Arista Networks, Inc.

Welcome to the second quarter 2020 Arista Networks Financial Results Earnings Conference Call. During the call, all participants will be in a listen-only mode. After the presentation, we will conduct a question-and-answer session. Instructions will be provided at that time. If at any time during the conference you need to reach an operator, please press the star followed by zero. As a reminder, this conference is being recorded and will be available for replay from the investor relations section at the Arista website following this call. I will now turn the call over to Mr. Curtis McKee, Director of Corporate and Investor Development.

Sir, you may begin. Thank you, Operator. Good afternoon, everyone, and thank you for joining us. With me on today's call are Jay Shrielal, Arista Network's President and Chief Executive Officer, and Ida Brennan, Arista's Chief Financial Officer. This afternoon, Arista Networks issued a press release announcing the results for its fiscal second quarter ending June 30, 2020. If you would like a copy of the release, you can access it online at our website. During the course of this conference call, Arista Networks management will make forward-looking statements, including those relating to our financial outlook for the third quarter of 2020 fiscal year, longer-term financial outlooks, the potential impact of COVID-19 on our business, industry innovation, our market opportunity, the benefits of recent acquisitions, and the impact of litigation, which are subject to the risks and uncertainties that we will discuss in detail in our documents filed with the SEC, specifically in our most recent Form 10Q and Form 10K, and which could cause actual results to differ materially from those anticipated by these statements. These forward-looking statements apply as of today, and you should not rely on them as representing our views in the future. We undertake no obligation to update these statements after this call. Also, please note that certain financial measures we use on this call are expressed on a non-GAAP basis and have been adjusted to exclude certain charges. We have provided reconciliations of these non-GAAP financial measures to GAAP financial measures in our earnings express release. With that, I will turn the call over to Jay Shreve.

Thank you, Curtis. Thank you, everyone, for joining us this afternoon for our second quarter 2020 earnings call. To start with, I would like to address the once-in-100-year coronavirus global pandemic and reiterate our commitment to employee safety and customer response. At Arista, we recognize our role and responsibility in supporting global communications and cloud infrastructure during these mission-critical times. We are adjusting to the new work-from-home norm and expect this will continue throughout 2020 until vaccines or therapeutics emerge. We are working closely with our supply chain and contract manufacturers to improve lead times and support our customers in their business continuity initiatives. Back to Q2 specifics. We delivered revenue of $540.6 million for the quarter with a non-GAAP earnings per share of $2.11. ACARE services and software support renewals contributed approximately 22% of revenue. Our non-GAAP gross margins were 64.7% influenced by software and services mix. We registered a record number of million dollar customers as a direct result of our enterprise vertical traction and continue to march towards our goal of one to two new customers a day in the second quarter. In Q2 2020, Cloud Titans was the largest vertical. The enterprise is once again our second largest performer, followed by Tier 2 cloud service providers and financials tied for third place, and service providers in fourth place. Our vertical mix in the second quarter was consistent with the prior trends that we shared with you in May. Cloud titans, approximately 40% of the mix, enterprises, including financial services, approximately 35% of the mix, and providers, approximately 25% of the mix. In terms of geography, international competition was 19%, with the Americas at 81%. Regional EMEA traction was slower than normal due to push-out of big customer decisions and reduced international cloud-type spend in the quarter. As we predicted, Arisker's cognitive campus portfolio met its first-year 100 million objective ending June 30, 2020. While we are pleased with the traction, we believe that our new enterprise prospects will take time in this COVID era. We remain hopeful and optimistic to double in the next five to six quarters to $200 million, of course, depending on market conditions. In terms of new innovations, last week, Arista introduced the cloud EOS router for the edge, supporting dynamic path selection services across all major cloud providers. Some of the key features include AWS Transit Gateway integration for seamless automated provisioning, elastic consumption of Cloud EOS across the three main cloud providers, AWS, Azure, and GCP, declarative provisioning of multi-cloud through HashiCorp's Terraform, and a dashboard based on Arista's Cloud Vision for centralized visibility of enterprise and cloud resources. As we review 2020 at the mid-year point, there are clearly macro issues outside our control. The number of confirmed COVID-19 cases has increased sharply in July. The economic recession is looming, and trade wars are escalating. Despite this, Arista experienced a cloud-tightened recovery and enterprise strength in Q2, offset by extended sales cycles in the campus sector. We're also pleased with the recent leaders' recognition in both Forrester's Wave for open programmable switches in business-wide SDN, with a top score in strategy category, and with Gartner's Magic Quadrant for data center and cloud networking. Gartner designated us a Magic Quadrant leader for the sixth consecutive year. These placements truly validate Arista's premier status in networking. I believe we are well positioned in cloud networking for data center, routing, campus, and new multi-cloud monitoring capabilities, enabling network migration from legacy to modern. In these unprecedented times, customers seek our compelling advantages, and we expect to emerge even stronger than many of our industry peers. We hope you're all being safe and well. I will now turn the call to our CFO, Ida Brennan, for financial details. Ida?

