Arista Networks, Inc. (ANET) Earnings Call Transcript & Summary

All right. Good morning. Welcome, everyone, to Day 2 of the Deutsche Bank Tech Conference. I'm Matt Niknam, communications analyst here at DB. We're very pleased to be joined by Arista's CFO, Ita Brennan. Ita, Welcome to the conference.

Good morning. Good to be here.

Thank you. So this is going to be a fireside chat format. If anybody has any questions, just raise your hand and somebody with a mic should be walking around the room. So maybe a high-level question, Ita, just to start. As we close out 2022. Can you talk about what you're most focused on in terms of top priorities?

Yes. I mean I think supply still remains a big priority, probably will into next year, really trying to resolve things so that we can meet customer desired dates and shorten lead times, et cetera. I think that's going to be an ongoing effort here through probably into next year, and then we'll see from there, right? It's just -- we've taken lots of steps. We have seen some improvements around decommits. You saw that in our Q3 guide, but still that we commit dynamic is uncertain, and we'd like to see kind of some resolution around that. So I think supply is going to remain a key item well into next year and probably shape the business more than -- certainly more than it would normally and probably even more than demand, right? So that's important. I think from a demand perspective, it's very much continued to execute on the strategy of expanding and diversifying the business and the revenue base, both products continuing to grow that campus footprint, continuing to expand that, the product footprint. And also obviously, the customer -- diversify the customer base, and we're seeing good traction in enterprise, with enterprise customers across various verticals. And that's very important, continuing to execute on that is important, adding new logos, expanding kind of the use cases we have with the enterprise customers. And then obviously, cloud is an important part of the business. We're proud of the position we have with cloud and continuing to maintain that and continuing to work on those partnerships and those co-development type relationships which are so important there. And that's obviously a key focus and something that the engineering team is very aligned to focus on.

Got it. So obviously, macro has been very topical of late although demand, obviously, you've got a lot of secular drivers that have maybe strengthened some of the demand tailwinds in terms of growth this year. So I'm wondering, what are you currently seeing from a demand perspective across your key end markets? And maybe if you can talk about some of the major trends and themes driving recent strength.

Yes. I mean I think we still haven't seen any, I would say, kind of negative impact on the demand side, right? You have a cloud business that is coming out of a period of relatively slower investments layered on top of a new product cycle. So that demand has been healthy and continues to be healthy. On the enterprise side, you have where a share gainer logo -- new logo adds position there. So we do see that continuing. We're adding, I would say, capabilities for the enterprise. I think the alignment of kind of cloud networking with kind of what enterprises are doing from a technology perspective is very strong, and that's certainly helping with those new logo wins and additions. So again, we haven't seen any change on the demand side. I mean obviously, with all of the discussions, you have to watch it very carefully. We are watching it closely. But in terms of what we see in the business, it has remained pretty healthy so far.

Got it, got it. and so maybe we can double-click on the Cloud Titans vertical. It's obviously been one of your strongest in recent quarters, but alas, there's always concerns around the slowdown, digestion period, whatever may occur. Can you share what you're hearing from your largest customers? And maybe talk a little bit about the visibility you have into their plans?

Yes. I mean I think, honestly, supply is probably the biggest topic with those customers in this environment. So you really are in a world where you're coming off of a slower investment cycle. There is new products being deployed, new use cases, and they are anxious to obviously make the investments that they want to make and continue to deploy. That's really where the focus is with them. We've talked about having roughly 12 months of visibility because of lead times and where lead times are. That's certainly helpful in terms of planning on the supply side. So I think if anything, their cycle is getting extended just because the supply isn't kind of there to allow them to go as fast as they want to go. All of that said, I mean, we do think that cloud spending, investment in networking is up and to the right over time and grows significantly over time, is it cyclical, maybe. But again, I think you don't see that for a while in this kind of supply-constrained environment.

Yes, yes. And so maybe if we can move to enterprise. That's been another area seeing heightened investment in recent years and where you've been able to actually take a lot of share. Can we talk a little bit about demand among your enterprise base? And then also talk about, are there specific verticals that maybe you've seen more strength or maybe more pressure?

