Arista Networks, Inc. (ANET) Barclays 2022 Global Technology, Media and Telecommunications Conference (Transcript)

Arista Networks, Inc. (NYSE:ANET) Barclays 2022 Global Technology, Media and Telecommunications Conference Call December 8, 2022 4:55 PM ET

Company Participants

Ita Brennan – Chief Financial Officer

Chuck Elliott – Director of Business Development and Investor Relations

Conference Call Participants

Timothy Long – Barclays Bank PLC

Timothy Long

Hi. Thank you, everybody for joining Tim Long here, Barclays analyst for comm equipment and IT hardware. Happy to have Ita Brennan with us, CFO of Arista, Chuck Elliott, up here as well to potentially help out if needed. But thank you for coming Ita. It’s great to have you here. It’s been an exciting few months for you guys with some really positive news. So that’s great.

Question-and-Answer Session

Q – Timothy Long

So maybe we’ll start off with that. I’m sure a lot of folks here were either at or listening to the Analyst Day, talking about 25% growth for next year. So maybe give us a little color on the strength coming off of 40% plus year. What gives you confidence in that? How do you feel about visibility? What are the pieces that are really contributing to that strong growth?

Ita Brennan

Yes. I mean, obviously, our approach to guidance has always been to kind of look at multiple different scenarios of kind of getting to the number that you end up kind of settling on and making sure that you have multiple different paths to get there, right? Because obviously, there’s things change over time, and that gives you some resilience to the number. We are obviously coming off of a period where we’ve seen good investment on the cloud side of the business. Again, coming off of a period where that spend had been lower, right? 2021 cloud was 30% of the business. So we expected cloud to recover. We expected to see that recovery kind of linked with some of the new products and some of the new products that we brought to market and it’s great to see that happen, right?

They would like to have gone faster even in 2022. I think it’s been a constrained supply environment. We have extended lead times that’s giving us a little bit more visibility to what’s happening in that business. But there’s underlying strength in that business, again, after a period of probably lower or under investment prior to that. So that’s obviously one tailwind.

If you look across the enterprise piece of the business, we are taking share on the enterprise. We’re winning new logos, adding new logos. We’re seeing very good acceptance of the overall enterprise solution now. It’s no longer just data center. It’s data center plus campus plus WiFi plus a lot of the management tools and visibility and security offerings that we’ve been doing. That complete package is being very well received in the enterprise, and we are seeing kind of good momentum in kind of the addition of new logos and then growth within those logos when we win.

You look at specialty cloud service provider, I mean again, specialty cloud is also very interested in the new products, new technologies, again, are looking to invest around those products. So you have a pretty strong momentum across the board, and it’s been pretty much constrained by what’s been a very difficult supply environment, right? And that was the dynamic in 2022, and we think that continues to some degree, at least into 2023. So I think the combination of the momentum and the diversity plus kind of the visibility because of supply all gives you multiple different ways that you can think about that 25%.

Timothy Long

Okay. Great. And then the – probably similar answers, but a little bit longer term growth of 20%. How do we read into that? Just the durability of these share gains and strength in the end markets is longer lasting or anything else that pops up that helps you over the more long-term?

Ita Brennan

Yes. And when you think about the longer term, I think what we tried to do at least is to, when you look back at the business, you see this very consistent kind of 20% plus five-year CAGR, right, from when we went public back in 2014. So again, we’re looking at kind of that 2020 to 2025 period and thinking about that as a 20% CAGR. I mean, obviously, that does imply some slowing of growth off of the levels that we’re seeing now in those outer years. It’s not that we have perfect knowledge of what exactly those numbers will be in those outer years, but we do think that there’s potentially some cyclicality to some of the kind of – particularly the cloud accelerated spending that we’re seeing now.

And that’s kind of recognizing that without necessarily thinking you have the perfect number for what that will be. And of course, internally, we’re obviously driving to grow all the other pieces of the business and drive that growth to offset that as much as we can, but we still think there is likely some deceleration in those outer years largely because of cloud.