Thanks, Sushri, and good afternoon. This analysis of our Q2 results and our guidance for Q3 2020 is based on non-GAAP and excludes all non-cash stock-based compensation impacts, certain acquisition-related charges, and other non-recurring items. A full reconciliation of our selected GAAP to non-GAAP results is provided in our earnings release. Total revenues in Q2 were $541 million, down 11% year-over-year, but at the upper end of our guidance of $520 to $540 million. Approximately 6% of this decline relates to the recognition of $38 million of product-deferred revenue in the second quarter of 2019. In addition, while overall demand in Q2 was reasonably healthy, we continued to experience some COVID-19-related supply challenges, resulting in extended lead times and somewhat constrained shipments for the quarter. Service revenues represented approximately 22% of total revenue, up slightly from 21% last quarter. International revenues for the quarter came in at $104.7 million, or 19.4% of total revenue, down from 23% in the first quarter. While the shift in geographical mix on a quarter-over-quarter and year-over-year basis was largely due to the location of deployments by our Cloud Titan customers, we did see some push-out of larger opportunities in our in-region businesses also. Overall gross margin in Q2 was 64.7%, above the midpoint of our guidance of approximately 63 to 65 percent, with the last quarter at 65.6 percent. Operating expenses for the quarter were 144.1 million, or 26.7 percent of revenue, down from last quarter at 149.3 million. R&D spending came in at 91.6 million, or 17 percent of revenue, consistent with last quarter. Sales and marketing expense was $41.9 million, or 7.8% of revenues, down from $46 million last quarter, with lower marketing and travel-related spending. Our G&A costs came in at $10.6 million, or 2% of revenue, down from last quarter at approximately $12 million. Our operating income for the quarter was $205.7 million, or 38.1% of revenue. Other income and expense for the quarter was a favorable $8.3 million, and our effective tax rate was approximately 21.9%. This resulted in net income for the quarter of $167 million, or 30.9% of revenue. Our diluted share number was 79.3 million shares, resulting in a diluted earnings per share number for the quarter of $2.11, down 13.5% from the prior year. Now turning to the balance sheet. Cash cash equivalents and investments ended the quarter at approximately $2.8 billion. We did not repurchase shares of our common stock during the second quarter. As a reminder, we have previously repurchased $494 million, or 2.4 million shares, against our board authorization to repurchase $1 billion worth of shares over three years commencing in Q2-19. We expect to continue to execute opportunistically against the remaining authorizations. We generated $138 million of cash from operations in the second quarter, affecting solid net income performance somewhat offset by incremental working capital investments. We expect to continue to strategically increase inventory levels through the end of the year as we look to improve lead times and help buffer against any future COVID-related supply chain disruptions. Deals come in at 65 days, often 61 days in Q1, reflecting the linearity of billings in the period. Inventory turns were 2.3 times, down from 2.5 last quarter. Inventory increased to $327 million in the quarter, up from $262 million in the prior period. Our total deferred revenue balance was $578 million, down from $597 million in Q1. As a reminder, our deferred revenue balance is now almost exclusively services-related. The level of services deferred revenue is directly linked to the timing and term of service renewals which can vary on a quarter-by-quarter basis. Account table days were 59 days, up from 43 days in Q1, reflecting the timing of inventory receipts and payments. Capital expenditures for the quarter were $2.1 million. Voucherings are outlook for the third quarter and beyond. We continue to closely monitor the impact of COVID-19 around the world. Local operating restrictions remain in flux, and it is unclear when and how these restrictions will finally be resolved. While we make good progress on supply chain and manufacturing during the quarter, we still have some work to do to fully return to normal lead time. On the gross margin front, we would reiterate our overall gross margin outlook of 53% to 65%, with customer mix being the key driver. We expect to recognize some incremental COVID-related freight and supply chain costs in Q3 and Q4, with all other things being equal, would bound gross margins closer to the midpoint of our range. On the spending side, we'll continue to manage spending and investments carefully, prioritizing key projects and customer engagements, while benefiting from a natural COVID-related reduction in travel, marketing, and related expenses. You should, however, expect to see us add investments back into the model as incremental revenue solidifies. Finally, our guidance for Q3 affects our current understanding of COVID-19 and its impact on our business and supply chain. This is, however, an inherently uncertain situation, and we will need to continue to monitor and attempt to mitigate new challenges as the situation unfolds. With all of this as a backdrop, our guidance for the third quarter, which is based on non-GAAP results and excludes any non-cash stock-based compensation impacts and other non-recurring items, is as follows. Revenues of approximately $570 million to $590 million, gross margin of 63% to 65%, operating margin of approximately 37%. Our effective tax rate is expected to be approximately 21.9%, with diluted shares of approximately 79.6 million shares. I will now turn the call back to Curtis. Curtis?