I think what's been nice of late is just the breadth of kind of that enterprise engagement, right? We used to be very focused on some very limited number of different verticals or many verticals within that enterprise space. But it has really broadened out, and you've heard that from Jayshree on the conference calls, where she's describing the different customer engagements and they're becoming broader. You're starting to see it take retail, health care, commercial, manufacturing. I mean you really are seeing more breadth in terms of the use cases. And I think, again, it comes back to this idea that now that IT is becoming more critical to those businesses, all the cloud principles apply equally in that -- in those cloud -- in those enterprise environments, right? So I think it's broader. I mean obviously, our legacy is -- we started out with the financials and then expanded from there, but it really has become much broader of late. And we continue to win kind of anchor accounts, if you like, in new verticals and then develop solutions for those verticals and then expand from there. So I think that, that breadth and expansion has been very interesting, and the team is excited about being able to kind of interact with these different verticals and come up with different solutions for those variables.

And so if we think about sort of a lot of greenfield opportunity across newer verticals, that may, I think, realistically would blunt any sort of macro concerns or macro headwinds that maybe a more mature incumbent would see. And so with that in mind, I'm just wondering, are you seeing any initial signs that a moderating macro is weighing on customer purchasing decisions?

No. I mean look, I think we have to be realistic that if there was a major macro event, then it's going to impact everything, particularly across the enterprise space, right? I don't think we've seen that to date. I think customers are -- have investment plans. They've been waiting for some time to try to put those investment plans into play because of some of the supply constraints. So we haven't seen kind of people pulling back on that dynamic, right? But again, if there's a major macro impact, then that's going to impact particularly along recession-type impacts, and that will impact businesses for sure. I just don't think we've seen that yet. And then there's this very interesting dynamic of the supply environment layered on top of what -- some of those other concerns.

What about specialty cloud? That's another area where I feel like we spend a lot of time thinking about Cloud Titans. We talk a lot about enterprise, maybe the campus opportunity. But within specialty cloud, any color you can share in terms of the latest you're seeing? And maybe talk about how demand and their purchasing cycle may vary relative to the Cloud Titans?

I mean they tend to be very similar in terms of what drives their -- they certainly follow the technology cycles in a way that's similar to the hyperscale cloud customers. So I mean, they would have very much the same approach to implementing new technologies, wanting to be at the front end of those product cycles. So we have seen strength in that part of the business as well, right? And again, not surprising because they tend to operate in a manner that's similar to what you would see from the cloud, the hyperscale cloud.

So if we kind of put this together, I think last year at the Analyst Day, you talked about a goal of mid-teens revenue CAGR between 2020 and 2025.

That's right.

And 30% growth is the guide for this year. So if we think about the strength to date, demand still seems pretty strong. What do you see as the biggest risk to hitting these targets?

Yes. I mean look, we've been -- because of supply and again, supply is the constraint, right, on the business as we sit here today, so we kind of took the decision that we kind of manage this kind of quarterly until we get some more kind of consistency on the supply side. So obviously, we didn't kind of guide Q4. We certainly haven't done anything specifically beyond that. We'll talk some more about longer-term targets at the Analyst Day in November. Yes. I think Jayshree has talked about double-digit growth next year. I think that's achievable, assuming there isn't something very disruptive on the supply side. But I mean, supply is constraining and shaping that. So we'll probably pump that question to the Analyst Day. And hopefully, we'll have at least another quarter of kind of execution on some of the decommit stuff, et cetera, and we can take a better view of that.

Got it. Okay. So I was going to cross off my next question, but I'll ask it anyway. Any additional color you can share in terms of what we can expect at the upcoming November Analyst Day?

Yes. I mean look, I don't think it's going to be that different to any -- to your typical format of business update and some technology stuff. And then obviously, we'll talk about the numbers as part of that.

So are we going to get -- I mean, is it fair to assume we may get a guide for next year? Or is it too early?

I mean we'll certainly -- I mean, I think we'll certainly take a look at what we know at that point around that.

Okay. Okay. We'll sit tight. So we touched a little bit on supply chain. Obviously, it's been really topical. I think even this morning, I was seeing headlines of another peer talking about decommits and supply chain issues resurfacing. But we've also heard some others talk about early signs of relief in terms of supply chain constraints. So can you talk to what you are seeing on that front?