Timothy Long

Okay. Yes, maybe we’ll dig into cloud first as a vertical. Obviously, you guys get a lot of questions on that. You mentioned the 30% last year going to 45% this year, new products that maybe unpack that a little bit, what’s really driving it? Because I think even Microsoft was kind of down a little bit for a few years there, and it seems like a big reversal year. So how much of this do you think is maybe a risk to taking some share or new workloads or new parts of the network, new products? What’s really – what’s underneath it, particularly for the big two?

Ita Brennan

Yes. I mean I think there’s – again, you had this lower investment and then COVID, which was largely a driver for a lot of these businesses, right, that COVID period and then a supply constrained kind of meshing in on top of that, right? So I think there is and now you have a new technology is again available to kind of help them grow their business and do some of the things that they want to do. So I think it’s a combination of all of those things. You have 400-gig, which obviously is being deployed into whether it’s AI use cases, data center interconnect some of the high-speed use cases.

You have more efficient, power efficient, more denser 100-gig and 200-gig configuration products coming out of that silicon as well, right? So you’ve got a whole new slew of products that are more efficient, both power and cost and address more of the capability needs that you have to support AI, et cetera. So that’s definitely a part of what’s happening. And then I think you do have this dynamic of maybe underinvestment or lower investment plus more demand through COVID coming into kind of a supply environment, which is really almost helpful in that it’s stretching out some of these deployments just because of the lead times that we’re having to deal with.

Timothy Long

Right. Okay. And then thinking about maybe a third leg to this at some point, you’re in another big cloud player. I think that business is growing pretty well. What can you say about kind of diversifying when you look at the Cloud Titan opportunity as you look out over the next few years here?

Ita Brennan

Yes. I mean we do have, what would be sizable businesses except that they’re dwarfed by the M&M customers and the role that they play. I think that there is announcements talk quite a bit about this make-and-buy model. And certainly, it’s worked very well with kind of some of the existing customers. So there is interest in that model in terms of where can you – what can the brand is Arista solutions bring to somebody who’s been traditionally more kind of white box and internal developments. That’s something that we continue to propose technology there. I think it’s a difficult change for anything to make. It takes a lot of time. It’s not in our near-term kind of numbers and projections. It’s more something that we will continue to kind of engage with and see if we can find the right model that works there.

Timothy Long

Right. And understanding you’re the financial person, maybe Chuck wants to jump in. But is there a lot more development to scale Google or an Amazon or an Apple or one of the other big hyperscale players or is it kind of just evolutionary off the similar type of products that you have now. So in addition to like proving the use cases, you also need to put some real development and co-development in it with those players.

Ita Brennan

I mean there’ll be a lot of development in a situation like that just because you’re starting from a very unique place, right? So yes, there’s definitely a ton of work that will go into that. And there’s just kind of the actual evolution how would that evolution ever happens. So again, I think it’s obviously something that’s very interesting on the one hand. But on the other hand, it’s not something that we kind of think about in the near-term.

Timothy Long

Right. Okay. And understanding you don’t talk orders or backlog, but maybe in terms of visibility, obviously, this customer base started giving you longer visibility. So where are we in that continuum? Do you think we’re at the point where we’re going to start to see maybe your visibility, maybe not go back to what it was pre-COVID, but lower than the year plus that you’re getting from this customer base?

Ita Brennan

I mean it’s all linked to supply, right, into the lead times on the supply side. Jayshree had talked about six to 12 months and the last earnings call. I think we still see some supply issues to be resolved kind of well into next year. I mean if you fast forward to whenever you solve all the supply problems, I suspect you’re going to see kind of lead times from customers come back in as well, it’s not necessarily a bad thing, right, which is back to why we don’t talk about orders and backlog too much is because lead times – if lead times are driving that, then that’s – it’s not really a leading indicator of the business, right? So I think first, it solves the supply issues and be able to kind of commit with confidence to some shorter lead times to customers, then I think you will see some of that visibility go away, but that’s probably a good thing. I mean a year is a long time in tech, right, to be waiting for products to be deployed.

