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

All right. I think we're going to get going here. So welcome, everyone. Thanks for coming. Good turnout for the last day for F5 here. So that's great. My name is Rod Hall. I'm the infrastructure technology analyst at Goldman Sachs. I have the great pleasure of having Francois Locoh-Donou with us. Francois, welcome.

Thank you, Rod.

I'd say, it's great to see you in person. It's been a long time.

I know. It's good to be here.

The CEO of F5. So I think you have a safe harbor you wanted to read and then we'll jump into it.

If you don't mind, Rod, thank you. So before I respond to your great questions, 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 these risk factors. Thank you, Rod.

Great. All right. Well, let's jump into it. So I thought the way to kick off here would be -- I'd like to give you a chance to kind of say whatever you want about F5 and kind of where things stand. I personally thought maybe the background on the software transition because that's been such a big focus for the company, would be a good place to get going. So maybe if you want to say a few words about that, anything else you'd like to say about kind of where the company is today as you see it.

Thank you, Rod, and it is a good place to start. Going back in time, what we saw over the last 5 years was that the world of applications was going to change pretty substantially. And for F5, we have been primarily a hardware infrastructure vendor, supporting applications but largely mission-critical applications in the data center. And what we saw was that applications would now be deployed, not just in data centers, but in public clouds, in private clouds and increasingly at the Edge. And so the bet we made was that if we were in a position to provide security and delivery solutions for every application and every API anywhere, then the market -- the addressable market for the company would be much larger. So we bet on the growth of applications, and we bet on the more distributed nature of applications going forward. And we are seeing both happening in real time. The pandemic, of course, has accelerated the growth in applications. Because all companies have put extraordinary focus on their digital channels and doing more business and transacting more with their customers through applications. And of course, applications are becoming more distributed, not just because they are lifted and shifted in public clouds, but because they are now increasingly in this container native, more microservices type environments, which means they live across a multiple set of places. And so F5 now is in a position to support all of these applications to secure and deliver all of these applications regardless of where they live, both in hardware and in software. Now of course, the major growth driver for us has been software and will continue to be software. Software is now -- we've reached a point where it's now actually more than 50% of our product revenue. And we think software will continue to be the growth engine for the company because a lot of these newer applications are built in environments that require software, either it's perpetual software or term subscription or SaaS solutions.

Yes. No, I mean the history of F5, it's always kind of in my mind, been a software company in a way because you've had software margins. And it just so happens that historically, the software was packaged in hardware. Now you're moving to software packaging. I mean, that's kind of how I've always thought about the company, so great.

That's right. Yes.

So strategically, you've grown your software offering through organic growth, but you've also made some acquisitions like NGINX, Shape, Volterra. So could you give me an update or give us an update on how you're seeing that portfolio coming together and how things look from here? Maybe good to start with NGINX and how that fits into the portfolio? But anything you'd like to say about those acquisitions and how they fit together?