Yes. I mean the situation we're in is that these small number of parts that have this kind of -- we're seeing -- I think the industry at this stage is seeing decommits across a kind of a small number of parts. And that's unpredictable by its nature, right? I mean it's -- you have a plan and then somebody decommits against the plan, right? So it's very hard to say with certainty kind of what's going to happen next, right? And that's why we've been pretty careful about kind of doing this on a quarter-by-quarter basis for now. I mean we did see some improvement. If you take the guide for Q3, there's some assumed improvement in shipments embedded in that, right? And then we'll see where we are for Q4. I think it's -- I'd like to see a number of quarters kind of together where you see some incremental improvement and some consistency before you could really claim victory, honestly on that. But we did see some improvement in Q3.

Okay. And then as of last quarter, I think you've actually got about $4.5 billion worth of noncancelable purchase commitments, and that's actually up $2 billion since the start of the year. So literally, within 6 months that number has nearly doubled. What's the rationale for such a significant commitment? And then how do you get comfortable with the risk of maybe potential obsolescence or overcommitment if there is a macro downturn?

Yes. I mean look, I think the big step up, just to clarify, that was really just time, right? You started to address 2023 and beyond more than what we've been doing before, right? I think there's -- some of those parts are multiyear commitments where we think there's obviously, some risk. It's not a risk-free strategy, but it is a risk-based strategy, right? You think about products that have long lives. You think about components that have a lot of usage across different product sets, et cetera, and you build it off of that, right? Some of it's in conjunction with customers, not all but some. We're staying very close to the key customers, understanding kind of what their plans are. But yes, I would say it's a risk-adjusted strategy in the sense that obviously, we're not doing it on products that we think have shorter life but we're looking at things that we'll be selling for many years. So if time does shift, if demand does shift, then it's really a cash commitment as opposed to an excess and obsolescence outcome. But again, it's not risk-free, but it's definitely a risk. We definitely think about it from a risk perspective.

And this is really -- does it spill into '24, in terms of the...

For some components, it's multiyear. Yes, for sure.

Got it. And then just to sort of tie this together, is there like a latest expectation you have? And I know -- I don't want to like ask you to look too much into a crystal ball. It's tough to predict, but what's sort of the latest expectation in terms of when these constraints largely dissipate and lead times maybe get closer to pre-pandemic levels?

I think on the last call, we talked about the second half of 2023, but I think that's probably our fifth guess here in the last 3 years. So it's been -- I just think there's a lot of moving pieces on the supply side and we're working through the shocks that we've had and that we know of. And then hopefully, there's nothing else that kind of comes out of the woodwork to put more pressure on the supply chain. It would be nice to get a breather for sure. But I think it is an uncertain time still.

Fingers crossed. So we talked a little bit about -- obviously, you made a lot of inroads over the last several years with Cloud Titans, with enterprise. As we take a step back, from a high level, what do you see as a key differentiator or a competitive advantage for Arista relative to peers? Why do you win?

It still all goes back to the core strategy and architecture, right, into the team, right, of building a cloud-based network that depends on software in a much greater way that software drives a lot of the quality and reliability. It drives the speed with which you can iterate and operate, right? That was hugely relevant, obviously, for cloud in the beginning and remains that way. It allows you to move through product cycles. It allows -- the team is exceptionally good at looking at new opportunities with those customers, new use cases and helping kind of build out road maps around new things. You saw that with kind of switch routing. You started to talk a little bit about AI. There's always something new that you can do with them when you have that level of talent in the organization, right? I think on the enterprise side, what's been fun for the team is that now, all of those same principles apply. It was always the theory that, that would happen, right, that pieces of that would happen. But you really are starting to see enterprises be focused on all of the same things. And the network is no longer an e-mail carrier, it's a fundamental part of the business, and it needs to perform in the same way as cloud. So anything that we can do to increase kind of their efficiency, drive simplicity in that -- across that enterprise footprint is very valuable to enterprise customers, right? And the same simplicity that was so important in cloud is now important in the enterprise as well, and the network needs to stay up, it needs to -- it can't be taken down for different reasons. You can't interrupt services. All those same principles are now relevant in the enterprise as well. And I think we do that better than the other players, right? And that's something that we continue to focus on, and we'll continue to kind of add capabilities that are important to those customers and kind of -- but all built around that same strategy and thesis of simplifying and reliability in the network.