Thank you, Ida. We are now going to move to the Q&A portion of the Arista earnings call. Due to time constraints, I'd like to request that everyone please limit themselves to a single question. Thank you for your understanding. Operator, please take it away.

We will now begin the Q&A portion of the Arista earnings call. In order to ask a question during this time, simply press star, then the number one on your telephone keypad. If you would like to withdraw your question, press the pound key. We ask that you pick up your handset before asking questions to ensure optimal sound quality. Your first question comes from the line of Rod Hall from Goldman Sachs. Your line is open.

Yeah, hi. Thanks for the question. I wanted to start off with the revenue trends and just ask on the verticals whether you're seeing, since we didn't see the seasonality on the quarters last year, whether what you've seen here is normal or similar to what you observed last year in terms of the cloud and other verticals, movement from Q1 to Q2. Sure. And then secondly, I wondered if you guys could comment on the supply impacts. I know that you said there was, or we calculated about $14 million of impact last quarter. Has that all come back to you here, and is there any further ongoing impact on revenues in the guidance from supply chain issues, or are those mostly solved now?

Hi, Rod. This is Jayshree. So in terms of seasonality, I think things got better in Q2 than we feared from COVID in general. We obviously experienced a very strong comeback from the Cloud Titans, but we also had strong traction on all the other four verticals as well, and they pretty much continued in a fairly linear fashion throughout the quarter. And in terms of seasonality, manufacturing, we are still experiencing inventory issues. We have very long lead times. We are improving them. We look to improve them, but we don't expect it to be back to normal until probably Q4.

Okay. We're still getting a little bit of drag on revenues, Jay Shree, because of the supply chain stuff, but maybe it sounds like maybe not as much as last quarter.

Yeah, we're improving from Q1 to Q2. We'll improve again from Q2 to Q3, but we won't get back to normal until Q4.

Great. Okay, thank you.

Thanks, Rod.

Your next question comes from the line of Eric Superger from JPM Securities. Your line is open.

Yeah, thanks for taking the question. Back on the supply chain question, Did it prevent you from shipping from further upside in the quarter or were you able to, because I think you had indicated last quarter that you had inventory, sufficient inventory for the June quarter. Did your supply chain prevent you from being able to make the shipments that you anticipated to make?

Well, given the guide we gave Eric, we were able to make a quarter. But if you're asking could we have made more, we could have made more if we had more inventory.

Okay. Very good. Thank you. Thanks, Eric. Thanks, Eric.

Your next question comes from the line of Jason Ader from William Blair. Your line is open.

Thank you. Good afternoon, guys. Can you talk about the demand environment for the Enterprise Data Center? I'm not talking specifically about campus, but about Data Center. It seems like you had another good quarter there. Just a little bit of why you're winning there. What are some of the learnings, Jayshree, that you've had over the last several years selling into that market? Because I know it's a pretty competitive market with a major incumbent. So just talk about your success there. the environment, and why you're winning.

Yeah, no, I think we're winning because of really three reasons, quality and support. Customers are seeing us as being superior to our peers in every which way in that department. Our innovation is getting stronger and stronger. And then our market recognition goes from industry analysts and our customers, and, you know, customers tell other customers. So many of these enterprises were aware of us and are now starting to make real decisions on us, right? both existing ones with greater land and expand opportunities, as well as new ones. Our sales cycles in the enterprises can be long, and so this is really work in progress that's been going on for several quarters that we're seeing materialize in Q2. This may sound obvious, but also another aspect of our enterprise's success is that the familiar customers are deepening their expansion with us for many more use cases. So not only are we winning new prospects, but we're also getting more land and expand opportunities. So I think we've always pointed to the enterprise being our shiny star and our brightest opportunity, and now we're seeing that across the board in all the regions.

Great.

Thank you, Jason. Your next question comes from the line of Nita Marshall from Morgan Stanley. Your line is open.

Great, thanks. I just wanted to kind of get an update on how you thought COVID, you know, was changing how Tier 2s were thinking about build versus leveraging public cloud, just given the acceleration some of them might have seen due to kind of COVID or work from home.

