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

Good morning Paul Silverstein, senior analyst communications networking equipment. It's my pleasure to have with us Arista Networks. We have CFO, Ita Brennan. I think I've known since the start. [ Thank you for joining us. ]

Happy to be here. Thanks for having us.

So my first question for you is, what is new different and meaningful that you haven't discussed ad nauseam [ and why ]?

Wow.

Before we get into the serious stuff.

I mean what is new? I don't know that there's a ton that's new, but it's not that long since we had our earnings call, so that's probably not surprising. We have some new products this morning if you want to go take a look at that. We do product announcements just from continued innovation around combining some of the low latency stuff that we've done before with some of the routing, layer 2, layer 3 capabilities all into a single product platform, so I think that's interesting to go take a look at. Again, just part of a continuing sleuth of new products and new releases over time. Sometimes we get fixated on the 400-gig line and the 300-gig line, 100-gig line. There's a ton of new products and new capabilities that happen kind of on the back of those kind of speed and platforms. So it's always interesting to see kind of just the speed with which the team is continuing to release not only new hardware products but also new software features and capabilities on an ongoing -- I think that's -- so keep an eye out for those. Other than that, I mean, from my seat, it's a lot of supply, supply, supply. It's just new, exciting, something that...

So that's actually a perfect segue. Before asking about the more substantive stuff, let's take a minute on supply chain, especially in the wake of Cisco comp [indiscernible]. Even with the China lockdowns coming off, there's meaningful risk that there's some [ particularly in ] port congestion [indiscernible] freight over in China. My sense is it's not as meaningful an issue for you, but let me let you respond. Any thoughts you can...

Yes, I think our direct [indiscernible] probably a little bit less, right? But we're always conscious of the fact that if you had a prolonged disruption in China, then that will start to show up and [indiscernible] across I would say pretty much all vendors, right, it's just a matter of time. So we think that we're starting to see some relief [ fast ] [indiscernible] be kind of how long it takes and how fast it comes back up. But our exposure is more secondary [indiscernible] supply chain can prolong the problem, I think, and it's manageable [indiscernible] work out of...

What would you say that was out here [ in ] that for long. It's now been, I think, 6 weeks from start. And if one factor's -- and I recognize that's not a given, but if one factor's in the port, likely [indiscernible] mitigate congestion, I would have thought that would be prolonged for you and most other companies [indiscernible].

I think that we have some ability to [indiscernible] part of kind of what we've been doing on the planning side. But again, if it were 6 months [indiscernible] type disruption, something like what we saw at the beginning of COVID, where you really had kind of very severe broad based disruption for [ some time ], that's when we start to see some [indiscernible].

But Ita, you're clearly indicating that at this point in time, the lockdown, you don't see incremental issue either from an ability to fulfill demand standpoint or from a cost [standpoint]?

I mean, I think, again, a couple of weeks ago, obviously, we contemplated all of the things that are happening in the world at the time. All things being equal, I think that's probably [indiscernible].

[indiscernible] third question, any thoughts on why it's so different for you? And what they spoke about [indiscernible] was $1.4 billion, $1 billion shortfall, their expectations [indiscernible] quarter based on China lockdown [indiscernible] sound like it...

Hard to try to kind of bridge to somebody else's [indiscernible] talked about what we were [indiscernible] far more immediate problem was [indiscernible] part, we talked about that kind of [indiscernible].

All right. Let me ask you one more question [indiscernible]. You, I think, more or sooner, earlier and more aggressively as anybody else in terms of both your purchase commitments, including going direct to vendors who you previously I don't think went direct to in order to better manage and you ramped up your inventory massively. And ironically, we've had conversation [indiscernible] about the magnitude, but I trust it goes without saying that, that reflects your visibility to demand. I think you and most of the companies are forward ordering to the tune of 4, maybe even 5 or 6 quarters from big customers gives you meaningful [indiscernible] it explains why you [ issue ] purchase [indiscernible] I'm going to let you respond [indiscernible].