Well, yes. So the -- so when we embarked on this mission to secure and deliver every application anywhere, one of the things we start making build versus buy assessments. Of course, we always want to do things organically, but there are times where you know you're not going to get there organically and you can accelerate things for your customers with acquisitions. NGINX was really -- the goal was to get into container-native environments. And NGINX had a form factor that was a perfect fit for container-native environments, and we felt we could build on top of that security capabilities and other capabilities. And so we now have seen very significant growth in these cloud-native applications that we support with NGINX, often actually married with traditional applications that are being modernized. We also felt that the -- increasingly, the challenge for our customers dealing with applications is the issue of application security. And so we wanted to broaden our security portfolio. We made the acquisition of Shape Security, and it's now allowed us to have a very -- probably the most complete application security portfolio in the industry, covering web application, firewall, encryption, description, API protection, DDoS protection, and we can now serve all applications with these security capabilities. And more recently, we have acquired Volterra to be able to offer all of these capabilities to our customers as a SaaS solution. And we spent the last couple of years doing a lot of integration to make sure that F5's capabilities are available as SaaS to our customers. When you forward to where all of this is going, the ultimate challenge that our customers have, our customers are generally large enterprises and service providers around the world, the ultimate challenge they have is that the world of managing and securing applications is still largely very manual because applications are more distributed. They're in different environments. They're dealing with tools fall, a lot of manual handoffs between teams. And what we want to do is bring way more intelligence, way more automation to the world of delivering and securing and managing applications. Not unlike what, for example, Workday has done to the world of talent management or what Salesforce has done to the world of selling, bringing more intelligence and analytics to these environments. Ultimately, we think the same is going to happen to the world of applications. And we are positioning F5 to be the company or one of the companies that brings that level of intelligence and automation and analytics to the world of applications. But to be able to do that, you have to be able to support all of these applications, regardless of the environment they're in, regardless of the form factor that they require and integrate the portfolio to be able to give your customers a single pane of glass through which they can see what's happening with all their applications. And that's really what we're going. That's what we've called the world of adaptive applications. It's to bring that level of intelligence and automation, and we think that's the challenge that customers will face.

One of the -- that kind of raises in my mind a question about the -- how far back toward the development process you might reach, there's the -- I was talking to somebody earlier about DevOps tools and how that's such an interesting area and probably persist through whatever kind of cycles we might have. But I'm curious how early in the application development stage you might like to go. I mean would you -- I would assume you want to pick up right when the application is being developed because nowadays they get developed, they get deployed and then they scale.

Yes. So historically, and I would say probably still for the foreseeable future, F5 gets involved more when an application is going into production. So we're not in the development side of applications, but we are in the runtime side of things, where applications are going into production. However, we have moved along more and more left in that environment, especially with NGINX, where we are right there when -- both when the application is born and shortly when it's going into production, and increasingly, as you know, security is shifting left. And NGINX actually has been a great vehicle to bring security earlier in the development life cycle of the applications or early when it's going into production. So for now, we're still more involved into production. But over time, I think our scope will grow.

I was going to say because we always -- we do a lot of -- we do some development on my team. So we do quite a bit of data science work, work in data analytics, and we'll develop something. And then there's always this hiccup as we deploy it, and we can do that very easily now with cloud services and so on, but then there's this hiccup where we want to scale it down. And that's where the load balancing and all these -- and security and so on gets applied. But that always -- it's always kind of a -- it's not a very easy process to get from that first workload that we initiated to something that's a little bit more scaled. So...

And it's -- one of the things you see -- I'm not surprised by your experience there because a lot of our customers are experiencing the same thing. And what you see is a lot of these tools that allow people to build applications, at least early prototype quickly and especially if you're building containers, Kubernetes is a good example of that, are phenomenal, but they don't deal with networking at scale, failover security, all of the issues that a large enterprise really cares about when you're going into production. And so NGINX for example, has become a very powerful complement of Kubernetes precisely to deal with that issue. But even beyond that, one of the reasons that we have built the portfolio that we have built, including offering our security capabilities and Software-as-a-Service, is precisely because we saw over the last few years a number of developers for speed reasons. Say, a company had standardized in production on F5, but developers for speed reason would want to go into a public cloud, build an application quickly and use native tools there because it's faster, it's cheaper, it's free. But then when they want to turn over the application to their teams and go into production, all of the friction that you mentioned happened. Well, you haven't thought about security, you haven't thought about resilience, you haven't thought about all these things. But there wasn't at the time a form factor of F5 that you can standardize on and say to your developers, go use F5 and start there. But now you can because of these capabilities that we've built, and it allows us to basically cover all the angles for a large enterprise.