Competitively, have you seen the landscape change at all in the current environment? And then maybe as a follow-up, is Cisco still sort of the primary competitor? Or are there others you typically see in RFPs?

I mean when you think about how the market is structured still, I mean, it's just likely to be Cisco in most cases, right? It's just they're the largest incumbent. So you'd expect to see them. That's where we see folks. And we're still focused on that kind of larger enterprise, large mid-enterprise part of the market, and that's very much kind of there where we would see that large competitor.

Yes. Okay. On the topic of WiFi, I know this maybe was a bigger topic of discussion a couple of years ago, but have you seen the competitive risk from WiFi alternatives change at all, particularly in the current supply-constrained environment?

I don't know that it's changed like fundamentally or if there's been some major change. I think it was settling into a -- kind of a make-and-buy type model in any case, even before kind of some of the supply chain stuff started to happen, right? You've heard Anshul talk a lot about this idea of co-development and kind of make-and-buy being a model that works really well for these customers. So I think that hasn't changed significantly. I think the supply and the complexity of supply in the shorter term is probably -- that's certainly something that some players are in a better position to help with than others. And again, Anshul talked a little bit just about how important it's been to be able to come to the table on the supply side as well with those customers. But I think fundamentally, when all things all shakes out, I think it's -- I don't know if there's been any real fundamental change to how that market is structured.

So we talked a little bit about the enterprise inroads you're making, specialty cloud, campus opportunity. So there's several larger growth opportunities you have in front of you over the next several years. Are there 1 or 2 where you see the most upside potential? Like what should we get most excited about?

Now you're back to the building blocks, right? I mean the whole idea is to have these building blocks of growth. We talked about this at the Analyst Day last year. You're never going to be able to predict perfectly how each of them will play out, right? And we've seen that again this year when you think about it. But having enough kind of building blocks of growth is important, right? So obviously, cloud is always going to be a big part of our business. You've got the enterprise piece, I think, is well on the way to be a consistent contributor to growth. Specialty cloud, again, making inroads, and service provider is still an area where we could do more there. We probably have more work to do, to do more there. But it's probably the piece of the business, it's the building block that you'd like to kind of add more strength to over time. So that's something that we do still. We're still making investments there. We're still doing work around that. But that's probably the one where you say, okay, can something be stronger. It's around that service provider.

On the campus side, so obviously, you've put out a target to double campus revenue, so $400 million this year. I think the commentary on some of the recent calls implies you're progressing pretty nicely, and it's been more of a supply issue than a demand issue at that. As we sort of look forward, do you need to make more investments to scale up that campus business?

I think, again, it's just like the enterprise, overall enterprise was for us. If you look back 3 years ago, people continuously questioned kind of could you actually build that business. And again, it's an incremental process, right? And you've seen that enterprise business grow very nicely. Same thing with campus, you start where your strength is and then you continue to build the business. So I think you will see us -- we're adding direct sales resources that are addressing both the enterprise data center and campus. The campus is obviously a big piece of that. And we are building some channel footprint, again, which has -- it has to be targeted. It has to be with partners that we feel like we have a win-win situation with. It will take time. Again, it's more to address kind of -- if you think about the campus market, about half of the campus market is large enterprise and then the rest is kind of mid to SMB. We're going to work our way through that TAM in the same way as we have on the data center side, and the focus initially is large enterprises, that's direct. I think we're continuing to do that. We're starting to do the channel piece to broaden that out. And we have some time to do that, and I think it will take some time to do that.

Good. So we spent a lot of time talking about growth drivers and building blocks, but let's maybe pivot the discussion to profitability and margins. Supply chain constraints, obviously, have weighed on gross margins in recent quarters. How should we think about an inflection in gross margins? When do we get there? And I maybe ask this in the context of you starting to see some of the benefits of pricing actions later in the year?

