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

All right. Perfect. Welcome, everybody. We're excited to have NetApp here. Before we get started, I'm going to read some safe harbors both on my account and on NetApp's apart. So for me, for important research disclosures, please see the Morgan Stanley research disclosure website at morganstanley.com/researchdisclosures. If you have any questions, please reach out to your Morgan Stanley sales representative. And for NetApp, today's discussion may include forward-looking statements regarding NetApp's future performance, which are subject to risks and uncertainty. Actual results may differ from these statements made today for a variety of reasons described in our most recent 10-K and 10-Q filed with the SEC and available at our website at netapp.com. We disclaim any obligation to update information in any forward-looking statement for any reason. All right. With that exciting start.

Nice job Meta.

Exactly. All right. So for everybody, I'm Meta Marshall. I cover the networking and enterprise storage space at Morgan Stanley. We are delighted to have Mike Berry here today, CFO of NetApp.

Thank you. Thanks for having us. We're delighted to be here as well.

All right. Perfect. And so maybe just kind of starting out with current macro environment, you guys are seeing some headwinds to both kind of the premise and cloud segment, is there any difference to the timing of how you guys see these 2 businesses recovering?

Yes. So again, thanks for having us. So while we think that we really like our position going forward, certainly, the macro headwinds have affected both call it, the hybrid cloud business and the public cloud business. We saw a lot of large enterprises. We talked a little bit about this last quarter as well. U.S. hi-tech as well as service providers really take a step back. I think that, that also applies to what we're seeing in the cloud. Not only us, but I watch a lot of the software names, and I get all of the research and the optimization that's happening has also happened with us. We saw some really nice growth early from some of our larger customers in cloud, as they deployed project-based workloads and as those build up and came down. So I think the macro plays into all of it. There's a couple of nuances for us, but largely, it's pretty much the same. In terms of the recovery, it looks like calendar year '23 will be tough. We feel a lot better about the second half of our fiscal year, which also aligns much better with some of the industry growth as well that they're calling for in storage.

Got it. On the cloud portion of the business, you guys continue to have a unique opportunity with your native cloud solutions. What will be the biggest driver over the next couple of years? Is it the expansion to AWS and GCP? Is it better alignment with applications like SAP or kind of the realignment of product and sales efforts on the cloud op side?

Probably the first 2 more than anything. So let's go through those. So yes, we feel really good about the cloud products we have. The native integration in the first party we have with Microsoft, that's ANF, and with AWS with FSx for ONTAP. We feel really good about from a growth, you should expect to see better growth in cloud storage just because that's where, hey, SAP, VMware, a lot of those workloads move to the cloud. We're seeing more net new workloads in the cloud versus data migrations, especially in a tougher macro. From a cloud ops perspective, we had the issues in Q4, but we're largely through that. Spot continues to grow very nicely. You should expect to see cloud ops be a stable grower, not huge growth, but the majority of that growth will come from cloud storage.

Okay. Got it. That business has been more difficult for you guys to have visibility into. But a couple of quarters of this cloud digestion in the books, what have Wave -- I'm sure you have been looking for Wave to gain visibility there that you found most effective there?

So we -- and you talked about the go-to-market a little bit. So if we talk about cloud storage and we'll separate Cloud ARR into cloud storage, which is about 60% of the business. Within that 60%, call it, about 2/3 is consumption. So this is versus subscription, which is more of the CVO, bring your own, our ONTAP product. But when you get in that consumption business, getting visibility into what those customers are going to do, is we could do a little bit better, and we focused on this a lot. Number 1 is understanding the workloads that they are deploying. So we talked about we saw great growth in those project-based workloads. And now as we talk to them, we want to be able to forecast take because at some point, data will build, it will build and then it will come down. I think that in addition to that, we probably, from a go-to-market perspective need, and we talked about this a good bit on the call, need to align our teams better with their sales structure, so that when they're in trying to burn down those big commitments and understand where those customers are, that we are aligned a little bit better with that. And we'll make those changes going into fiscal '24. I think the good news about this is, hey, especially those larger customers have largely -- that data has dropped, the ARR has come down. So going into Q4 and then into fiscal '24, we don't have that, call that, big piece in ARR, which we know is going to optimize down. Hopefully, it just grows from there.

