NetApp, Inc. (NTAP) Earnings Call Transcript & Summary

Well, folks, thank you very much for joining us here at the Raymond James Technology Conference. We're in person in New York City, which is kind of exciting in itself, although it's harder to get up early again. So I'm joined from NetApp, Lance Berger, who is Senior Director of Investor Relations. And so I know for you, it's 6:45 a.m. on your body clock.

It's great.

But I've prepared an outline. I want to go through a couple of thematic questions, help the audience if we have some time or we'll check in.

But let's sort of step back. You reported earnings last week, and it was probably a little bit disappointing for some folks. Some -- a little bit of slowing. The numbers came down a bit. So maybe start out with sort of the update on how you're seeing the macro environment over IT spending? What's -- what are your thoughts on the current situation and how it's changed?

Yes. I think if you look at the last, let's call it, 3 months, right? I mean, kind of midway to 2/3 way through the second quarter. I think we saw a meaningful slowing in kind of deal cycles as far as elongation and appetite by certain verticals to continue to kind of what I'd characterize as normal/aggressive spending. And we tried to be prudent as we looked out into the back half. I think the house view on the macro, we've had this for a little bit of time, which is as we turn into the new calendar year. So if you look at calendar '23 IT budgets, for example, we don't necessarily think they're going to be down year-over-year. But I think if you're a capital allocator putting capital in the ground in this environment, you are more likely to be conservative in the first part of the new calendar year, and that's kind of what's baked into our view. Obviously, calendar 3 ends in January and calendar 4 is April, which is kind of right in that sweet spot or not-so-sweet spot for kind of that dimension. And I think we all are hopeful that as we get into the back half of Calendar '23, things normalize, but that's what we're seeing right now. And when you talk about where we're seeing the weakness, we talked about on the call it being kind of U.S. high tech and service provider, and then some pockets of geos in Europe.

And I think when I reflect on IT spending in all categories, we spend a lot of time thinking about, let's call it, the post-pandemic environment where companies have elevated backlog and supply chain issues have dogged companies, and we're now seeing a recovery. In that context, what's the situation in terms of NetApp's business in terms of how much backlog you've had and what are your experiences with supply chain issues?

Yes, it's interesting. First of all, I'd like to give a shout-out to our supply chain team. They've done a really amazing job in kind of navigating us through that. What I would tell you is, as Mike said on the call, we ended Q2 with, what we would call, normal seasonal backlog. So if you look at throughout the history of NetApp, it's at a fairly normal level for exiting Q2. I think we were pretty clear in Q4 -- Q3, Q4, Q1 that we had some elevated backlog from a seasonal perspective. So that's -- we don't give backlog figures per se, but that's directionally where we're at.

And within the supply chain, the storage market is, I'll say, special and that you've got a dynamic where memory prices are in decline. And in many other markets where they're dependent on other components and shortage, they're paying premiums, and so we've got a variety of movements of the inputs. So when you reflect on your supply chain issues, how do we sort of think about the net result where memory may be cheaper and abundantly available, but other components may be constrained. What's the net impact on NetApp?

Yes. I think the net impact on NetApp is that you won't see the goodness in the supply chain or parts of the improvement until Q4 in our product margins, right? So we've had this dynamic where I think George and Mike made a decision very early on that we did not want to build a massive amount of backlog and delay shipments because we couldn't get a hold of analog pieces. So when you think about the -- what's weighing on product margins, the last 4 quarters, it's really been kind of the $50 million to $60 million per quarter, which is a big number in the construct of our revenue for the premiums we're paying in the open market for analog pieces. Now obviously, with the consumer vertical really slowing down, that's loosened it up a bit, and we see kind of really nice trends to those premiums alleviating. So that -- but that's still -- we have to run through that historical inventory because, you know, to get it to show through in the P&L. So that's where we're at on the analog piece. I think on NAND pricing, it always takes 3 to 4 quarters or 3 to 4 months for that to show up in our P&L. So we expect some goodness from NAND pricing and product margins in Q4. I would say, Mike has reiterated the construct of getting back to mid-50s in the fullness of time. We haven't put a timestamp on that because these things are very fluid, but we feel good about product margins going into fiscal '24.

So I want to ask 2 questions following up on that around your pricing. So had you implemented price increases during the period where we were constrained? And how do you think about, I'll say, industry pricing? So I don't imagine you want to tell me you're going to put prices ahead. But do you think industry pricing gets reduced as NAND...