Yes. I mean look, I think the supply chain situation is -- impacts gross margin in a lot of different ways, right? You've got direct things that you can quantify easily like the expedite fees and the decommit broker purchases, which we've talked about, the 200 to 300 basis points type. But you've also got just inherent inefficiencies in your manufacturing and in your supply chain, whether it's higher freight costs or it's underutilization because of decommits and other things that kind of just increase your cost base. You need to solve the decommit problem and then you can get some momentum and then some of that will go away as well, right? Obviously, the broker impacts would go away if you were able to solve the commit problem, right? So there's a lot of things to do on that side of it, and they're something of a moving target like we talked about, right? If you could hold all of those constant which you can't do, right, then you could come and look at pricing and say, okay, now I should start to get some benefit from pricing at the back end of Q4 this year, and that would carry through into next year. So we will get some benefit from pricing, it's just hard to anchor everything else. So we'll have to look at those 2 things combined, but there is some benefit from pricing that we should start to see heading into next year.

And then longer term, as we think about a mix shift towards -- cloud side, obviously, has been growing very robustly, you're pushing more into campus. How does that affect the ability to return to that sort of 63% to 65% margin range?

Yes. I mean if you think back to -- our range originally was constructed around kind of -- heavy cloud content was 63% and that end of the range. And then as that started to come off, you'd see higher -- you saw that play out in 2021 where cloud was a smaller part of the business when we were operating at 64% or 65% gross margin. So now the cloud is back, if everything else was equal, you'd expect to be kind of at the bottom end of that range. So again, is it possible to get back to the bottom end of that range even with the heavy cloud mix? I mean it's possible, but I do think -- I wouldn't take it to the bank tomorrow because you have to resolve all of those supply disruptions that we talked about, right?

Yes. Okay. Pricing, we've referenced it. And I think you've had 2 price increases over the last year, and we talked about the first one starting to benefit in 4Q. What's the magnitude? I mean I've heard 5 to 10, I believe, for the first one. And then the second one seems like it was a little bit more specific, but maybe if you can help us frame the magnitude of those benefits.

Yes. I mean the first one was broader and was in that 5% to 10% range. The second one was more targeted to specific products where we had kind of -- because again, for us, given the customer set, you have to be pretty transparent around the cost increases and why you're doing these price increases and stuff. It's not simple to just do kind of a blanket increase and not explain kind of what the drivers are. It's more of a negotiation just given the size of a lot of the customers. So that was more targeted to products. It's still in that kind of 5% to 10% range on those products, but more targeted for sure. So yes, we're averaging out to something less than that on the overall base.

Any pushback?

I mean most customers understand this, right? And they want it to be over. I mean the pushback, I think, is more they want it to be over more so than it is that they blame us or that there's anything different competitively about this, right? But -- I mean, everybody is tired of supply issues, right? But again, I think the world we're in right now is that we're going to be wrestling with some element of that for some time.

Are you -- I don't mean to get you to telegraph anything, but as we think about -- I've heard some companies talk about the cost curve beginning to bend, others say it's still pretty inflationary. So anything sort of on the horizon in terms of incremental pricing actions you're contemplating?

I think it's all going to depend on the supply and how we see that, right? I mean, again, we've tried to base it very much on cost and what's happening with on the supply side, so that's going to drive it. And I think it's just a little too early to say whether there's been any kind of improvement there at all.

So we talked a lot about gross margins. So if we think about operating margins, I mean, you've already achieved non-GAAP op income margins of about 40%. That's well ahead of peers. As we sort of look forward, is there a longer-term target you have in mind and maybe if you envision reinvesting some of that strength here into more R&D or sales and marketing?

Yes. I mean, again, at the risk of overplaying the Analyst Day, we will talk about that more in November. Our last kind of discussion has been around a 38% operating margin, plus or minus. I mean the model is such that when cloud accelerates, obviously, we talked about the gross margin impact, but the operating margin impact of that is obviously that you get a lot of flow-through to the bottom line as well, right? So it's accretive from an operating margin perspective, even if it is -- causes some pressure at the gross margin line. We're investing on the sales and go-to-market that's mainly targeting enterprise and CapEx and some of those other things. So that cloud investment is light on sales and marketing. And obviously, it's -- the R&D is the important part of that, right? So that causes in periods where you have accelerated growth, there is a good flow-through to the bottom line, right? That's kind of a self-regulating piece of the model, if you like, right? Even though the gross margin gets pushed a bit, the operating margins are healthy.