Okay. Got it. That's helpful. You just talked about it. Hi-tech and new workloads have been the driver of that cloud storage business. And you talked about the majority of the growth coming from new workloads. What is that process? Or are there ways in which you can kind of incentivize customers to move their premise workloads to cloud?

Yes. So most of the focus, especially in a tougher macro. And I'll say as the CFO of NetApp, we have the same discussion in terms of the economics of migrating data to the cloud, probably not as appealing as deploying net new workloads. So this is where SAP, VMware come in, high-perform AI. I know it was the big thing people are talking about now because now you don't have to procure all of that hardware, you can burst it into the cloud and then it allows you a lot more flexibility. So this is what we do with the hyperscalers. Not only are we in with our customers as they make that cloud journey, helping them move to the cloud, but also, there's hundreds of billions of dollars of commitments out there with the hyperscalers as how do we burn down that commitment, and that's where we come in, in terms of utilizing their product or ours to help that. And a lot of that focus is really on net new and net new workloads. ANF is very focused on the high-performance workloads because, again, we deploy hardware in their data centers. So you're going to see great performance there. FSx, it's software. So that's been more, call it, secondary workloads, but it's moving its way up. And then GCP has a little bit of both.

Okay. That's super helpful. I think that just took care of my next question. You've had a year of reorganization on the cloud ops business. Where are we in the scheme of sales changes or product changes that you wanted to make?

Yes. So in cloud ops, we talked in Q4 about as we integrated those businesses that we had some issues in terms of just sales capacity. We've largely rectified that. And so Spot in all 3 quarters of fiscal '23 has largely hit our expectations. So from a sales perspective, a lot of that is largely done. We have 1 sales leader, and she's done a great job of pulling those together and also integrating Instaclustr. So from a go-to-market, largely check, we feel good about that. From a product perspective, there is some integration between CloudCheckr and Spot because they do a little bit of the same, not really. CloudCheckr brings great MSPs. We sell into the service providers that something Spot didn't have. The other thing CloudCheckr has is FedRAMP compliant for some of their products, which is great. So we're bringing those into Spot. So it's mostly from a product side. And then we just recently said, "Hey, we're going to pull back on Spot PC, the VDI business and focus it where we think we can drive growth. So we feel good about where we are with cloud ops. A lot of work going on still there, but it's not a big lift. It's just combining those road maps as they go forward.

Okay. So it's mostly a product no longer.

Mostly product. Yes.

Okay. The -- you've talked a lot about Spot doing quite well. We've talked a lot about cloud digestion and rationalization efforts, which should clearly help Spot. Has there been a way to differentiate this product against kind of other cost optimization products?

Yes. So we feel really good about Spot being a full FinOps solution. So we have multiple products in there. Spot Eco, which is, call it, almost entry level where you're going to manage your reserved instances and your savings plans. And then we have a product that also helps as you deploy Kubernetes in the cloud, and then more of a -- called Elastigroup which think of that as a much more higher scale. So we have all that full breadth and then we'll take CloudCheckr and add it in, in terms of MSPs, and then also the FedRAMP piece. The other thing that we differentiate is we work across all 3 of the hyperscalers as well, which allows you, no matter where you're going to deploy -- now most of it will follow the market share of the hyperscalers in terms of most of it being with AWS, but we feel really good about that set of FinOps. There's a number of competitors coming in there. This is also where we can judiciously use our go-to-market resources to help drive that as well. And it is a little bit of the flip side of the optimization that we get on cloud storage. You see the benefits in Cloud Ops.

Okay. Got it. So kind of breadth of solution and kind of targeting this for go-to-market. Okay. Another question that we've kind of had and have posed to multiple companies, but you guys have a great insight into is just how do you see some of these cloud optimization efforts? Is this a year of a digestion period and we reemerge at the same trajectory? Or are the customers telling you anything differently about how they look at these cloud transitions going forward?