I would be very surprised if us or our competitors cut what I call list price, right? It will really be on a deal-by-deal basis. And the storage industry is attractive but also interesting from a few different areas where the stickiness of the incumbency is very real, right? And that's why you don't see the market share numbers move around a ton. So if you're in a renewal situation or a customer that is just kind of a repetitive buyer, then you may not necessarily see a ton of pricing pressure. But if you're going after a new account, that's where you strategically use the NAND declines. But it's -- what I would say is the players have been very rational thus far. And I just -- with the margin structure of the industry, I don't see that changing. I mean, I think if you look at some of the peer sets, some are much more reliant on storage to boost their overall cash flow and some are storage only where you can't hide, right? So I just don't see it becoming irrational. It's been fairly consistent.

So I want to paraphrase it to make sure I follow it. So we can imagine markets like PCs and servers, we could see more price competition and price pressure than in a market like enterprise storage. Is that right?

Well, I would -- I don't want to speak too poorly of servers and some of the other subsets. But I mean a lot of it is commodity based, just like with NAND and DRAM, I mean, it doesn't really matter how great your customer is. It's a commodity in a lot of ways, and it really impacts the pricing, right? And we've always positioned and believe that our value is in the software which is obviously not a commodity.

Now at the beginning, you mentioned some challenges, in particular, high-tech verticals. And I want to unpack this a little bit and that what I would have guessed coming into earnings season that Europe would be a problem because of foreign exchange movement, and that what we've heard is Asia Pacific has actually been slow. And then in terms of enterprise, you did mention some elongation. Maybe help us understand what that means?

Well, I think what it means is I think we've all seen what's going on in the Valley, right? And you have -- it started with software and then it was semiconductors, and now it's the big consumer electronic companies. And now, it's rolling into the enterprise companies. I think people are taking a really hard look at their P&Ls. And I think before a CFO makes the decision to cut people, they're going to look at other areas to conserve on. And I think we're going through a consumption phase. I think it's not dissimilar to what you've -- we've all seen what's happened with the hyperscaler CapEx at times, right? They will come out and say, okay, we're taking a pause, we're in absorption phase. And then every single company that you cover, you try to figure out who's exposed to that. I don't think it's dissimilar. So I feel good about the long term. I mean, some of the biggest, greatest companies in the world that are our customers and the biggest consumers of data, frankly. And so the confidence that they'll come back to the well is very high, and we're very sticky customers with them. It's just -- I think it's more want of timing than anything else.

And I wanted to get a little bit of help with you explaining the relationship with public cloud. And you've got business, public cloud services and products selling into that vertical. But I think people get confused as to who the customers are and what they're doing, what the trends are...

Sure.

So maybe if you could help folks understand NetApp's relationship to public cloud as an avenue to market?

Sure, absolutely. I mean, so if you look at it, we have, obviously, deep partnerships with all 3 of the major hyperscalers. They're different, but the same. So if you look at Azure NetApp files, we have our equipment on the back end of their data centers. It's a go-to-market piece that's actually run by Microsoft, so it's a Microsoft-branded solution. It is in the carrying bag of the quota-bearing reps, just like any other Microsoft services. Same thing with Amazon FSX for ONTAP. It's a first-party service sold by Amazon but branded as FSX -- with ONTAP as the underpin of the technology, but it's really being sold by their sales force. And GCP, a very similar arrangement. So it's a great sales motion for us because if you think about our P&L, right, when we sell a system, we have to have a sales rep on the street that is touching the customer. We get the leverage in our P&L. In the fullness of time once this business matures, it will hopefully start to show up more where we -- it's another go-to-market piece where the cost is very different than at the sales and marketing line.

And I want to talk a little bit about your exposure to U.S. dollar and translation. So I believe you do some of your sales in local currencies, and so you have a translation impact. Can you help us quantify what's going on with strong U.S. dollar and...

Yes. We talked about it. We entered the -- we gave our original guidance in Q4, it was a 200 basis point headwind to revenue. Now it's sitting at 500 basis points. And for a company that cross cycle is growing mid-single digits in the core hybrid business, that's a pretty big movement, right? So we have -- we sell about 70% in U.S. dollar, about 20% in euro, about 3% in yen and then 7% in about -- in all other kind of collectively. So it's a pretty meaningful headwind where we sit today. And hopefully, as we kind of go through the rest of this year into next year, that at least normalizes, right? I can't remember the last time that the euro and the dollar were at parity, so hopefully, that doesn't last very long.

And do you have any, what we've been calling, natural hedges where you've got engineering in Europe or you do have...