Thanks. Yeah, no, I think, you know, there's two ways to really look at Tier 2 cloud meters. Sometimes They make investments and from a cost perspective they can very much justify their investments as much more cost effective in the long run than going to the public cloud. But other times they have a hybrid combination of also going to the public cloud to manage their cost structures. So you see a bit of both. It's very cyclical in nature and it really depends on which tier two cloud you're talking about. So some start on their own cloud and continue to grow the workloads, and we experienced some of that in Q2. And others slow down a little, and we experienced some of that too throughout the year. We did have some delayed decisions that we hope we can pick up in the second half of the year. I think a piece of the Q2 cloud that's doing well for us is especially CDN. With the work from home, as you rightly mentioned, you know, the aggregate of all that video bandwidth is starting to create investment. It's not that the video itself is large, but all of the 4K and 8K flows add up to an aggregate bandwidth. Great. Thank you. Thank you, Mito.

Your next question comes from the line of Amit Daryanani from Evercore. Your line is open, Amit.

Thanks for taking my question, guys. I have a question for you. In the last couple of quarters, we've seen your revenue decline, you know, 10% to 12% year-over-year, while hyperscale capex has continued to go higher. And I totally realize hyperscale capex is a lot more thing than just ERISA spend. But as we get into the back half of this year, you know, the expectation is hyperscale capex will slow down somewhat. And in that context, Should we feel comfortable that Orisa's revenue decline starts to diminish and shrink as we exit the year? Or how do we think about that?

Well, Amit, first of all, you're absolutely right to say there's no one-to-one correlation between network capex and the massive cloud-tightened capex. But if you look at the cloud-tightened capex that was reported, some were very high, some were strong, some were sluggish, some were down. And so I think the Titans did experience a stronger spend overall, and that did affect our Q2. But as we said in the last call, and I want to reiterate, we originally thought we'd be flat to down. Because of the strong CapEx, we now feel good about flattish, which is an improvement from down. So, of course, we'll monitor this in the second half, but we believe that the overall for the year, you know, will be a flat – for CloudTitan. Got it.

Thank you. Thanks. Our next question comes from the line of Paul Silverstein from Cali. Your line is open.

Thanks. Good evening, guys. J3, relative to the longstanding concern among investors about you all getting picked off to one extent or another, whether from internally developed slash white box solutions or from third-party competitors like Cisco and Arista, specifically, obviously, referring to CloudTitan's Any insight you can share with us regarding your competitive position, any changes for better or worse?

Thanks, Paul. No change for better or worse in the quarter. As you know, Arista introduced the switch abstraction interface, so we continue to have a long history of commitment on Linux, EOS SDK, containerized EOS, virtual EOS, DISAG EOS, Sonic support, FBOT support, ONL support. So we embrace white boxes, and our customers embrace Arista's blue boxes, too.

Just to be clear, I'm not just referring to decisions that impacted your quarter revenue, but decisions that might have been made that could impact the latter half of this year or next year for that matter.

Yeah. Maybe Anshul, if you're there, we can get you to get some color, but we don't see any change in second half.

Go ahead, Anshul. I would say there's a lot of noise around this topic and obviously created by parties that don't have much to lose. But, look, we are engaged, as you mentioned, in multiple projects, including a lot of the roadmap for our customers. It's not just a box they have to choose. They have to build a network that scales for their business, and we are very heavily involved in their 2021 and 2022 architectures. So the high-level message I would give is don't worry. We will be fine.

So I apologize, but is that don't worry, notwithstanding share loss, you'll be fine because there's so many projects you'll grow, notwithstanding others gaining some ground? Or are you saying that there will be incremental projects and you will not lose share?

I think so. Paul, I think we can continue in detail, but I think the message you're getting is our fundamental thesis is unchanged. We're not seeing architectural shifts or competitive losses. In fact, recent data validates that we're number one spot for the third consecutive year on 100 gig, and our share is increasing every quarter.

Thanks, Paul. Okay.

Your next question comes from the line of Jim Suva from Citigroup. Your line is open.

Thank you so much. Maybe if we just look even, you know, longer term with 400 gig coming up, yet we have a coronavirus pandemic that we're working through too, and it's harder to meet and have teas and coffees with people in person and such. Any visibility you have on, you know, 400 gig coming, whether it be discussions, are they more fruitful, taking longer, testing labs? How should I think about that?

Right. Thanks, Jim. Well, I think, as we've always told you, 400-gig trials began at the end of last year. But as we said last quarter, we will be delayed from 2020 to some of them may go into 2021 due to all the things you mentioned, COVID-related slowdown. You just can't deploy 400-gig over a virtual period. collaboration conference. You have to do new product qualification. You have to get the optics in. That said, we're very pleased with our progress in 400 gigs, and we're winning tens of customers. In fact, to date, we've won over 50 customers in 400 gigs. So we're pleased with the progress, but many of them are in trials and early deployment cycles.