Timothy Long

Right, exactly. And you could work down some of your supply commitments that you’ve put in place.

Ita Brennan

Yes, and we’ll start to do that anyway because you are seeing some lead times improve, right? It’s just you need everything to improve to be able to solve the customer problem. But we will start to see improvements and there are improvements on some parts and lead times. So we will start to kind of focus on the purchase commitments and start to manage those.

Timothy Long

Right. Okay. You did mention 400-gig in one of your earlier answers. It seems some of your competitors who are lagging you on the technology front, tried to use that as an insertion point. It doesn’t really seem like it’s happened. So maybe kind of what’s your view on and the team’s view on where we are in the 400-gig transition? And do these transitions matter for market share gain or not?

Ita Brennan

No. I think every product cycle is an opportunity for everybody, right? Yes, I think we’re very happy with kind of how the products proved in and the acceptance of the products. There’s always some new use cases that you get to play in. We’ve won some WAN opportunities, for example, to something new. The AI area and just kind of some of the technology discussions around that are very interesting. So it’s always an opportunity to do new things and to look at new things for us as well as the competition is seeing it is something where they can come to take some share from us. But for us, it’s more – you just get more capabilities to bring to these customers and figure out where can you expand kind of your share inside those accounts. And that’s been the approach for a long time, and it’s worked very well. Again, I think in this cycle where we’ve been able to kind of find places where we can really add value for these customers.

Timothy Long

Okay. And if you go underneath – if you go to the specialty cloud, are all the dynamics the same minus the white boxing as far as share opportunities, new use cases? I mean, are those still driving – being driven by that piece of the business?

Ita Brennan

Yes, there’s a lot of similarities, I think, from a technology and a needs perspective with that specialty cloud piece of the business as well, right? I think we saw good, strong kind of visibility from them as well. We’ll probably solution more of that business as we head into next year. The hyperscale has kind of come first, right?

Timothy Long

Okay. Excellent. Maybe let’s go over to the enterprise and campus part. So talk to us a little bit about the – you’ve gone from $100 to $200 to a little under $400, I guess, for campus this year, we’ll see the final numbers, I guess, in a few months. What’s driving that success? How much of this is strength that you maybe have in the data center side? And maybe just walk us through kind of what’s helped those share gains?

Ita Brennan

Yes. I mean I think just from the business perspective, we’ve seen a pretty balanced kind of data center first, campus first kind of wins in that space, which has been very gratifying wasn’t necessarily how we thought it would have played out initially. So we are seeing good wins that are being campus-driven, if you like. Maybe Chuck can kind of talk a little bit just about why the solution is resonating so well with customers across those different pieces of the business. But we are seeing good traction on a campus first basis as well, which has been good to see.

Chuck Elliott

Yes. So I mean, I think at the end of the day, customers are increasingly buying into Arista because of the simplicity of what we offer and the fact that it works. So simple means that operationally, it’s lower cost for them, guys get to – engineers get to go home Friday night instead of spending the weekend troubleshooting and the products just work. And that results in lower cost to their business. So especially cloud guy might be providing a service to their end customers. That service is up and running, their business works, business model works when it’s down because there’s an outage on the network. Somebody’s geared it worked the way it was supposed to. That’s bad. That hurts right.

So what we hear from customers is, why did you buy Arista because it works. And now why are we able to deliver the quality that makes it work? It’s because we have a single operating system, a single bit of software that runs on every platform that we sell. EOS is running everywhere. We don’t have one operating system for the campus, one for the router, one for the data center, one for security, one for this, one for that. We have one operating system that runs everywhere.