We do. I think if you -- right. So if you look at the OpEx side of it, it is a natural hedge. So going down the P&L, obviously hedge to revenue, negative hedge to revenue. In product margins specifically, we buy all of our components in USD. So that's a meaningful headwind and I think it was about a 300 basis point headwind year-over-year on product margins, which is not immaterial. And then in the OpEx line, it's a reverse of that. It doesn't cover, obviously, order of magnitude, all the revenue down draft. But we do have a big engineering presence in India and other pockets around the world. So it's been -- it's somewhat of a natural hedge in the OpEx line.

And when we think about NetApp in particular, so you're largely a specialist in enterprise storage. How do you think about what differentiates you from your competitors? What's sort of the NetApp secret sauce or?

Yes. I mean, we've been the best file OS on the planet for 25 years. I think that really is what brought -- really brought the hyperscalers to the table, right? So when you look at the hyperscalers and you think about their strategy and what they're trying to accomplish, there is a massive subset of the workloads on-prem that they simply cannot -- they can't service, because the file OS layer on the storage side becomes the bottleneck if they're using their native solution, right? So they saw that, and I think that's what really led to the partnerships is the ability to go after SAP workloads, AI workloads, genomics workloads. I mean, anything that's super high performance. Now I think if you look at on-prem, our secret sauce really is the ability to service that workload or any subset of workloads across the entire spectrum. So when we go into a customer, we can -- we're kind of agnostic to where the data resides. We're agnostic to what media it rides on, whether it's spinning disk, hybrid disk or all-flash, we feel like our all-flash is very competitive with the rest of the market. I think the interesting thing, and I think this is sometimes missed by the investment community, is that the makeup of the market share dynamics between spinning disk, hybrid and all-flash is structurally different. So if you look at the top 5 vendors within all-flash, we make up 85% of the market, right? and that's including IBM and HP, ourselves, [indiscernible], right? If you go down and look at the spinning disk part of the market and the hybrid, I'll include hybrid in that, the top 5 vendors make up 50% to 55%. So there's a lot of hoo-ha out in the market around the market share shifts between the top 3 to 4 vendors in all-flash. I think the reality in the fullness of time is all this value is accruing to those top 3 or 4 players. And I think if you look out 10 years from now, I think that's a tailwind for all of us.

And where is your business mix now between spinning disk and flash ?

I'd say, it's probably 50-50 plus or minus, depending on the quarter. But trending -- obviously trending towards all-flash.

And I guess one of the arguments I've heard is that as NAND prices come down, flash will be able to displace spinning disk use cases faster and faster. Are you seeing that? Do you believe that argument?

Absolutely. I mean, I think that TCO argument has been kind of on the table for 10 years and people are -- have been forecasting the demise of hard disk drives for a long time. So clearly not at cost parity because Moore's Law applies to everybody. But I do think in the fullness of time as NAND pricing comes down, it just incentivizes. I think one of the other things that people miss is that NVMe is the protocol that was written specifically for all-flash, and all the storage vendors are ready to go. The reality is it has to be written on the application side of the house for it to be meaningful. In other words, if the application is not written NVMe, doesn't matter that your storage box is NVMe-enabled. So it does -- for a new protocol to take real solid hold in the industry, it takes 5 to 10 years. We've all seen that in past cycles, and we're probably in the fifth inning of that. And I think the next 5 years will be a really interesting turning point from that perspective.

Now there's been a theme across a lot of data infrastructure of migrating to more consumption-based business models, recurring revenue business models. And one thing that I've observed is NetApp's put a much greater emphasis on the public cloud services and what you've been doing there. But you have an offering called Keystone, which is an as-a-service offering. And I guess, I'd like to just get an update. I don't feel like we hear about it, and part of the reason I'm intrigued is because some of your competitors talk so much about their as-a-service business; NetApp talks about there's less. So it's not that you don't have it. What's the thinking? How is that business? Does it fit into the strategy?

Yes. You know, it's interesting. I think one of the reasons we don't talk about as much as the competition or, maybe I should say, the competition talks about it more because we have a cloud offering, right? We view as if a customer really wants to go as-a-service, let's do it as-a-service as opposed to a leasing model, right? And look, I mean, if the box is on-prem, you can call it whatever you want. It's still a system on-prem. It really -- it's a financing tool, right? And there are some of our competitors that include interest income in their ARR number, right? So when I look at it, I'd say, if a customer really wants to go as-a-service, something that can burst up and down where is OpEx and you want true flexibility, we view the cloud as the path for them. To be fair, just like we do spinning disk and hybrid and flash, we are truly agnostic to what the customer wants. If the customer comes to us and says, we want a subscription model on-prem, then we have Keystone. If they want to do perpetual model like the traditional IT sale, we have the rest of our core business. And if they want a migration in cloud, we obviously have the cloud.