Yes. So we've obviously had a lot of discussions with our customers as well as our wonderful hyperscaler partners. And I think all 3 of them have talked publicly about, hey, it's probably another 2 to 3 quarters. They see it even more than we do because of the breadth of workloads and certainly at a much higher scale. I think this goes a little bit to the economics of cloud versus on-prem. And in a tougher macro, it's a little bit easier to, hey, I have the capacity on-prem. Data migration, we talked about that, probably lower priorities. We think over time, hey, the growth in cloud is an enduring growth and it's going to continue. There's just so many more reasons to deploy in the cloud from flexibility to being able to burst up and down. That's what we've seen a little bit. So we feel really good about cloud growth in a tougher macro. I think people will pull back. Hopefully, it's something as we get to the second half of calendar year '23, gets a little bit easier. And from our perspective, if you kind of pull out those big customers, we've largely tracked with their growth as well. But we -- hey, we think it's a great growth opportunity long-term. We're also learning how to run that business in a tougher macro.

Got it. One of the more exciting announcements that you guys made on the earnings call was just the introduction of the C-Series as a way to kind of improve competitive positioning in all-flash. What type of customer or application type does this help you with? And just how do you look to gain share within a market with where there's -- is this not a competitive [ listness ] market?

Yes. Well -- so we're super excited about the introduction of C-Series. So we have to had QLC in our lineup. It's been more focused on the higher-end workloads. So we've largely focused on ANF and then you have Hybrid, which is going to be a combination. This to us is a great add because it all runs on ONTAP. So any of our customers that use ONTAP, on-prem and in the cloud, that journey fits really nicely. It's going to be workloads that are going to be more capacity based, but as the price of NAND comes down and the total cost of ownership gets better with all-flash, we like that as well. It also, for us, will allow us to go into net new, whereas before, you're probably not going to get net new with a hybrid environment. So that -- while it doesn't expand enterprise storage, it allows us to go after that group. So we feel really good about that. The sustainability, obviously, with all-flash, is better. The other thing we like is if the high end of all-flash has a certain margin profile, and then Hybrid is going to be lower, the C-Series is going to be higher than Hybrid, a little bit lower than high end, but as some of those workloads probably move down, we feel good about that from a margin perspective as well. So we're super excited about the full lineup to be able to go out with that.

I know this question was asked on the earnings call, but I think it's helpful to just kind of rehash it about, what kind of gives you confidence that it's not a cannibalizing product?

So for us, we think that there are workloads that they just run much better on high performance. And this is, as we saw in the last couple of quarters, U.S. tech, large enterprises, service providers, things like AI, ML, the other workloads that run much better, we don't think will come down. For us, it's as much about being able to replace the 10K drive market and to be able to move them up into all-flash. So at least in the next, call it, several quarters, we think it actually should be a tailwind to margins versus a cannibalization of the high end. There's just different workloads that run a lot better. It doesn't have the performance characteristics that you need to run those high-performance workloads.

Okay. Got it. That's helpful. You talked about maybe 2 different things on the earnings call, which is that customers traditionally have a dollar amount in mind that helps insulate against price fluctuations on the flash market, but also seeing kind of customers opt for smaller drive maybe. What are you just seeing with demand on the flash side given macro conditions and just given where we are with NAND pricing?

Yes. So 2 separate pieces there. If we take the demand in terms of especially the large enterprises and our U.S. tech, they almost always procure the high end because, again, your previous question. We saw that demand come down, which had an impact on all-flash. And that also then made its way to the product margins because those are going to be higher margins. Within the purchases that we saw, we did see lower capacity, as we call it capacity per terabyte come down a little bit, because folks have probably less dollars to spend. They need to fit their storage requirements in that. So we got hit with a little bit of both in the quarter. Again, that's not a cannibalization. It was, hey, demand wasn't as much as we would have liked. And then within that, we did see that come down. To that point, hey, NAND pricing will help, I think, over time with total cost of ownership and sustainability, everybody rightfully so has that as a driver. So they're going to probably move from 10K drives into that market.

Okay. Got it. That's helpful. Are there considerations on margins just as prices fluctuate in the NAND market? We've gone through a year of supply chain challenges, where you may be holding a little bit more inventory that maybe exasperate these issues one way or another?

Yes. So a lot in there. So let's talk about -- so inventories for us, we had a certain demand forecast for the year. And gosh, it's almost been about a year ago when we had that little contamination issue in NAND. We did bump up our inventories at that point to more than anything, lock-in pricing. That's taken about 3 quarters to work its way through the balance sheet. Last quarter we're down about $70 million in inventory, the vast majority of that was SSDs. So now we're in a much better level. We've not been able to benefit from lower NAND because of our inventory position. That now is going to help in Q4 and significantly helping in fiscal '24. The other thing for us that we've had that other folks at least haven't talked about are these premium issues where we've had to procure parts for low-value analog parts in the open market. And that's been about $40 million to $50 million a quarter. It's a big number. That is going to go away in Q4 as well as next year as well. So as we look at the COGS situation, our product margins, we feel really good about that going into next year. In addition, gosh, darn, hopefully, FX doesn't become -- at least it's flat, not a headwind as well. All of that will help product margins as we go into next year.