Yes. Look, I think we've tried to take a risk adjusted [indiscernible] right, at what products have long eyes ahead of them, where are we seeing are working with customers there that was kind of a [indiscernible] of the product life cycle [indiscernible] life cycles, et cetera. And you get to the point where the [indiscernible] because your funding components go to [indiscernible] the risk go in excess. Now that you can have a little bit of that, but I think if you pick the right products [indiscernible] right discussions with customers, that helps to mitigate that. It's easy to have the what's the worst case scenario of taking this also to the other side of the line when you don't make these [indiscernible] where you can't satisfy customers and you're paying probably [indiscernible] expedite a lot of things in the short term just for that. Try to take a balanced view, continue to kind of risk [indiscernible] what we're doing, try to pick the right products with long lives [indiscernible] those products [indiscernible].

Trust [indiscernible], but there's obvious concern of investors from macroeconomic question [indiscernible] but [indiscernible] there's been no signs to date forward-looking metrics that would concern you about attenuation.

I mean we haven't done anything in the business that supports that, I mean that's a roughly up to folks. We hear the same concern. But in terms of looking at the business, haven't seen [indiscernible] in the business that which at this point [indiscernible] the discipline, you're hearing kind of around the economy and the macro and what your [ seeing ] bottoms up. Obviously, it's something we'll continue to touch [indiscernible] kind of cautious by nature. But...

Before I ask about your different customer market, I want to ask a big picture question. Ever since I've [indiscernible] the sector, stating the obvious I've seen young companies that grow up that [indiscernible] all comes to culture. And from the get go of founders [indiscernible] and your top R&D and sales staff, the best of the best, that it's not about any one product, people behind the products. But the ability to maintain that culture over time, it gets more challenging, more challenging. Larger people eventually go on to put more exciting. What's the risk that you're approaching or near that point with social benefits you have...

Look, I think one of the reasons why the company attracts [indiscernible] that exposes that innovative culture is very much alive. You have folks like [ Antral ]. You have like second level [indiscernible] embody that. We spend a lot of time on recruiting. And for somebody to come into the team as a leader, there's a lot of kind of betting that happens [indiscernible]. It's a highly technical product customer-focused organization. It's very entrepreneurial in terms of how we operate, fast move. So I think it's all about continuing to keep that focus, not that we're not adding new people, we are. Not that we're not kind of [indiscernible] especially in the organization, but making sure [indiscernible] culture and actually welcome that culture company [indiscernible] to a company that kind of focuses on those things, kind of targeting those people. I think that's really because you get through a very [indiscernible] cost environment.

From day 1, you've been largely about share gain, which is not to say that you're not impacted by macro, but you could gain share and grow dramatically in a nongrowth market. That said, strikes me -- and I made this observation previously in print, this is the best in market, carrier, cloud and enterprise that I've seen. But obviously [indiscernible] market growth. Much of your growth going forward to the extent you can...

I think enterprise, for obvious [indiscernible] winning new logos and campus business which is starting from doubling that, right? So that's a piece of the market where we're making good progress, but we're still very early in terms of do we have some verticals like [indiscernible] now we're slightly more penetrated, but we're still winning new [indiscernible] others where [indiscernible] have our sales, commercial, industrial, right? So I think that piece of the business for sure, there's a real focus on new customers' momentum that we think that's around that. Solution continues to add to the facilities so that we continue to address a broader set of [ price cuts ]. The nice thing about the enterprise market has been when you get that anchor customer any vertical, you have some things for them and then you can set up and go process the other customers in that vertical. So some really good traction around new logos and new wins in that price piece. So it's definitely, I think, a big driver of the campus piece alone. It's a big driver [indiscernible] I think about the other pieces of the business, some of it is just handsome [indiscernible] and continue to grow with us.

