Cisco Systems, Inc. (CSCO) Earnings Call Transcript & Summary

Yes. It's going actually really well. The teams that have built the Silicon One technology is -- that they are so talented and have done just an exceptional job. And we have -- what we announced, as you know, is we would actually present our technology to our customers. And in this case, it's primarily web scale and large service providers who are interested in this, which would be -- we'll sell them integrated systems like the 8000 Series. We'll sell them our silicon to go into white box, white box switches, white box routers, whatever they like to do. Or we actually will sell them our software that they can run on their own hardware platforms. And we've actually been successful with the 8000 in certain franchises. We've got -- we now have, I think, shipped 14,400 Gig ports. So that's moving forward. We have our software running on certain customers' home-built hardware. So that's -- in many cases, that's in labs and some of the work that's going on. And then we have customers who have bought silicon that are piloting some white box solutions with our silicon, as many of the customers are looking for silicon diversity, as you know. And so it's going well. I think it was a big contributor to the cloud performance this quarter in that segment. And so I think it's the right strategy. I think customers appreciate the flexibility. And in many cases, customers, you believe they're going to go the silicon route, and they test the silicon. And there's a lot of work that has to go on with writing their automation platforms. They have to write into the software development kits of the silicon. So that commitment is a really big deal. And so what a lot -- what we see is a lot of them begin looking at silicon and ultimately buy either integrated systems or a combination of some integrated systems and some white boxes with silicon. So it's -- we're in the very early days of this, but we're very pleased with the progress so far.

Have you seen anybody, just out of curiosity, try to use their -- try to get you to shoehorn other people's operating system onto your silicon? Maybe they like your silicon, but they use a different operating system.

Yes. There's an open source networking operating system called SONiC that we have run on top of our hardware. So absolutely, that is another option that we've exposed to customers as well.

Okay. Since we're on the data center track, I'll ask you about 400 gig a little bit here. So how does the -- and we were -- by the way, we were really positively surprised by the proportion of revenue in service provider, that 25% that's web scale. That was a lot higher than I thought it would be. Or that it seems like you're doing a lot better there than maybe any of us thought. But I'm curious how that business develops this year for you. We're expecting a rebound in hyperscale spending. That spending continues to be pretty strong. What do you think happens in '21? Does that accelerate further? What do you expect there?

Well, I mean, I think there's a few things to unpack. First of all, I think if you believe that there is a continued transition to the cloud, then they're going to need more infrastructure, right? They're going to need more -- they're going to spend more capital to support that transition. For us, I think the good news is it's one of the few areas in our business where we have a big opportunity to gain share. So that's another positive. I think that the one thing to remember is -- and the reason we gave a trailing 12-month growth rate is, this is going to be a lumpy business. Because if you look at the prior 4 quarters, not this quarter, but the prior 4, we saw growth rates anywhere from 17% to 74%. So it's going to be highly dependent on in which quarters these big customers are actually making purchases for certain franchises. But that being said, I think they're all -- when you begin to upgrade your core, every layer of your network, data center interconnect, the wide area, everything has to upgrade correspondingly. I had a friend years ago who said, "Networks are just a series of moving bottlenecks that you continue to try to solve." And I think that what you -- when you start putting in the 400 gig technology, it just creates the need to upgrade other tiers in the cloud infrastructure. So I would suspect that this will be an opportunity for us, and we just have to continue to deliver on the technology. And I feel good about where we are, particularly relative to where we were. And we wanted to -- we needed to share the 25% because, first of all, when you see the service provider segment up 5%, the web scale portion of it was a significant driver. And I knew for a fact if I shared with you that it was up 100%, you were going to ask me if that was on $15 or was it meaningful. And so we wanted to make sure you knew that it was 1/4 of that business. So -- and that really did drive the 5%. That was a huge driver of the 5% growth in service provider this quarter.

That's great. Well, what about where you're seeing traction? You talked about people buying switches more than -- maybe they look at the silicon, they buy a switch instead. But where are those things being installed in these networks, in the hyperscale networks? Are we talking top of rack? You see more aggregation activity, WAN? What layer of the network are you seeing more activity in?

Well, you tend to see -- like top of rack, there's a lot of discussion around taking the silicon and doing white box. In the core networking infrastructure is where you see these boxes going in. You see them going in -- you see upgrades going in and interconnect between the data centers where you need to have high balance. So it's really at different phases of every tier. Many of these cloud players obviously have, like, they define tiers of their network Tier 0, 1, 2, 3. And we're actually -- there's defined pathways to own projects for all of those tiers. And so we're working through all of them. So there's a lot going on over the next few years. And I think it's just like we've seen with service providers over the years, you just have these constant upgrade cycles year after year after year. And I think we finally put ourselves in a position where we have great technology where we can be super competitive, and we just have to go win those franchises.

Okay. That's great. So let's talk a little bit about the nonhyperscale, nonweb scale part of service provider, the telcos and cable operators. One of the things that we've been tracking pretty carefully is the development of the disaggregation of hardware and software and routing now. AT&T's recently announced they're going to use your iOS XR software on white boxes for routing. But wonder if you could talk a little bit about that trend. Do you see that elsewhere? How interested is Cisco in that business? Where in the networks do you think it lives? Does it live more in the edge of the network? Are we talking more core routing as a possibility? Just kind of talk us through what's happening there.