I look forward to seeing that deployed, so my life gets easier. So I wanted to ask you a little bit about your strategic thinking, security versus ADC or application load balancing or application delivery sort of products. A couple of points to make to start the question off or start -- maybe start -- queue your answer a little bit. You bought Shape clearly a security play there. We've had COVID. Security has gotten a lot more interesting or it's always been interesting, but it's morphing, it is something a little bit different, especially as people work remotely and so on. So I'm curious if your strategic thinking around security and the importance of security versus some of the other things that F5 can do has changed as we have moved through COVID this last couple of years? Or if you look back 2 years ago, would it have been about the same in terms of your thinking?

So a couple of things. First of all, part of our fundamental belief system is that there is complementarity between the 2. We got into security because F5 has kind of had a strong secret sauce around proxy technology that is technology that sits in front of applications and intercepts all the traffic and makes decisions about what to do with that traffic. That is a great position to build security from. Now historically, it would have been more of our application delivery technologies would drag on security. More and more, we're seeing a lot of our net new use case or security and they often attach application delivery use cases. And so if you look at the focus and evolution of the company, yes, it is increasingly more security centric because more and more of the challenges that our customers face are security related for their applications. So as applications move into public clouds, the security, the attack surface grows. As applications get distributed into micro services, and now there's all kinds of east, west traffic between APIs, even within the cluster of applications, the attack surface grows again. So there's a lot of new technologies that are helping developers build applications faster, scale applications faster. But all of these new developments also increase the attack surface, which creates more opportunities for security for companies that are focused on securing applications. And that's an opportunity that we're looking to capture.

Right. Okay. That's helpful. So it's kind of a situation where the application delivery part of this, let's call it, the technology stack for you is there. And now -- but there's -- security is this ever evolving thing, and you're in a great position to kind of continue to evolve with those trends. Is that a good way to say it?

Absolutely, absolutely, yes.

Well, no presentation, no fireside chat, although I always think these are funny but there's no fire, but it's a fireside chat. None would be complete at this conference without asking about demand. And how you see demand developing? We've seen a couple of companies, Dell, Pure, talking about long -- elongated sales cycles. At the conference here, Pure was saying, but those sales are still completing. It's just taking longer to -- people longer to make decisions. It seems like the companies that have talked about it have said that it's mostly for large enterprises. They every -- commercial size companies, small companies are really not affected. And then there are a bunch of companies that are saying, hey, we haven't seen anything change. So I wonder if you could comment on how you see the demand environment, what kind of conversations you're having with enterprises these days and maybe bigger picture on their budgets, not necessarily just with regards to F5.

Yes. What we said that we had seen -- generally, we said the demand has remained pretty strong. On our last earnings call that was in July, we said, look, in Europe, especially, we did see this phenomenon of elongated sales cycle and more approvals and back-end loaded linearity, which we attributed to, I think, a couple of factors. One is the deterioration in currency exchange, which has continued with the dollar getting stronger for Eurozone countries. And also the fact that all equipment vendors in general because of component price have increased pricing. And so when you're in a country that has had this change in currency with the U.S., and you're facing U.S. equipment vendors, then your effective price has increased pretty substantially. And I think that, that over time, will cause those customers in large enterprise in those countries to be really careful about the prioritization of their IT spend and their projects. So that was the dynamic that we pointed to.

What do you think people -- have you had any conversations yet or is it too early to be having conversations with CIOs about how they're thinking about next year? Like are they thinking about reducing budgets? They just don't know yet? What -- any color you can give us on those kind of conversations?

I think CIOs are looking on the horizon like all of us to see what are the potential clouds. And is the recession coming or not, will it affect their budgets or not. In our case, our conversation are frankly centered on 2 areas. One is the growth in application traffic, which, frankly, that growth continues. And it's driven by the fact that most of our customers are involved in different stages of digital transformation, but becoming more IT-centric, trying to do more of their business through digital channels, and those initiatives are continuing. And the other set of conversations we're having with them is about securing those applications and the worries about ransomware, the worries about identity theft, of malware penetrating organizations, all of these things are still absolutely top of mind for our customers. So in both cases, we feel those are not really negotiable aspect. If you have demand and your applications are growing and you have more users and more traffic and more revenue coming, even if it's more expensive to have the infrastructure in these applications, you're still going to support it. And as you know, security has become a CEO and Board issue for most companies, and it's also something that generally folks want to make sure they have the right posture...