And we have one management platform, CloudVision, that runs everywhere. So you can be much more efficient and especially in this day and age where there’s a tremendous – notwithstanding the layoffs that have been going on, there’s still a tremendous shortage of technical talent out there. So if you can get by with fewer people and do a better job delivering what you’re trying to deliver from an IT perspective, that’s a big win. We give that to them.

Timothy Long

Okay.

Chuck Elliott

It’s all about.

Timothy Long

Excellent. One of the things that came out of the Analyst Day that I took from it was a little bit more optimism about wireless LAN. You’ve had Mojo for a while. So maybe talk to us about the importance of an integrated solution and having a stronger wireless LAN offering and how that helps you hit some of these targets in campus?

Ita Brennan

Yes. I mean obviously, Mojo was a starting point for us on the wireless side of the business, right? It got us a team and some technology and then we’ve been working to integrate that with CloudVision, right, and to have – because again, on the wireless side, a lot of the benefit is coming from the management and the control of that WiFi and not necessarily kind of the box and the chip itself, right, which is pretty common, right? So the – I think what we’ve seen is good progress in the ability to integrate that with CloudVision managed that with CloudVision. And now it’s kind of seen again as an extension of kind of the end-to-end solution that we’re offering. That’s resonated very well with customers, right? It took some work to get to the point where it’s fully managed by CloudVision. But I think now that, that is – it’s a very logical add-on to the networking piece of the offering.

Timothy Long

Right. Okay. Can you talk a little bit about kind of different go-to-market that’s required for campus? I mean obviously, if you want to fully go over the full TAM, you need a lot more channel. So just talk to us about your approach there, probably more selective with where you’re going to go out because Cisco, although you might have some technical issues at a pretty strong channel reach and presence. So how do you go about that?

Ita Brennan

Yes. I think it’s everything and it’s in the right time approach pretty much how we approach a lot of things, right? But if you think about the campus market today, it’s what a $12 billion kind of market and roughly half of it is in the hands of large enterprise, right? So that’s kind of our – and is held pretty closely by one player. So that’s an obvious starting point for us, right, that it would be that large enterprise customer. It’s still mostly direct sales model, strong leverage across between the network and the data center piece and the campus. So that’s kind of our starting point. That’s what’s driving kind of the growth that you see today.

At the same time, you do need to start to build some channel capability because, obviously, you’ll want to get out of that 50% at some point, right? So lots to do there. But there is the other half of the market, which is more mid-market. That’s where the channel can be even more useful in terms of driving sales. So we are kind of driving a channel effort. It’s targeted. It has to be targeted to your point because you need to find partners that can strategically kind of align with us, and they’ll probably typically be more technical, and they’ll kind of fit our overall kind of go-to-market and approach to the product. We are making some progress there. It’s going to take time. It’s not going to happen overnight. But the more pull that we create, the more except as there is a Arista across the board in the enterprise, the easier that’s going to become, right? Because partners will see that demand from end users, and they will look to be part of kind of the Arista franchise as well, right? So I think that’s something that we’ll develop over time. It’s not necessary to drive kind of our current momentum, but we do need to be making progress so that we’re able to kind of address more of that TAM over time.

Timothy Long

Right. Also at the Analyst Day, you talked about whether it’s the earnings call, I forget, orders being better than sales for this piece of the business. So just curious, your view as you’ve gone from the $100 million to $400 million-ish. Did COVID help this or hurt this piece of the business because there were times where probably you were going to be in a bake-off that didn’t happen. So what do you think the net effect was on the slope of your market share move?