Yes, that's the -- I would say it's primarily going to be at the edge. You see it in the enterprise, too, even with SD-WAN, where you have the ability to disaggregate the software and run it on sort of x86 hardware in performance-agnostic kind of environments, not agnostic, but where you don't need super high performance. One of the interesting caveats I have to make around sort of our focus on software is that we still build incredible hardware. And the need for incredible hardware is somewhat underappreciated in the industry today. I listen to some of the pundits on the TV shows, and they're very dismissive of hardware. But I will tell you that they certainly appreciate the performance that the Internet provides them every day, and that is all based on incredibly fast hardware that's running underneath as well as very solid software. So I think that you have the ability to disaggregate where you don't need super high performance. I think that it's much more difficult to do in the core where you need really tight coupling with the software and the underlying hardware to optimize the performance. But for us, you asked how interested we are in that business. I said over the last few years is we're going to provide our technology to our customers in the way in which they like to consume it. So it's very simple to me, and I told our teams years ago. It's like, if they want to buy software, and we want to be religious about it, then we'll sell them nothing. If they want to buy software, we should provide them the best software we can possibly provide, which is what we've done, which is a great relationship we have with our friends at AT&T, as the example that you gave.

Okay. And then more broadly, on telco spending, I mean, just kind of focusing on the U.S., but I guess, other regions, I'd be interested in your thoughts there, too. But the C-band auctions were, let's just say, a lot more expensive than anybody thought they would be. And we're wondering how that affects capital spending this year, CapEx for the telcos. How you see that translating through to your own revenues in the U.S. And then maybe you could comment on what's going on with telcos for you globally as well.

Well, you don't spend a lot of money on spectrum and then stop. So I think that the investments will follow. We talked a lot about the 5G build-out, and it is actually happening, which is good news. We talked a lot about it over the years and tried to prognosticate as to when it would start. The good news is it has begun. I'd say the primary focus for the carriers today is in their radio networks, the packet core technology, which we do provide, the orchestration, which we do provide, as well as sort of mobile backhaul. Very few have moved into the core upgrades. Some have. But we're working with, I think, 35 providers around the world right now on some aspect of their 5G infrastructure. And I think that as we move further into these deployments, that's actually when it becomes more positive for us. Because that's when the core networks have to be upgraded to accommodate the bandwidth requirements and the increased traffic flows. And as we see more and more applications developed for 5G, that will put load on these networks. And while we can't see it today, I think if you go back to 2007, 2008, when 4G came out, no one had any idea what's going to happen with the application on these devices. And then think about what we do today with them. So I think that we don't know what innovation is going to come yet. But as that happens and as they deliver enterprise services, we will benefit from helping them deliver those enterprise services. Because that's where we have tremendous value, where we have synergy between a service provider and our enterprise base and also the infrastructure to support the bandwidth. So we're optimistic about the next few years for 5G build-outs.

Okay. And one thing I guess I wanted to ask you about 5G is where you see yourself participating in the most? Is it mostly the packet core? I know in Japan, you've even done some end-to-end systems integration. Like where is the biggest opportunity for Cisco as 5G continues to roll out?

Yes. I think that -- look, we have -- we're going to continue to add to our portfolio so that we can provide more and more elements of the 5G infrastructure for our customers. I think in the traditional world, you'll see us participate in most -- every aspect with the exception of the radio network. And then as we move to open RAN, we'll continue to look at opportunities. You mentioned Japan. And we have done a lot of work on orchestration, packet core, backbone, data center equipment, those -- the disaggregated radio network that requires data center infrastructure and cloud orchestration. We've done that work. So I think that -- it's going to provide us opportunities up and down the software stack as well into the high-performance hardware that we build in the core networks. So lots of opportunity.

Okay. And just wondering, when C-band rolls out, a lot of that will be small cells, probably in dense urban areas. We don't know for sure, probably. Just curious, in those small cell deployments, where does Cisco play? Where do you have the most opportunity there, do you think?

Yes. I think that -- look, aggregating the underlying infrastructure to support that bandwidth will always be a key element of what we do. I think the extent -- how that market plays out, we're watching very closely. We're looking at our options in that space. We'll certainly look at things like margin structure as to whether it makes sense for us to get into some of those areas. So it's a little early, but our teams are deep in the analysis, and we'll share more as we make some decisions over the coming months.

Okay. The other thing we've been thinking about, Chuck, quite a bit is we look at the cost of the hyperscale, the web scale people's networks, particularly their core networks. But those networks are getting more and more extensive, as you know, over time. And then we think about the cost of providing that network capacity versus the cost for a telco. And it just seems like it probably is a lot cheaper economically for the web scale guys to deliver that capacity. And then the kind of capacity they're delivering is huge relative to telco networks. So just curious, strategically, where do you see this going in the future. Do you think that telcos might eventually start to lease or use their networks, ride networks on top of some of these big web scale networks that are being built? Or do you think they mostly just keep their networks completely separate from those web scale networks over time?