I mean big spike in security demand, for instance, when the war in the Ukraine kicked off site. So yes, the world is not feeling like a more secure place these days. I don't think so.

If anything, it's feeling less secure, with the war in Ukraine being one aspect, but also there have been a number of pretty visible breaches of SolarWinds that woke everybody to the issue of supply chain security. Log4j last December was a big factor that got everybody to realize, wow, any application that I have that is facing the web really needs to be protected. So I think the -- that dynamic is continuing.

Okay. I wanted to dig into your software growth target. You've said 35% to 40%. You reiterated that recently for fiscal year '22. I wondered if you maybe could talk a little bit. You did say there's more macro risk. I think everybody is saying that about targets right now. So you're not alone in that. Could you talk a little bit about drivers there? Like what do you think the mix of drivers there is? What are you maybe most excited about within software? What are the best short-term drivers? Maybe what are the medium-term drivers? Just talk to us a little bit about the drivers within that software growth target.

Yes. So I think the growth in cloud-native and container-native applications is an important driver for us. We serve this need with NGINX, and we also serve it with our security capabilities, some of which have been ported on NGINX today, but that continues to be an important driver of growth in software for F5. We also have -- one of the other drivers in growth is the modernization of existing applications. So this is customers that are using our BIG-IP technology and either want to lift and shift applications into the public cloud or want to modernize portions of these applications and move to software implementations of BIG-IP. A third driver is security. We've just talked about that. Largely, a lot of new security implementations we're seeing are in software when you go to -- we see a bigger security attach rate when customers move to public cloud because they're more worried about their security posture. And also when applications get more distributed, people are more worried about security. So we continue to see that as an important growth driver. In the service provider world, 5G -- 5G core, 5G implementations are a growth driver for our software. And then there are newer catalysts that I think will play out. F5 Distributed Cloud, which is our new SaaS platform, we launched that in February of this year, which initially, we focused on Software-as-a-Service for security. So all of our capabilities, web application firewalls, API protection, DDoS are all available as SaaS. And that's a newer catalyst. Of course, it's starting small. But I think over time, it will be an important catalyst of our software growth drivers because we will add more capabilities to this platform so that our customers can consume F5 as a service. And there's a number of applications that we couldn't touch in a hardware format or even in a software subscription form factor that we can absolutely serve as SaaS.

Right. Okay. One of the things I wanted to -- I've been meaning to ask you about BIG-IP is software versus hardware. One of the challenges back in the olden days before you can do BIG-IP in software was, I buy this software and instantiate it on a piece of hardware, and I get a certain amount of capacity, and I can only buy it in those units of capacity. So I'm curious whether is that one of the things that causes people to decide, hey, I want to buy it in a hardware form factor or software form factor or are there other drivers? What -- I'm just curious what drives a person or a company to decide to go software versus hardware in that situation?

I think a big -- probably the biggest driver is when you move to software-defined environment, and this isn't just F5, it would be a number of other solutions that would move to be in a software environment, generally, it allows you to have more flexibility around where you deploy your licenses. So if you have uncertainty and say, yes, I may need these licenses in -- on-prem here in this transition, but I may need to move them to a private cloud, I just don't know yet, or I may need to move them to a public cloud, I don't know yet. If you have the right software agreement, you can deploy your licenses everywhere. So it's an insurance against the uncertainty. It gives you more flexibility. It also typically allows you to automate your environment faster. If you have the right API from these software solutions into your orchestration solutions, then you can automate the service as an infrastructure personnel, you can automate the service that you deliver to developers or platform teams. And so those are a couple of the real reasons why people move to software environment. But of course, to do that, you have to be ready. Now the -- but the richness, I think that CIOs find in the portfolio that F5 has today is that we are able to serve their needs in all of these form factors, so hardware, Software as SaaS, but also in all the commercial form factors. So if you want CapEx in hardware or CapEx in perpetual license software, we can do that. If you prefer to move to an OpEx model on a subscription basis, we can do that, or in SaaS basis. And I think especially in a macro environment where things may get tighter next year, the ability to do any one of these models, I think, is going to become important because every organization is different around where they have pressure on their budgets. Some, it's on their CapEx; others, it's on their OpEx. And so the ability to move from one to the other is I think one of the strengths of the business model of F5 today.