Ita Brennan

Yes, it’s interesting because it’s obviously puts and takes. There was a while kind of in early COVID where we said, Wow, this could be bad for this business, right? The offices are closed, people are home, who’s going to worry about the campus, right? But then over time, you start reading about the campus is a lot of things, right? And I think what the team has done a good job of is kind of expanding where we play, right? Whether now we see a lot more retail opportunities to see a lot more healthcare, education, there’s more – I think COVID drove more pressure in some ways on campus networks and the management of campus network. So that’s been positive, right? So I think we went from a place where initially, it seemed like it could be a problem, right, there actually be more complexity now around those campus footprints, and that’s kind of an opportunity actually for us to insert.

Timothy Long

Okay. And how about some newer technologies like SD-WAN, you’re starting to talk about we’ll see something more on that next year. Is the campus an enterprise and environment that’s ripe for you to just add new TAMs to? And how do you think SD-WAN plays into this, obviously, a pretty high growth end market right now, but very crowded.

Ita Brennan

Yes. I think the approach at the end of the day is back to the simplicity that Chuck was talking about, right? I mean we tend to go into these markets and do these things when customers kind of want you to do that. So if you have a solution that’s kind of EOS into end and then there’s a piece kind of in that chain that’s not, right? There is – there tends to be over time, a pull from customers to go solution to that, right? So I think that’s what we’re trying to do with that enterprise routing offering is really to say, okay, can we give customers the ability to have the U.S. end-to-end and to have all of the visibility and other benefits that they have where we solution that with a partner today, if we can get to an integrated, more integrated solution, that will be more powerful for customers. And customers are looking for that so that they can benefit from kind of that end-to-end solutions.

Timothy Long

Right. Now will you need to add more security capabilities and IP over time because it’s obviously a meaningful piece of the SD-WAN architecture? You probably have a lot of the routing capabilities now. How about the security piece?

Ita Brennan

Yes. I think our focus on the security side will be around kind of securing those network flows, right, and securing there’s encryption, there’s visibility monitoring. We’ve been doing some interesting stuff with Awake and with the capabilities that we acquired with Awake, putting those sensors in line into switches so that you can do a better job kind of looking for anomalies in the flows of the network. I think that’s how we’re thinking about security today. We’ll see how that evolves over time. But it’s really more focusing on using the advantages that we have with the U.S. and the visibility we have with the U.S. to better monitor and look for anomalies inside the network.

Timothy Long

Right. Okay. Maybe if we could touch a little bit on telco and service provider another big TAM. You guys have had this routing journey where you start with some basic use cases and move it up. So how are your views of disrupting the more traditional telco market where there’s a few established players, but it tends to be pretty software-intensive margin-rich business. So I think it would fit very well into the model for Arista?

Ita Brennan

Yes. I think the approach has been and kind of has to be at the end of the day, is not to try to go back and retrofit to the kind of the classic router, either products or honestly, software stack to some degree, right? I think that would – we always felt like that would have been a mistake to try to go backwards. When you think about the power and the cost improvements that you’re getting with these new kind of switch router platforms, et cetera, I mean it’s clear that that’s kind of the direction that eventually a lot of this TAM will go, right? It’s taken longer probably than we would have initially thought coming into this.

We’ve done a ton of work on the routing stack for cloud, lots of new protocols. We’ve done work with service provider customers to kind of help bridge them to those new protocols. And I think that has to remain kind of the approach, right? That’s the approach that ultimately we think will work, right? The more capable – those products become – the broader that gap becomes between kind of what’s deployed traditionally and then what a telco cloud type network can bring you from cost, power, efficiency, everything else.

We do believe that eventually you’ll start to see more of those dollars get solved with that solution. And we’ll have to be patient and continue to work on helping to bridge the gap for sure. We’ve kept all the sales resources in place to kind of support those customers, and we’ll just have to be patient in terms of waiting for that transition to happen.

Timothy Long

Right. Okay. Yes, they tend to be a slow-moving bunch. So I’m assuming that there’s got to be a long-term partnership, pretty risk-averse. So…

Ita Brennan

And starting from a typical place. I mean, starting from with a lot of complexity. So it’s not an easy transition, but I think the benefits are becoming more and more compelling all the time in terms of making that transition.