How big is the Keystone business?

We haven't sized it, and it's -- I would say it's not material at this point. Yes.

Not at this point. So that was sort of leading into your public cloud strategy. So I think there's some -- a little bit of maybe helpful to educate folks as to what are the components and the nature...

Sure. So if you look at kind of the makeup -- let's just take it as a kind of a 100% pie, if you look at it, 60-40 storage versus cloud ops. Within cloud storage, we have 2 major offerings, cloud volumes ONTAP, which is really where the customer brings their own license that they purchase directly from us, they take it to the hyperscaler and run it on that commodity hardware. But it's really -- that part of it is really driven by the NetApp sales force. And then the bigger part of our storage layer in cloud storage is CVS. So that's Azure NetApp files, FSX for ONTAP and GCP is where the actual hyperscalers themselves are taking our solution, embedding it into their platform and selling as a first-party solution. So that's the 60%. And then cloud ops. You have Spot, which is made up of Spot and CloudCheckr, which is really compute optimization. You have Instaclustr, which is open-source database. And then you have Cloud Insights, which is our historical OCI software which was sold to on-prem customers that we have transitioned to a SaaS model. And so those are -- and that makes up the other 40% of the...

And I think you've provided some disclosures as to what percentage of the customers are new to NetApp, and candidly, I've been surprised by that. Could you help us understand what's making -- essentially drawing those customers that aren't historically NetApp customers to utilize you in the public cloud?

Well, I think they're going to the hyperscaler partners and looking for specific solutions, right? So that's on the CBS side of the business, because that's where the hyperscalers are really driving the sales motion. And when they have a relatively high-performance workload, if Azure can't service that workload with their bare metal offering, they are incentivized to win that business by offering them Azure NetApp Files. So I think that's really what's driving the -- like the ANF and FSx where they're coming aggressively after new workloads that maybe Microsoft and AWS haven't been able to service in the past, and they're excited to win that business. I think on the optimization, cloud ops side of the business, they are looking for verticals that help them optimize their spend. Spot was a fairly well-known brand within that ecosystem as well as Instaclustr. And so what you're getting is kind of cloud native customers that haven't really been exposed to on-prem storage before now being exposed to one of our properties.

And on the recent earnings call, you talked about some slowing down of consumption-based activity. What's influencing that?

Well, I think if you listen to the earnings calls of Google, Amazon and Microsoft, they spoke pretty clearly about this, right? I think people are trying to -- I think there's -- any capital allocator in this environment is privately thinking incrementally more carefully around each dollar that they put in the ground. And I think as they go through that, they're trying to optimize their storage spend. I mean, people talk about it being an OpEx model, but the good thing about that for the customer side is it's an OpEx model that is also very flexible. So in this environment, if you can look for areas to potentially reduce your OpEx exposure, that's one of the places you're probably looking.

And I guess one of the big themes is around just public cloud adoption. And so NetApp is may be a little bit unique and that you've got an offering that's public cloud and you've got an on-premise offering. So net-net, when you think about this trend of enterprises embracing the public cloud, is it a net benefit? Is it a net headwind, but you've got some offset? How do you characterize that sort of big picture transition of public cloud adoption by enterprises?

I think it's absolutely net positive for us. I mean, we have, what, 12%, 13%, 14% market share on-prem. If you want to move your SAP workload to the cloud, there's one place you can do it, right? So I think George's commentary internally for a long time has been if we can't get 12% to 14% of the high-performance workloads moving from on-prem into the cloud, then we and our partners are doing something wrong.

And when you think about this position and success in the cloud, is there a cross-sell opportunity? Have you seen cases where you've secured a customer in your cloud services, and then they become a customer on prem? How prevalent is that?

Absolutely. I mean, it's becoming increasingly prevalent, especially when they're doing high-performance workloads in the cloud, and they're starting to -- they mature as companies, they have very sensitive data that they don't necessarily want on the cloud and they're looking at how to diversify their footprint. We've absolutely picked -- this has been our thesis for 5 years, and it takes a long time for these things to maturate, but we feel good about that. I don't know if we're taking questions.

Sure. Go ahead. And we'll repeat the question for folks on the webcast.

