth in our Horizon 2 outlook. Francois, I don't know if you want to talk specifically about the 5G architectures?

Yes. I think the -- Jim, so the -- when you look at 5G, to take full advantage of the opportunities that are presented by 5G, things like IoT use cases in the enterprise, things like network slicing and things like the -- potentially the new world of applications that 5G enables. It does require, frankly, capabilities in networks, which carriers have, but also strong capabilities in IT. And so I think you're going to see the ecosystem evolves. And the announcement like the AT&T one and as you already mentioned, I think we'll see more announcements around carriers partnering with players that have more of an IT and cloud DNA and heritage to bring those two things together to be able to really take full advantage of the new opportunity. In the case of F5, we feel very well positioned for this because as you all know, F5 has very strong heritage in networking, and we have been in service provider networks for a long time. But we also have a very strong position in IT and not just legacy IT, but some of the most modern IT practices around microservices, cloud orchestration and so forth. And by the way, we've accelerated those capabilities with the acquisition of Volterra. And so the opportunity for us, I would say, is really, if I put it in 3 chapters, right now, what we're seeing is upgrades on 4G core networks, hardware capacity, upgrades in existing architectures because of the incremental traffic that's starting to come from 5G devices and 5G radio access network. The second chapter of the opportunity for us is going to be when 5G cores are virtualized. And we have already some design wins in this area to help carriers move to a fully virtual core. We have virtualized functions that F5 has that are needed in these 5G cores, and that will turn into tailwinds for our software business. My expectation, we will start to see that in the back half of 2022. Beyond that, our Volterra platform is generating quite a bit of interest from service providers, because it is a way for them to also leverage a universal edge platform to insert their capabilities at the edge and have some 5G capabilities in a box to deliver to their customers. And I think that will be an opportunity for us down the road.

Got it. I know we're pretty much out of time, but I did want to sneak in two quick ones. Obviously, you talked about F5 being a secure application delivery company. Should we expect more product development to come about or perhaps even M&A on more of the delivery and microservices side or on the security side of the business? Then additionally, Frank, how are you guys thinking about more software disclosures as we move into next fiscal year?

I'll take the first part, Frank. Being -- in terms of product development, Jim, first of all, we're very happy with the portfolio we have now, because if you go back to where we were -- when we were addressing just the traditional ADC market for just traditional applications, primarily on-prem, that was a $2 billion market with low growth. When you look at the markets we are addressing now, I mean we expect them to be around $30 billion by 2023. Now within that $30 billion, there's a bunch of integrations and things we're doing to capture a bigger slice of that addressable market. For example, we are porting Shape capabilities into BIG-IP, making Shape available in NGINX. We're porting our security stack on Volterra, as I mentioned. Some of the API capabilities of NGINX are going to find their way on the Volterra platform. Some of the BIG-IP capabilities will find a way on the Volterra platform to be delivered as a service for customers that want a pure SaaS consumption model. So our focus is really around these organic developments. Security is a space that's moving always very fast. And so we'll continue to prioritize organic development in security beyond even the capabilities we have today. But there may be a time and place where we feel that for time-to-market reasons, there's an opportunity we want to acquire rather than the build. And in the model, we have left ourselves the flexibility to do that. We've shared we intend to use about half of our free cash flows for share buyback and the other half to have the flexibility, if needed, to move on an acquisition.

Yes Jim, and the second part of your question, about as the -- as more of our software is driven by ratable consumption models, we'll continue to evolve what we disclosed for metrics to investors. Today, it's a small portion of our software revenue that's driven by a pure ratable approach. So ARR and net new ARR can be confusing without the context of some of the other components. And so for this reason, to date, we've really focused on total software subscription revenue and that mix, but we are evaluating additional metrics, and we'll continue to explore which ones to share as we continue to track those and evaluate those internally.

Got it. Well, sounds great, guys. Keep up the great work, as always, and thanks for joining us. I believe that we'll actually end the fireside discussions for today, but special thanks to you, Francois and Frank, for joining us. And take care, everybody.

Thanks so much, Jim.

Thank you, Jim.