Have you seen any trends that way? Have you seen people preferring OpEx more than CapEx or vice versa? Have you seen anything developing as this environment has gotten more uncertain? Or...

We -- what's interesting is we're seeing people move back and forth. Now I have so much pressure on my CapEx budget, but could you do that as an OpEx model for me? Yes, we could, and we could structure that way. And then later on, the pressures are on OpEx, and we're able to serve them on CapEx. This is why you've seen our subscription business in software has grown. It has grown, of course, much faster than our perpetual license business. But our perpetual license business is still there and is strong and sometimes you'll see these big deals in perpetual software, it's because people want a more of a CapEx model at a point in time.

So it's like investors, they keep changing their minds, like maybe there's going to be recession, maybe there isn't, maybe I'll do CapEx, maybe I'll do OpEx. Okay. COVID, let's talk about COVID benefits a little bit. I wonder -- when a lot of companies we talk to and then you're in the midst of COVID and the impacts of that are the lack of impacts, you have one perception. And then as you look back now on that period, let's -- I hope we're just looking back on it, and I hope we don't have too much more of it to go...

I'm not sure. I'm not sure yet.

But as you look back over the last couple of years, what are your reflections? What do you think the COVID impacts on the business have been? Or do you think there have been any?

I think there was a trend before COVID where we could see that a lot of our customers wanted to become more IT-centric. They wanted to digitize their business. They were in early stages of digital transformation processes, largely because they knew that it was the future for their business, whether you're in retail or manufacturing or healthcare or banking across all of these sectors or even automotive that we were seeing more and more of our customers built more of these applications. COVID accelerated that significantly. And I don't think we will go back from COVID. And I'll give you a couple of examples about that. Most -- say, most large banks around the world, but -- even specifically in the U.S., have been trying to move more and more of their customers to use these mobile banking applications because they have a digital engagements, and it's a better engagement with their customers. The folks who have not moved to mobile banking amongst their consumers because they were still -- absolutely wanted to go to the branch, during COVID had to move to use these mobile banking applications. Once you use them, you're not going to say, okay, now I'm going to go back to the branch and do all the things I was doing there because you now have crossed that chasm and you're using the technology. And I think we're seeing that across a number of sectors. And so there was an acceleration during COVID by necessity, which we've seen has continued around the growth in application traffic across most of our customer verticals.

So is this more of this forward shift to digitalization and the -- got it, taking the business out of the physical world.

Yes.

Okay. So product differentiation, let's talk about that a little bit. Could you talk just about what you think your competitive moat looks like. How you're protected from various different competitive threats that exist out there? And you can take it however you want it. It might be good to review a little bit what you think the key competitive threats are for the business and then talk about how you're protected against them.