Yes, I wanted to ask, and maybe it's a follow-on to that topic. But you had outlined maybe high tech U.S. enterprise and areas in Europe are a little softer right now. How do you compare that with the cloud side where you're seeing demand signals? And what gives you confidence in the back half of...

So to paraphrase the question for the webcast, basically, we've seen some softness in enterprise in Europe. Should we anticipate the risk of softness from the cloud customers and the cloud operators?

Well, I think the reality is we have seen that, right? I mean, I think on the last -- the most recent earnings call, we talked about the slowdown in cloud. All 3 hyperscalers talked about that slowdown. I think it's more of an optimization as opposed to people taking workloads offline and moving them back on-prem for, let's say, underutilized capacity. I don't think that repatriation is what's driving the slowdown in cloud. I think people are really looking just to optimize our OpEx spend. And I think until we kind of get through this macro headwind that everyone sees out there, then I think that's probably a reasonable way for capital allocators to be behaving.

Got a question?

I think your peers done the same thing in cloud and are offering it through Amazon, like why are you sensitive?

So just to repeat the question for the webcast, why is it that NetApp has these partnerships with the public cloud providers that your competitors don't?

I think one of the very specific areas of what I would characterize portfolio holds for the hyperscalers, let's call, 5 to 7 years ago, which is when we started these discussions, was an enterprise-grade file operating system. And I think when you look at enterprise-grade file operating systems, we were at the top of the list for the partners, and I think George realized this trend. I think you have to realize every competitor has a different makeup in the revenue stream, right? I mean, if you're Dell and HP, you're protecting a lot of the stack that AWS and Azure are going after. So I think they were a little more reluctant to shake hands firmly with the hyperscalers. And I think George's view was kind of like I reiterated before, we have 10% to 12% market share on-prem. If we can have 100% of high-performance workloads in the cloud, why would we not go that way? I think his other view was that this is not a tidal wave that NetApp is going to stop, the movement to the cloud. So he embraced it really aggressively. I think that was actually part of his vision in winning the CEO role, what he did.

So one of the things that I sort of knew is NetApp is generally thought of as an enterprise storage supplier, yet series of acquisitions that you made over -- time's a blur now, within 2 years, CloudCheckr, Spot among those, aren't strictly storage solutions. And so how do you think about the general direction and how should the investment community characterize NetApp?

It's interesting, right? I mean, I completely agree with you. We've kind of made a strategic pivot. I would view it as data optimization, right? Data optimization both on-prem and in the cloud. We've extended that, obviously, beyond storage with CloudCheckr and Spot and Instaclustr. But that's kind of how we view our role in the ecosystem.

And so you went through a period with a number of acquisitions. I think the messaging, and I want to verify it is, we're sort of done for now with acquisitions. We're in the execution phase. So how are we thinking about use of cash? And what is the possibility, probability of more acquisitions when do you think that comes back into the mix?

Yes. I mean, the only thing we've committed to is through the end of fiscal '23. I want to make that super clear, right? First, it was the -- we had the kind of the blip in Q4 of last year, so we took a pause in first half. We now have these macro headwinds. We want to focus internally in getting the properties appropriately integrated, so we will take a pause through the end of this fiscal year. And as a result of that, Mike kind of upped the ante on the buyback for the rest of the year. I wouldn't get over your skis on magnitude of incremental buybacks, but it will be more than 100% of free cash flow that we will return to shareholders this year.

And is it fair to say that you're still sort of on the pause on acquisitions and in the execution phase on the acquisitions you've made? Or should we sort of say at some point in the not-too-distant future, we might see acquisitions come back into strategy?

Yes, I mean -- yes, our strategy and development team is always looking, right? So I don't think Mike meant to suggest that we've paused the acquisition engine. We just don't anticipate closing acquisitions in the rest of this year. So we will always be looking for great fits at the right price.

And are there any sort of key points we should be thinking about in terms of use of cash, whether it's expenses coming back because travel comes back, or any kind of debt issues? Anything in terms of use of cash or expenses that you see coming up over the next year that is not captured?

No. I mean, we've made it -- it's interesting. You look at the incremental FX headwinds and the down draft in the top line guide. And what we said is we've absorbed most of that, the vast majority of that, in kind of disciplined spending. And so I wouldn't expect us to get over our skis on really amping up the OpEx line until we get further visibility into how the macro is going to play out.

Believe it or not, we've run out of time. But folks, thanks for the questions. Thanks for joining us. This was Lance Berger with NetApp.

Great.

Thanks, Lance. Great seeing you.

Great. Thank you. Appreciate it.