Yes. It's a great question. It's an emerging topic. What I would say is that, first and foremost, is enterprises are effectively building virtual networks on top of the Internet, right? I mean that's what SD-WAN, and that's where we're headed, is that we're creating virtual secure connections as opposed to dedicated point-to-point. And I think you could see over the next few years, any combination of carrier, cloud, backbone, for transit, I think you'll see all of those as options for customers for moving traffic between locations, given that it's a virtual overlay. And I think that what we want to do is build our technology in a way that's agnostic to whatever that customer chooses and then partner with all of those players to provide that capability to our customers. The one thing we're hearing from enterprises is that to the extent they make a choice for one provider or another provider, they just want to maintain future optionality. What they don't want is to get locked in to a certain architecture that reduces their flexibility over time. As you know, they love optionality relative to how they move workloads. And they have applications running in multiple public clouds. They have stand-alone SaaS providers. They've got private data center apps. They've got increasing -- you're going to see, increasingly, elements of applications running at the edge. So they're going to -- they really want optionality. But the good news is there are going to be lots of options relative to how they move traffic around the world.

Okay. We've got about 5 minutes left, and I want to make sure I remind people that there is an opportunity to ask questions. [Operator Instructions] I've got one here for you, Chuck, from the audience -- well, a couple, actually. One of them is on ESG priorities. This question -- the question is, what are your top ESG priorities? And then how is management compensation tied to ESG?

Yes. It's a great question. So we're super-focused on sustainability. We've been under -- we've had significant initiatives around greenhouse gas reductions. We're 100% renewable in the U.S. We're 100% renewable in certain countries in Europe. Our products are -- we're working on and have implemented a lot of the circular economy elements relative to getting our products to our customers, out of our customers and keeping them out of land fields. So sustainability is a big deal. And I think if you've seen, in 2019, Barron's rated us the #1 most sustainable company in the U.S. And I think last year, we were #2. So that's certainly important. Diversity is incredibly important. And we've been making progress, both inside the company. 27% of our employee base now is female. And I'll tell you, that's been a -- that's a difficult number to move when your headcount is relatively flat, which is what ours has been over the last few years. But we've taken it from, I think, 20% to 27%, and a big focus there. There's obviously ethnicity focus now for us, both in our boardroom as well as at our executive levels. And governance has always been a super important issue for us. And giving back to our communities, it's been something that's been very important. So all of those things are critical to us going forward. The good news is these are not things that we're starting to focus on now because of the focus on ESG. I'd like to think our teams have been leading on this. And there's an element in our performance, that is every one of the executive officers, in particular, has an element of this in their actual priorities for each year, that are assessed every year before we make decisions around bonus payouts. And so it's a big focus.

All right. That's great. And then I've got one other question from the audience is, on Wi-Fi 6, how does that drive -- or how do you see that driving the industrial segment? Does Wi-Fi 6 enable industrial automation? Do you see a lot of that? Or is it more in campus deployments and that sort of thing?

That's a good question. I mean clearly, it's in campus deployments. And I think that just gives you big natural upgrade there. In the industrial space, what you'll see is for nonlatency-dependent applications, Wi-Fi 6 will be fantastic. But this is where you hear the discussion around private 5G. And we're working hard on those kinds of solutions. Our teams are building technology that will accommodate both Wi-Fi 6 or private 5G, depending on what the customer would like to deploy. And so I think it's a combination. It's a little -- if it's not latency-dependent, I think Wi-Fi 6 will be a great solution. If it is latency-specific, that's where you're going to see us work on these private 5G solutions in areas like manufacturing, where you need real-time and very low-latency communication.

Great. Okay. Well, I always like to wrap this up with your long-term outlook. I think I asked you 10 years [Audio Gap] But what do you see -- what is Cisco going to look like in 5 years? How does COVID change that? How has it changed your vision of the company and where the company is headed in the next 5 years?

Yes. I think that -- look, we're going to continue on the path we've been on. This transition that we've been going through is -- when you've operated in a certain way for 30 years, it takes a little while. So as I said on our Q4 earnings call, I think the good news is we achieved 51% of our business coming from software and services. The bad news is 49% still isn't. So we got a ways to go. And some of that's still going to be the high-performance hardware, but I think continuing to add more and more software into our business model. I'd say the thing that will likely accelerate as a result of COVID is our delivery of a lot of our core technologies as a service. And I talked about -- think of taking SD-WAN and cloud security and tightly integrating those and offering that as a secure cloud on-ramp service for our customers as opposed to them buying the technology and integrating them themselves. So I think that you'll continue to see more and more as-a-service technology coming out of us in the business model and an increased, continued focus on increasing the software content and, frankly, still building the best hardware that we need to run the Internet at the speeds that are going to be required over the next few years.

All right. That's great, Chuck. Thank you for all of that. Thanks again for joining us at 5:00 in the morning. I know it's a big ask to ask you to wake up that early. So really appreciate it, and thank you, everybody, for joining us.

Thanks, Rod. All good. Talk to you soon, buddy.