I would add that you have been very early in terms of products. You're still -- your product portfolio, if I used to want to reference [indiscernible] have a fresh -- all fresh on the products. And obviously, you're going to features of functional product platforms. Just by way of example, correct me if I'm wrong, I think it was only the last 12 months that you introduced a core campus platform that not only does it enable you to address the headquarter's building, city, whatever, but having a core solution set that would drive incremental revenue.

I think having this ability to kind of go to an enterprise customer and talk from -- all the way from kind of campus routing has been a huge addition, right? And it's definitely part of some of them that we're seeing there, right, that you're now not kind of coming in and speaking to a piece of their overall networking if you're able to address the whole thing. And that cross-sell capability, we kind of understood that it's kind of our net customers would want to address kind of campus with Arista, but we've also seen new campus wins where we can sell back into [indiscernible] some cases. So I think that's definitely been powerful being able to kind of add that kind of in with a more featured CloudVision and a continuous more features CloudVision that's across the whole thing. It's definitely an important part of the structure. It doesn't mean that we won't have new products back to the opening comments, right? We will continue to have new products within that, but [indiscernible] and the software structure will remain the same across those.

Just as a reminder to the audience, going from 0 to $100 million to $200 million [indiscernible] you're doubling that again this year, $400 million enterprise campus. Your enterprise isn't larger than enterprise campus. Those plans would have been [indiscernible]. That said, you've only been able to point on to the extent the [indiscernible] rises that you and any other organization tell your sales force, panel-present. It's still relatively modest. But you made the point that you're doing exactly what one would expect going from cloud to high end of the products [indiscernible] and you don't necessarily need that scale with quality of [indiscernible] point. That said, how much of your growth this year [indiscernible]...

I mean, I think about it is that a lot of the growth that we're seeing now is still being driven by kind of a direct sales new logos across customers and also, obviously, that cross-sell and penetrating further existing customers. We're starting to see some channel, partner sales efforts, but that still needs work to continue to kind of invest there, target partners that we can be successful with. I mean I think it has to be broad kind of incrementalism, right, but it really is kind of an incremental [indiscernible] partnerships hopefully be successful with them and then you can add more, and that will take time to continue to expand. We don't need that today to kind of -- to do what we have in front of us. But eventually, to your point, you'll need that [indiscernible]. So we are focused on [indiscernible]. But again, it's not something you can turn on overnight. It's something that you have to build with the right partners. So we're...

In the [indiscernible] channel, you both have the benefit of larger deal size customers, the scope of the network they're buying from you, the number of elements, et cetera, plus ongoing expansion, having well over 1,000 customers. [indiscernible]. How is the growth?

It's always a tough question to answer because usually new customers, they're initial. The contribution is small, right? And then as you're doing that initial kind of, I like to say it's hard to sell quality. But when customers or that's what drives us, we'll probably look at kind of that initial kind of incursion point so that we can kind of expand within that. So it's always hard to say how much [indiscernible]. How much of my revenue is new customers, it's always really very small. But it's because you've got that kind of land-and-expand type metric.

To be clear, you're not -- you all may have been very vocal in saying you're not for customer to go to this. But you're still selling a traditional model. This isn't about the initial with buffer revenue. That's not what you're talking about. Doesn't sound like you're changing that anytime soon.

I mean I think our view is for the operating system that's core to the product functionality, our customers at least still want to buy that as an initial sale of perpetual license, for feature sets that we're adding on top of that, be it CloudVision be it some of the visibility products and then other things that we'll do in the future that naturally fit as a SaaS product, we'll obviously sell those as a product. So it's really more not kind of forcing a fulfillment model on something that's the functionality of the product but other add-on software capabilities that we felt are good.

All right. I appreciate the specifics talk about any particular customer. I'm sure you've gotten the question directly, especially about Facebook, not your operating challenges with [indiscernible]. Personally, I think it's a better argument actually it's [indiscernible] model. But any thoughts you can share on what your [indiscernible], any thoughts about the [indiscernible].