I think the competitive threats to F5, I mean we're more in a competitive environment where we -- oftentimes, we can compete with folks and partner with them at the same time. The one that I think the -- if I go back over the last couple of years that the investment community has been focused on is the public cloud. It's -- the public cloud F5, your hardware company, things are moving to the public cloud, therefore, that's your competition and it's going to cause problems for us. What we actually have seen is that the public cloud has been a great thing for F5 because when our customers move some of their applications to the public cloud, first of all, generally, they move with F5 because it's too complex to do it without F5. Public cloud providers themselves recognize that, which is why we have phenomenal partnerships with them. But also, it creates a bigger security -- it creates a bigger attack surface, so we have a bigger security solutions and business coming from that. Now it's true that there are a lot of applications that are born in the cloud and will use the native tools of the cloud, some of which are competitive with us around load balancing and even some of the security capabilities. And the way we differentiate with that is really twofold. Number one is the vast majority of large enterprises don't have applications or don't want to have application in a single environment. They increasingly want to have the best environment for the app. This app is better suited for this public cloud. This app is better suited for that public cloud or even this micro service needs to be here in this public cloud and that micro service needs to be in this area. And so we are an infrastructure agnostic player. And so relative to the platform players, we can offer the same security capability, the same traffic management capability across all of these environments. And that reduces the complexity significantly for a CIO. And I would say that is -- being infrastructure agnostic and being able to support all these environments with the same technology is our primary vector of differentiation against platform players. Second vector, of course, is that we are a specialist. So application security and application delivery is all we do. So relative to a public cloud that is the Walmart of all the things you want to have there...

They would hate that...

It's probably not my best analogy, but...

Maybe.

This is a specialist versus a generalist, of course, in the areas that we have focused, we have a level of depth and feature richness and capabilities that for large enterprises makes a big difference.

Yes, that makes sense to me. What about -- VMware bought Avi Networks and integrated it with NGINX? What about that competitive angle? Obviously, things developing with VMware now too with the deal it's on. So can you comment about that a little bit? How big of a threat is that to F5?

I don't think it's a meaningful factor for us. I mean I said that before the acquisition, and it hasn't really changed. So -- but I would also say in the same way I mentioned platform vendor, VMware, it's in the same scenario. So Avi was integrated into NSX. And where I think VMware is bit focused on that solution is when a customer wants a private cloud in a box with everything in, including the rest of the VMware portfolio, they would drag Avi as part of that. But we haven't seen that been a headwind, frankly, for us because, again, large enterprises typically for traffic management and application security want best-in-class solutions. And ideally, they want the solutions to cover them across all their environments. So it's -- that's not a meaningful...

And it probably goes back to the specialist argument versus generalist argument to you specializing they're more generalized and probably don't have as much focus and particularly as Broadcom maybe acquires that business. So...

So we'll see how that plays out, but yes, it's the same argument around us, the focus that we have in the space.

Right, right. I -- another thing that I'd like to kind of ask you about is the BIG-IP historically was very -- there was a strong moat in iRules. It was very sticky. People would develop a lot of access rules and so on within it. And then it's very tough to move off, and that was one of the great, I think, foundational arguments for F5. Some of these new -- in NGINX, Shape, so on, talk to me a little bit about your thoughts on the stickiness factor there. And how are you thinking about a continuation of what's been a very good platform for the company?

So in NGINX, I think the stickiness factor is the bundle of capabilities. So you mentioned iRules as one of the factors of success of BIG-IP and absolutely, it has been. But the other factor about BIG-IP has been that it was a -- it is the technology that we've put in front of applications. And we have been able to consolidate a lot of capabilities onto a single platform, started with low balancing, but then we added authentication, web application firewall, encryption, decryption, et cetera. And so for our customers, it was, okay. Rather than having 5 or 6 vendors for this, I'm going to reduce my operational complexity. I have a single platform that can do all of that. The same thing is happening with NGINX. When you're going into production, just like you mentioned, your team were doing earlier, now you have to worry about, okay, I need resiliency, I need to be able to fill over my applications between clusters. That's the networking capability that I need. I also need to protect my API. Oh, by the way, I need to secure the applications. And so we supported a lot of these capabilities in NGINX. And so that's the stickiness factor. If you're getting that bundle of capabilities from a single very elegant, very lightweight piece of software, you don't want to go and use 5 other elements to do that. In -- on Shape, it's different. It's about, of course, protecting customers against automated attacks and bot attacks. Attackers retool very frequently. And for large enterprises where the stakes are high, they want the most sophisticated capabilities to protect against these attacks that are -- that can potentially cost millions or tens of millions of dollars. And Shape's AI technology allows us not just to protect against these attacks, but also to really understand the traffic and be able to protect against fraud. And even in some cases, be able to improve the experience of users going into an application to enhance the revenue coming from that application. So the stickiness comes there from these capabilities. And then on Distributed Cloud, which is our newer platform, that will play out over time, but we're playing the same playbook of bundling a bunch of capabilities...