Thank you so much.

Thank you, Jen.

Your next question comes from the line of Aaron Rakers from Wells Fargo. Your line is open.

Yeah, thanks for taking the question. I just want to go into the numbers a little bit. You know, Edith, can you help us understand or appreciate the variables to gross margin? At 59.5% of product gross margin, it looks like it's the lowest level we've seen, and I think If my memory is correct, your mix between the verticals really didn't change sequentially. So can you just unpack the impact that you're seeing on gross margin and how we think about product gross margin going forward? Thank you.

Yeah, I mean, I think there's, you know, the biggest driver is still the customer mix, right? But you're right, it wasn't that dislocated between Q1 and Q2. I think if you look at last quarter, we had talked a lot about, you know, being able to sell some inventory that previously we thought we couldn't sell or we wouldn't sell, and that has given us some pickups on the gross margin side. All of that type of stuff will flow through product margins, obviously not through the services. So now as you think about this quarter and honestly into Q3 and Q4, we have some incremental COVID costs that we're also covering around freight, logistics, you know, just supply chain costs of kind of prioritizing supply over demand. you know, over a cost structure, right? So we are going to carry some, you know, some burden of cost around those activities through the end of the year. That will all show up in that product line, right? So I think that we'll move around a bit even outside of the, you know, just the normal kind of customer mix that we've been seeing historically.

I guess just to be clear, there's been no change in pricing dynamics competitively in the market that you're seeing?

I mean, there's always puts and takes on pricing. That's how the market works, but nothing out of the ordinary or anything new or different.

Yeah, and just to clarify what Ida said, we are experiencing COVID-related greater cost structures, no change in competitive or pricing dynamics.

Perfect. Thank you.

Thank you. Your next question comes from the line of Simon Leopold from Raymond James. Your line is open.

Thank you very much for taking a question. Jayshree, I appreciate you giving the guiding comment on the cloud titans. I was hoping you could clarify what baseline you're using when you talk about flat, given that you had the significant revenue recognition last year. I guess the number we had been thinking you meant is, was a base of about $840 million in 2019. But I want to make sure I understand what you mean by flat, what you're comparing it to. Thank you.

Yeah, no, that's actually a great question, Simon. I was just simply computing year over year relative to, you know, the revenue we've given, but I didn't get into the color of deferred either.

You want to help me with that? Yeah, I mean, that commentary, Simon, is very much around demand, right? So it's not adjusting as the deferred. So, you know, if you think about the clouds, business year over year and you say, okay, the demand was flat, you still need to adjust out the $118 million of deferred on top of that, right? So you will still see a decline on a revenue basis for cloud year over year, even with the demand, as Jayshree described, being flat.

Could you tell us what dollar value you're using just to keep our lives simple?

At the top of my head, no, but if you take the percentages that we gave you roughly for the mix of the business and then adjust the deferred, you'll get there, right? I mean, you know how much the deferred was, $118 million, and you know what the split of the business was on a demand basis, so you can get there. We can take it up afterwards, too, if you want, but it's pretty straightforward. Okay.

Thank you very much.

Thanks, Tom. Our next question comes from the line of Lu Jin Ho from Bloomberg. Your line is open.

Great. Thank you for taking my question. So it seems you may have a new competitor in the switching market, Nokia. And it's not a competition question, more so of a market demand question. I'm just curious if the telco providers have changed their purchasing or their thoughts on switching, because I know that's been a sore spot for you guys for quite some time now. I'm wondering if 5G is finally coming into fruition for switching business for you.

Well, you know, I think there's very few details on exactly what the product is, but Nokia is certainly a good service provider company, and I believe while their entry into the data center as a fifth player will be difficult and will be tough to compete, and I don't really see any 400 gig products available despite the marketing, we do think, and we have respect for them as a service provider and especially combined with optical company. They might find some use cases there. So I don't think we'll see, like you rightly pointed out, I don't think we'll see a direct competition with them, but they may find some use cases combining with optical.

Well, is there an opportunity for that Telco data center to finally emerge when 5G starts rolling out for you guys?

We are already seeing opportunity with the Telco data center independent of 5G rolling out, but we certainly see that as independent of Nokia. That's a very real use case that Arista today participates strongly in the service provider segment.

Great. Thank you. Thanks, Vijay.

Your next question comes from the line of Jeff Qual from Wolf Research. Your line is open.

Thank you very much, everyone. I was hoping to ask about the trajectory in Cloud Titans. Traditionally, of course, you all have been very smooth with growth in the Cloud Titan vertical. In the last couple of years, it's been a little less so. I think many of us have been hoping that 2021 would be back to the smooth trajectory. And I'm wondering if the push out in 400 gigs defers that a little bit, or we should be just a little careful about how we look at 2021, given that 400 gig might come, you know, late in the first half or even the middle of the year rather than at the top of the year. Thank you.