So if we sort of boil this down to a cash flow discussion, how do you prioritize this use -- prioritize use of excess cash across the business?

I mean cash is interesting right now just because, I mean, we do have the purchase commitments. We are growing inventory a little bit. We are -- with the challenge with the decommits is, if you have 90% of the parts or 95% of the parts, you still can't ship, right? So there's -- and you saw -- we talked about a little bit on the last call, you saw it a little bit in Q2, but also I think for the next couple of quarters, you could see some of that dynamic until we get out from under the commit. So yes, so that's the first thing, is always like have enough means to do whatever you need to do to be competitive and maybe even forward thinking a little bit in the business, right, whether that's purchase commitments or from R&D or something else, right? That's priority number one. I think we have been returning cash to shareholders. I think we have like $300 million left on our second kind of $1 billion authorization. So we have been pretty active in returning cash, and we'll continue to do that. And then there's M&A. And M&A is a lever. I think we're all very conscious of the fact that it's a lever but you have to find the right opportunities and the right -- and it has to be the right fit for the business, the company, et cetera. And that's tended to drive us to smaller acquisitions, team, culture, technology aligned. And so we've done a couple of those. They've been super helpful in terms of building out, whether it was WiFi to build out the campus portfolio or it was a way to help us with security around the network. We had the visibility that came with the Big Switch acquisition. So those types of things, we will absolutely continue to look at those as well, but they've tended to be smaller in nature. There's not a ton of targets, if you like, around the network that we found to date that kind of meet all the different criteria, but we'll continue to look at that.

One question before I dig a little bit deeper into M&A. Maybe it's a less prudent question to ask in light of a current rising rate environment, but you have $500 million on your balance sheet and cash, no debt, generate almost $1 billion annually in free cash flow. So how do you think about the opportunity to maybe enhance returns via leverage?

Yes. I don't know that we're there. We actually have close to $3 billion in cash on the balance sheet and like all cash including marketable maturities and everything else, right? Yes. So I don't know that we see kind of leverage, particularly with the rate -- it's not free, right? And we have -- I think we have enough sufficient cash to do -- to drive a good kind of return program in conjunction with everything else that we're doing without driving leverage at this point.

Got it. So you touched a little bit on M&A. I'm just wondering if you can talk a little bit more about your approach. Any specific capabilities, products that you'd look to add inorganically to enhance the platform?

Yes. I think, look, we -- everybody has been around for a while on the management team. So we view M&A as a lever, right, but it has to be used carefully, right? And I think that's -- you see that kind of in what we do. So it's technology where you feel like somebody -- you can get -- you get an advantage from looking outside. WiFi was one of those where, yes, we had some WiFi resources in the team, but we didn't have a team, right? So we were able to add a team. I think the Security Awake (sic) [ Awake Security ] was an important one that -- we had a very strong networking knowledge base, but we didn't necessarily have a good security knowledge base. Adding Rahul and his team obviously helped tremendously with that, right? So now you've got that in-house knowledge and you can start to really look at kind of the security of the network and what we can do there. So that's the approach. I think the approach is -- culture is super important in getting people who have the same view on the importance of the product to the customer and then having it be adjacent in some way to the core of what we're doing, right? And obviously, then it has to meet our financial goals and everything else as well. But those other things are more important honestly, as a starting point.

Are you seeing any more sort of opportunities pop up, maybe private market valuations coming in a little bit? Just wondering in terms of what you may be seeing on that front.

I mean I do think kind of valuations are becoming more rational, right? There was a period of time there where it was difficult to get comfortable with some of the valuations. It is starting to become more rational, but again, you have to satisfy those other criteria as well.

Yes. Okay. And then last question to sort of tie this all together. If we're sitting here a year from now, what would you say you would have liked to achieve over the next 12 months?

I would like nobody to ever want to talk about supply again for at least a while. And I think just continuing to execute well in cloud and drive this expansion in the enterprise, I think those are front and center for us and being able to do both of those things that have the cross leverage between the technology that's been deployed in cloud to the enterprise is very important.

That's a great place to end it. Okay. We'll end it there. Thank you, Ita.

Okay. Thank you very much.

Appreciate it.

[ Yes, and more living for win ], but that's looking very challenged at the moment. That's worse than supply.