Timothy Long

Right. Okay. Maybe a little bit on the financial side. I mean you have a difficult job in that your revenue growth is really high and your operating margins are really high. So it sounds like the long-term model is kind of more or less status quo on the margin front. How do you not see more leverage? And I guess, what is that extra investment OpEx dollars? Is that just going to really help fuel that revenue growth line so that we can keep that growth at an elevated level? Is that how the model works over the next few years?

Ita Brennan

Yes. I mean, part of the kind of expansion that we saw this year, et cetera, it’s just coming from the sheer acceleration in the topline growth, right? And we’re not going to spend to the peak of that, right? So some of the expansion that you’ve seen has come from that, right? I think when we think about the long-term model, hopefully, we’ll get gross margins back to what we consider a more reasonable range. We’ll see if we ever get back to exactly where we were before. It just kind of depends on what happens with the supply chain and just the complexity of supply chain and how much of that can you really peel back, but we’ll certainly look to improve gross margins, and we’ll look to continue to invest in development and sales and marketing, right?

And I think we’re reserving the right with the kind of the long-term model, the 38% kind of operating margin, we’re reserving the right to be able to make some investments if we think they’re truly beneficial to the future and maybe make some of those investments ahead of what we see on the topline and what we see on topline growth. Again, it will come, obviously, with some value that we’ll articulate et cetera, if we were to do that. But I think we are reserving the right to do that. I think it would be foolish to miss opportunities if they present themselves where we could do something additive to kind of just the normal run of the business. So we’re kind of reserving the right to do that with that 38% operating margin.

Timothy Long

Okay. And then from a capital return standpoint, you also have options there. So maybe walk us through kind of the framework and how does the company feel about M&A to supplement technology in areas where you see real revenue opportunities?

Ita Brennan

Yes. I think you’ve seen us do M&A where we see technology or people, teams, smaller, but kind of, I think, well understood assets that we’ve acquired. I think we can absolutely continue to do that. When you think about bigger and larger M&A, I think with the scope of what we have in front of us that we can kind of solution internally with organic investments, et cetera, I think that just becomes harder for it to kind of meet the bar or exceed kind of what we feel like we can do with our own resources, at least for the foreseeable future.

I think on the return of cash, we would offset dilution, and that’s kind of a baseline. And then we’ve been somewhat opportunistic around returning more cash than that. This year has been a tough cash year in the sense that you’ve had a lot of working capital, supply chain investments, et cetera. But we have still been pretty active from a return perspective. So we’ll continue. We’d like to start to make some improvements to working capital in 2023. I don’t know if we’ll get all of that done, depending on the supply chain. But certainly, we will start driving working capital metrics kind of back to something more favorable. Again, we may not get back to exactly where we were pre-COVID, but we’ll certainly look to make improvements. And then we’ll continue to kind of opportunistically return cash.

Timothy Long

Okay. Maybe one more quick one and maybe not that quick, but macro, we’re hearing from everybody, but we haven’t heard it as much from Arista. Do you think there’s just such demand for the product category now that maybe there’s a little buffer from a tougher macro environment or is it something that still keeps you up at night?

Ita Brennan

Yes. I think we should think about it and try to understand what’s happening from a macro perspective. I think there are investments that customers want to make. There’s been kind of a protracted supply environment. I think customers can continue to make those investments, if they believe there is some manageable kind of disruption from a macro perspective. If it’s more severe and longer term, I think then we’ll have to be very careful about what customers decide to do. I’m just not sure we know which path we’re on yet.

Timothy Long

Right. Okay. Ended with two seconds, you’re professional. That is amazing. Very good. Thank you, everybody, for joining. Ita, Chuck, thank you so much. Appreciate it.

Ita Brennan

Thanks, Tim. Thank you very much for having us.

Be the first to comment

Leave a Reply

Your email address will not be published.


*