Thank you, Jeff. I'm going to say a few words and pass it to Anshul. When you look at 400 gig, as I said, that's not one you can buy over a cup of coffee. So there's an intense level of testing and proof-of-concept labs that go on. And while our cloud visibility and demand is much better understood for 2020, I think it will still take time to make this migration from 100 to 400 gig. And different cloud titans are in different stages of that migration. They're certainly all in varying forms of trials. But I think you're right to say some of them will get off to a good start in 2020, late 2021, and some of them will take longer. I think what I can confidently tell you is the combination of 100 gig and 400 gig is going to be the mainstream deployment for most of our customers. Anshul, you want to add to that on the 400 gig?

Absolutely. As you mentioned, there's no real stall effect for 400 in the cloud because the cloud needs to deploy capacity. They'll buy whatever is available in the market and qualify it today. The market is going to bifurcate, going from 100 gig into 200 gig, 400 gig, and 2 by 400 gig in the next one to two years. And I think we will participate very well in all of these factors. But, again, the cloud is not waiting. They just build with whatever they have today.

Okay. Thank you very much.

Thank you.

Your next question comes from the line of Samik Chatterjee from J.P. Morgan. Your line is open.

Hi, thanks for taking my question. Actually, you mentioned you're seeing a strong comeback from the cloud earlier, and I'm just wondering how you're feeling relative to visibility in terms of getting back to more of a mid-teens growth rate and spending from cloud customers that you had communicated at the 2019 analyst day, and how does this set us up for 2021, where I think consensus estimates seem to be largely embedding in a double-digit growth rate with the cloud?

Thanks, Salik. I don't think I'm ready to say we're going to grow double digits or mid-teens on the cloud. Our numbers are so large, we'll be happy to grow in the cloud period, right? And I think our faster growth and traction is clearly coming from the enterprise. But the large numbers are clearly being contributed from the cloud. So time will tell, but there's too many variables to make a direct correlation for that in 2021.

Yeah, it's a little early to start trying to call a 2021 view just yet with all the uncertainty that there is, right? Yep.

Thank you. Thanks, Monica.

Thanks, Timmy.

Your next question comes from the line of Taliani from Bank of America. Your line is open.

Hi, guys. Hopefully you can hear me okay. Yes, very clear.

Hi, Cal.

Hello. I'm going back to a lot of the questions that kind of danced around it. Last year, if I look at Q1, Q2, and Q3, you grew year-over-year 26%, 17%, and 16%. Pretty nice growth. And this quarter, yes, you beat the numbers, but year-over-year growth is – double-digit decline. So I'm trying to understand the color of the numbers beneath the surface, meaning last quarter you provided us with the numbers for the cloud vertical, and I was able to see what was the growth ex-cloud vertical. I'm trying to do the same thing to see where is the strength, where is the weakness, and get some color around it. So can you help us with two things? First is... what's the growth with and without cloud, just so we understand the growth in the other verticals, and second, can you give us color on the profile of the cloud growth, meaning is this an existing customer that started to deploy new architectures you mentioned in the past, and because of that the demand is going up, Or is it just demand just after a decline, you see kind of snapback in demand, et cetera? I'm trying to see if there is any big trend behind the recovery in cloud. Thanks.

Yeah, I mean, Tal, I don't think we're going to go kind of down to that level of detail. I think if you take kind of the year-over-year growth, you have about 6% of that that relates to the deferred revenue being recognized last year. You have, you know, a step down on Facebook, which I think we've talked about before, right? They were very active in the first half through Q3, actually, of last year, and then we saw that step down in the fourth quarter, right? And we know they're running at a lower level. We've shared that previously through this year. So those are kind of your big, you know, drivers on the cloud without us trying to get into different use cases and products and stuff because we're not going to do that. But those are the key drivers. kind of drivers around the cloud. And then we've seen strength, you know, in some of the other verticals. And Jayshree gave the mix of the verticals again this quarter. There was nothing that unusual in the mix versus, you know, the breakdown that we gave you on the last call in the quarter just gone in Q2. All right, so enterprise continued to do well, continued to grow, and, you know, cloud was better than we had expected. But still, when you look at it over the year, it's not a – you know, it's not a breakout. It's solidly kind of flattish on a demand basis now versus being, you know, down when we thought coming into the year.

Any color, maybe my second question, any color about the growth in the cloud vertical?