Yes. I mean it is hard to talk about particular customers, so I mean we'll pull back to kind of cloud hyperscale overall. I mean, I think what we've seen across the board is really coming out of a relatively low investment period in networking, and therefore, being -- and having good growth in their business as that means the interim period and then needing to take advantage of some of these technology changes and what they bring. There is some urgency, but they're looking to -- across the board. Working on kind of planning, supply capabilities, how do they actually make sure that they can have, what they need to deploy. And it seems to be kind of in line operational almost for them to continue to do that. I think these are the types of customers that tend to make plans and have the confidence to kind of continue to execute plans that we substitute from them. And they are very focused on kind of, I think, what appear to be kind of operational, certainly.

If I remember the numbers, this year, we're looking at $23 billion or $24 billion big point. Obviously, 2 of those are the customers you're not -- that's coming off the year of , 2021 [indiscernible] huge years. Young folks worried about a [indiscernible]. But any thoughts?

I think if you look at networking and the networking spend, 2021 when you only really saw kind of the return and start to drive some meaningful kind of revenues, et cetera, at the back end of last year. So they really are coming into this 400-gig cycle off again, up a period where back end of '19 [indiscernible] particularly on networking. Those seem like we're kind of we're in a different part of that networking cycle. Just when they were ready to kind of make these investments, you got hit by supply constraints and it's been this kind of stretching out of what they want to do.

I'm sorry, you just said don't worry, be happy?

I did not say that. If you ever hear me say that, Paul, you'll be very surprised.

But you think it's not a given that there will be [indiscernible]. I have my own thoughts, but no one can [indiscernible].

Yes. I mean, look, I think, again, it comes back to how fast can they deploy, what are they for, right? You've got lots of new cases, et cetera. I mean we've talked some time now about thinking their spend could be cyclical. And I think that's balanced kind of the question is then where are we into this thinking. That could be sooner. Now supply is high level kind of conversations we're having, so we'll have to see how it plays out. But certainly, it's still relatively early to deploy relative to...

You're fully aware of the [indiscernible], in particular [indiscernible], probably "what developed off of it." Ironically, one there has been a few announcements, still I think actively some evidence [indiscernible] to your [indiscernible]. Any thoughts to...

I mean I still go back to the white box and [ antral ] thesis of make and buy, right? I mean there's no doubt that all of these customers have multiple sources, everybody does. That makes sense, right? So some of the co-development work that we've done with beta is a good example of you get the benefits you bring to the table still have kind of [indiscernible] we're able to create [indiscernible]. I don't think that desire has gone away. I think there's a very good appreciation that they get value from both the business model that works for us. But yes, I don't think white box has necessarily gone away. I don't think it's gotten any stronger. It may have gotten a little in the short term because of some of the supply issues. But I think it's a part of the model that we've learned to coexist with, and it works well for the customer.

Moving beyond the big 4, you've been growing the rest of your [indiscernible] for scale as well as in that [indiscernible].

I mean they're following pretty closely the same cycle, right? They're at the beginnings of adopting some of the new products. They're active on to invest. They're -- I would say they're constrained by supply in the same way. We came out of pretty good growth in kind of the specialty cloud [indiscernible] last year. I mean that's continuing. So I think very similarly, kind of indexed the hyperscale out of the [ site ]. But following the same technology trends and wanting to adopt the same technology.

Margins. You were generating, you always guide [ 53, 65 ] [indiscernible]. So big back offering cost structure [indiscernible] first time in your history. But any thoughts on how long is [indiscernible] impact. If I may, from an operating margin perspective, if you [indiscernible] have always been pretty [indiscernible] about that [indiscernible] there's 25 plus percent probably [indiscernible].