More and more...

Comes with single platform -- exactly, and a single pane of glass.

Right. Yes, that's swiss army knife approach, it was very, very successful in the past. We're down to just a few minutes left. I want to be sure that we give investors here a chance to ask you questions if they have them. So does anybody in the room have any questions for Francois? If you do, raise your hand, and we'll get a mic to you. Yes, there's one here. And please say who you are as well, maybe where you're from before you ask...

Peter Jenkin from Herald Investment Management. If you can talk a little bit about the Edge environment and how that's changing. And how you fit in with sort of Edge 2 and the parts of the network that you're using to deploy that?

Yes. Thank you. So we believe that there are a number of use cases where customers will want to deploy their applications closer to their end users and closer to the Edge. These use cases, by the way, will range from use cases that you can serve from an Edge data center, which is typically what traditional CDN players or even the newer CDN players have done. But also we'll go to use cases that need to be at the far Edge where you need to be very, very close to a user, and it can't be served from an edge data center. And so the investment we've made with Volterra was really having a universal Edge platform where essentially we could run our security, our software stack to support applications on any compute footprint, whether that is in an Edge data center or whether it's in the trunk of a car or at a charging station for an electric vehicle. But basically, anywhere where an application needs to run with the lowest possible latency, that we will have that capability. So that's one aspect of an evolution of the Edge, some of these far Edge use cases that we see coming, and we are focused on as an opportunity. The other big change that we see with the Edge now is that if you go back over the last 20 years, the problem that was solved by CDNs was caching content close to users. And that was -- the problem then was users wanting to consume media, video, pictures, et cetera, or any form of application, but caching that content closer to them so the latency would be lower. Increasingly now, Edge isn't just about caching content, but it's about distributing applications and networking applications because applications have a bunch of micro services that need to reside at the Edge and need to communicate with other applications that are in other places. And so evolving and building an Edge for networking applications is different than for caching content, and that's where our focus is. Our focus, by the way, continues to be around application delivery, application networking and application security. But increasingly, having an Edge capability to do that is going to be very important. And that's where we're focused with the F5 Distributed Cloud.

So again, lightning round, and I hate to do this lightning round, the supply chain. That's the other thing that no conversation would be...

Incomplete...

Complete without, right? So where are we with supply chain? You've had your struggles with it, just like everybody else has. Can you give us kind of an update on how that looks to you now and how you see it progressing?

I would say in the first several months of -- first 6 to 7 months of 2022, we saw the supply chain situation deteriorating every month. It's where it's at, where it's been for the last few months, the last 2, 3 months is it hasn't deteriorated further. It's stabilized. But it's stabilized in a quite a difficult place in the sense that we're not getting as many components as we want to get, the -- we still are getting decommits from suppliers, and the prices we pay to get our components are extraordinarily high. And so all of these are factors that we continue to deal with, but things are not getting worse. And I would say, the other reason for optimism is that the -- generally, our suppliers have made commitments to increase their capacity over time, which we think is more of a early '23 when we really see that. We see those commitments are on track. We haven't decommitted from those things. And in our case, specifically, we've done work to design around the most constrained parts and also that work is on track. So generally, it's not getting worse, it's not great. It's stable where it is. And hopefully, in a few months' time, we'll start to see some improvements.

Yes, maybe some light at the end of the tunnel.

Yes, very.

Great. Well, we're out of time. So unfortunately, I've got a bunch of more questions, but great to have you here. Thank you so much for coming, and thanks, everybody, for attending.

Thank you. Thanks a lot.

Yes, good, Francois.

Thank you.