I mean, I think you can get there on the map, right? If you take out the deferred and you look at the growth on the, certainly on the revenue basis, you can get there, right? You know, we're saying for the year, we think the demand will be flattish for the year, right? So I'm not sure if there's much else that we can give you at this point.

What I was referring to was in the past you noted that some of the declines were because customers were looking to deploy new types of servers and as a result they reduced spending. I'm trying to understand whether there is any correlation between that part and between the growth in cloud. Is it just reversion to the mean or is there any big trend beneath it? and new architectures are deployed, et cetera, and that's why demand is going up.

So look, we have a number of use case expansion tiles in 2020. Usually they come in multiple flavors. Either it's adding an additional spine layer where they're building a super spine or regional spine. It could be expanding racks to increase the server density on the network IO. And these two especially, or they could be building out new data centers. So these are the three most popular use cases we see. And we definitely saw that this quarter and we expect to see more of that this year.

Great. Thank you. Thanks so much.

Your next question comes from the line of Alex Kurtz from KeyBank. Your line is open.

Hi, this is Michael Osborne. Alex, thanks for taking the question. As we enter the second half of the year and we start to get the op-ex levels for fiscal year 21, I guess how would you start to think about the timing of fiscal 21 leverage and the return to more normalized growth? Thanks.

Yeah, I mean, I don't know that you should expect us to drive, you know, a ton of leverage. I think we'll, you know, we talked about kind of reinvesting and bringing back some of the investments, you know, that we talked about taking out last quarter as we see revenue, you know, solidify and revenue growth solidify. So I think, you know, the operating margin targets that we've out there, you know, the long-term model is roughly the 35% plus or minus. We guide with 37% for Q3. So I don't know that we're looking to, you know, drive operating margins that are much higher than that. So we'll continue to invest, you know, for Fortunate to see kind of top line growth.

Great. Thank you, Michael.

Your next question comes from the line of James Fish from Piper Sandler. Your line is open.

Hey, ladies. Congrats on the quarter. But I just want to understand where you mainly saw the traction on the campus side. It's specifically on the geos, given what's going on lately in this environment, because most of us really aren't sitting at a campus anymore. And if I can just sneak in one related to this, were there any push-outs related to Cloud Titans? I just want to understand correctly, if the push-outs were primarily just related to Cloud Titans internationally, is that the right way to think about it, or was it something else?

Okay, let me understand, I get confused when there's more than one question, so let's start with the campus first. So your question was, you know, how did we achieve our goals when nobody's in the office? Well, quite simply, when nobody's in the office, if everybody's making planning decisions, it's easier to install them in the office because no one's there to interrupt with it. So what we found is that the engagements we've had with both existing customers and new prospects in 2019 really helped us achieve our first-year goal through the first three years. And there was very strong demand for I can't even point to a theme on verticals. It ranged all the way from mid-market enterprise to some of our large customers who are familiar with us, who really influenced our decision to be in the campus. One of the main reasons they want us in the campus is single EOS, high quality, one cloud vision, and then one ability to manage both your wired and Wi-Fi. What was your question, James, on the cloud titans?

I just wanted to make sure I was thinking about it correctly with the push-out related. It sounded like you guys said it was related to Cloud Titans Internationally. Is that correct?

No, no, no. What we tried to say was our EMEA business was weak because the Cloud Titans did not order in that country, so the European piece of our Cloud Titan was weak. And there were some push-outs of European customers that pushed out of Q2, which made that number weaker than normal.

Thanks for the call, Jayshree.

Sure.

Thanks, Jane. Thanks, Jane.

Your next question comes from the line of Ryan Coots from Rosenblatt Securities. Your line is open.

Great. Thanks for the question. Circling back to your campus opportunity there, can you maybe walk us through some of your channel strategy and how much is direct versus you're looking for third-party integrators and such to work with you?

Yeah, no, that's a good question, Ryan. Our channel strategy is still in very early stages and evolving. In fact, if you do channel checks, I think most of you don't hear enough about Arista. So I would say a lot of our first year success was definitely due to our direct customer focus, maybe as north as 80%. However, we fulfilled through the channels, and so our fulfillment through the channels was also 80%. But the impact of channels, I think, will kick in more in the second and third year than it did the first year. And I've got to give a shout-out to Chris Schmidt, Ned Chapman, and a number of folks who are working on a suite of channel programs that are extremely competitive all over the world.

Thank you, Brian.

Your next question comes from the line of Ben Bowen from Cleveland Research. Your line is open.

Good evening, everyone. Thank you for taking that question. Jayshree, I was hoping you could talk a little bit about how you feel the increased mix of work-from-home employees across broader enterprise is likely to impact longer-term investments in the campus from the enterprise customer. And then also, you mentioned a little bit about video as service provider, but I'm interested if you think there's any corresponding impacts as more people are at home, and that might be influencing service provider investments.