Yes. There's a couple of pieces to that. I think on the gross margin side, and we've long had our 65 with heavy cloud [indiscernible], we talked on the last call about 200 to 300 basis points just around kind of short-term [indiscernible], right? And that's -- normally, we don't [indiscernible] these things, but that's in terms of that's kind of the impact that that's been having and add on the market. If we get released from that or we don't need to do that anymore, that's where we're being surprised on a decommit in short term, and we have to go out and buy something on the market. If and when that gets better, so we should get that back, right? But we think that end of the year, probably into early next year before you see supply being [indiscernible], that's really what it takes to get full relief from that. On the other -- then the rest is back to customer mix and where are we kind of in terms of cloud mix [indiscernible] absent our normal kind of range. So we started to get a little bit of help from the pricing that [indiscernible] or assuming that still 300 basis points continues, give us a little help starting [indiscernible]. There's a lot of moving pieces, but I think the biggest driver is at and have to be...

Those are the price increases a year [indiscernible].

Yes, November.

November. And some of those companies implemented through price increase...

We are just -- talked about on the call, we were just in the process of our next kind of price increase, group up everything that we know of and added to that. That's kind of [indiscernible].

You're not repricing backlog. So there's a time delay before this resumes typically 3, 4 quarters given the size of backlog that you have. Those price -- now the second round of price increase, that fully offset [indiscernible] changed rapidly [indiscernible] necessitated the second [indiscernible] round.

It's scooping up all of the kind of prolonged kind of systemic pain. It's not necessarily solving for the [indiscernible]. That's just -- because again...

[indiscernible] speaking with the transitory issues, right?

Right. So it's looking to address the other -- all the other kind of systemic long sustain...

Presumably, once we get through the whole supply chain crisis return to normal [indiscernible] your margins are more or less prepayment?

I mean that's definitely kind of an objective is to push back to kind of normal. It will take time, right, because we'll look to recover some of those cost advantages. Customers will look to recover some of the price increases. We'll have to kind of navigate our way through that, but obviously [indiscernible].

All right. Let's finish with one last question on revenue, back to revenue. We're in a great demand environment on the supply side, well maybe some [indiscernible]. I listen to you, Juniper, Cisco [indiscernible] 5 major players, and surprising everybody's [indiscernible] whether it's backlog or be it all the forward-looking. Everybody sounds great off the products. We all know that Enterprise Network is growing. Any thoughts in terms of what you're seeing in the marketplace especially Cisco [indiscernible] incumbent?

For us, again, it's still largely one large incumbent because again we're playing at that kind of large enterprise piece of the market. And even the campus market, when you look there, I mean, that's like a 90% dominant incumbent player. So we -- that's -- I think that's probably true in most of the enterprise competitive situation that we're in, is that it will [indiscernible] player every now and again we'll see somebody else, but really still kind of competing for [indiscernible].

And then we're across that territory, [indiscernible]. We have -- you have this great [indiscernible] what do you most expect? Where could be the biggest delta upside or the downside [indiscernible]?

Yes. I mean, look, I think it's important for us to continue to drive the growth across [indiscernible]. It's been really exciting to see enterprise finally hit that tempo, kind of that business being 40% of the company last year. That's a big milestone for me and seeing kind of that consistent growth over time, right? I think that's -- it's an interesting piece of the business that can -- you have the -- if you are successful there, you have the opportunity to do so much more with software kind of solutions for those customers and continue to build on that business and it's early. But I think we've got enough confidence now that we can continue to execute verticals. But there's many, many verticals for us to go and play in that we're not even touching today. So that for me is very exciting. Then we need to absolutely take care of cloud. We will always have that as our bread and butter. But being able to kind of really see all of this technology now be relevant to the enterprise, that's been a big step forward for us. Always suspect that, that would be the case, but I think you're finally seeing it come to fruition.

Is there anything other than supply chain?

I worry about everything, Paul, you know that, reality. But supply chain is #1.

All right. On that note, I want to thank everybody for joining us. Ita, thank you for joining us [indiscernible] if you need any help on the rest of [indiscernible] please let me know.

Thank you, guys.