Thanks. Yeah, no, Ben, both are very good questions. I think it really depends on how long you think work from home is going to go on. If it's going to go on indefinitely, I think the campus investments will take longer. It will still happen. We were expecting to double every year, and maybe we won't double every year, but we will attempt to double every 18 months to two years, so... There's very much a campus requirement, whether you call it the headquarters campus or the regional offices or the branches, and Arista's technology where we're able to unify wired and wireless in a single pane of glass is powerful. But I think we will see the decision-making take longer, so I still see the opportunity as vibrant, but the time being longer. In terms of video and service providers, I think that's a really good question. When you look at our tier two, both cloud providers and service providers, the ramp in video traffic by itself, because it's compressed workflows, aren't that great. But the aggregate bandwidth is definitely causing more investment in both tier two cloud providers and service providers. It's a colo or a peering point. or a content provider. So we are seeing that the work from home is creating an aggregate demand for bandwidth.

Thank you.

Okay. Thanks, Preston.

Your next question comes from the line of George Nutter from Jefferies. Your line is open.

Hi, thanks very much. I guess I was... You were going to ask about, I guess, some of the bigger picture drivers. You talked about 400 gig earlier, but Intel recently delayed their 7 nanometer program around CPUs, including server CPUs, and I guess I'm wondering if that impacts the server refresh model that many big Internet content providers were kind of following. How do you sort of think about that bigger picture?

Thanks. Thanks, George. I think... When it comes to building a network, the philosophy many of the top enterprise and cloud clients take, and Anshul alluded to this, is build with what we have. So we haven't seen any delays in putting the network together because of server depreciation cycles or product acquisition cycles. So we don't think the Intel slip will have an immediate impact. It could factor in in 2021 or 2022, But right now I think we don't see any impact of the 7 nanometer because they need I.O. They need I.O. independent of which CPU it is. So we feel pretty good in the near term. Thank you. Thanks.

Your next question comes from the line of Sammy Baudry from Credit Suisse. Your line is open.

Hi, Sammy.

Hey, sorry about that. Thank you for the question. I just wanted to ask a little bit about the international opportunity there. We've seen a couple of quarters of year-on-year decline, and you talked about some push-outs. However, the European region, some of the other regions are probably going through something very similar as the United States, more cloud application consumption, and people working from home, et cetera. I just want to understand, what is it about the other regions that are not seeing the same magnitude of demand or the same dynamics as the U.S. from an architecture perspective, from a product perspective, how Arista inserts into it? Could you just give us any color on what explains the, you know, the decline versus some of the other, you know, other regions that are holding up relatively well?

Yeah, so I would say three things, Sammy. I think, first, we just got a late start on international investments, and we're still looking to see that pay off country by country, and each country has its dynamics. It's hard to lump Europe as one. The second is both quarters that you're referring to was influenced by the Cloud Titan and how they purchased. So they just tended to purchase more in the United States than they did in-country in the region, and that affected it. And then the third thing I would say is in general, Krishmit and Ashwin Kohli and Anshul and the whole team, we're humming more and we're having greater enterprise traction in the U.S. and bigger bets in the U.S. And we want to see the same and we have every belief and hope that we will see the same, but it's taking a bit longer in these national locations.

Got it.

Thank you.

Thanks, Amit. Thanks, Amit. Your next question comes from the line of Vinod Srinivasara from Oppenheimer. Your line is open.

Hi. Thanks for taking my question. I just want to talk about your outlook. To the extent that you're embedding supply and demand headwinds, would you say that the headwinds are mostly, say, like 75%, 80% supply chain related, or are you seeing or expecting some demand weakness or slowdown in any vertical? Thanks.

Thanks, Vinod. That's a loaded question that I may not be able to answer to your satisfaction, but I would say at least as it pertains to Q2, demand was strong and I wish we could have shipped more. And I think we will run into a little bit of that in Q3 as well, albeit it's a slow summer cycle. So we don't yet know about next year or we don't know enough about Q4, but certainly Q1, Q2 were pressured by supply chain. And the second half, we hope to respond to that supply chain and create more demand.

Okay, thank you. And is there – what would you say – can you just give us a sense of your backlog? Like how much of that has kind of been held up?

Yeah, no, we don't talk about backlog, but it's safe to say we have some.

Thank you.

Thanks, Ben.

So this concludes the Arista Q2 2020 earnings call. We have posted a presentation which provides additional information on our fiscal results, which you can access on the investor section of our website. Thank you for joining us today, and everyone be safe.

Thank you. Thank you for joining, ladies and gentlemen. This concludes today's call. You may now disconnect.