Got it. Just given the time that we're coming out of, you mentioned contamination issues, supply chain issues, I think a conversation we're having with a lot of companies is what level of inventory do you feel comfortable kind of holding after all of this? Just how are you looking at inventory considerations or just that being a source of cash that [indiscernible] in?

Yes. So great question. And it has been a big, I'll call it, drag on cash flow in terms of going into this year. So we look at it in a couple of different ways. You see, we disclosed total not only inventory on the balance sheet, but also what we have in our distribution partners. For us, we want to get back to a reasonable, I'll call it, weak supply of SSD. I think it will be interesting, you'll probably see a little -- it won't go down as much as we would like because we do want to make sure we have QLC. That's not a big market as of yet from a supplier perspective. So we want to make sure, if we see a nice bump in demand. Everything else, Meta, you should expect to see come down. I think from where we are total inventory, a little bit lower, not a ton lower, which is largely consistent with where we were 2 years ago before we got into all this fun. And the big thing there is just, hey, we've also had to hold those premium parts because we didn't know if we were going to be able to get them. And that was a big cash use. If it's $50 million a quarter, that was $50 million of cash going out the door for us.

Okay. Got it. A question we've gotten a lot from investors is just OpEx came in very low in fiscal Q3, but the guide didn't necessarily indicate a huge step down in post Q4 with the risk. How do we reconcile the 2?

Yes. Okay. So let's talk about that. Again, big Duplo Math, not exact. We're at about $700 million in Q2. We guided relatively consistent with that. The big drop in Q3, 2 things. The largest was, hey, we basically adjusted our accruals for incentive comp for the full year, where we had accrued a certain amount in Q1, 2. And then 3, we said, hey, we're connected with shareholders. We're not doing as well as we would like. So there was a big credit to incentive comp. The other part was we did, from a timing perspective, pull back on some program spending and some discretionary spend. So that was the, call it, $50 million quarter-over-quarter drop. The guide for Q4, call it, $675 million, brings -- we're still accruing at a lower amount, but pull that credit out, which is going to be 70% of that drop. And then we'd get a little bit of benefit, call it, 70% of the restructuring. That will then roll into Q1. So what you should see in Q1 is a little bit more drop from the RIF, but then we will also bring incentive comp, call it, back to the base level. So you won't see the, call it, the $650 million level more where we are with Q4 going into next year with a little bit of bump. All that depends on how we perform. And when I say incentive comp, I mean not only base incentive comp for the employees, but also commissions.

Okay. Okay. That's very helpful. Where was the risk mostly focused and, I guess, maybe there was a little bit of uncertainty or just as to kind of what directives maybe you pulled back on with that risk?

Yes. So when we -- think of it as largely consistent with our head count. It wasn't focused on one area. If it was 8% across the board, some were, call it, 5%, some a little bit more. We try to look very carefully at where we thought we had overcapacity, or we weren't seeing the return on the spend. And also -- and we've talked about it with our go-to-market going forward, we do want to make sure that we are focused the resources in the right spot. So it was largely consistent with the employee distribution. We did pull back on, we talked about Spot PC. We are integrating CloudCheckr into Spot. So there was some of that. You will see a little bit, and it's not a big number, a little bit of add back because we did cut a little deeper in certain areas, and we'll add back. We're also pushing, and the team has done a great job adding back on lower cost locations. So even though you may see head count go up, dollars will go up accordingly with that. The one commitment we will make is, I know a couple of years ago, we came and said, "Hey, we need more capacity with sales." We didn't do a good enough job, making sure that we had enough to go drive net new. We feel very good about our sales capacity based on where we are today. So going forward, we'll add back judiciously again where we cut a little bit deeper, but we also feel really good about our capacity and our level of productivity going into next year.

Got it. And I have a bunch of more questions, but are there any questions from the audience? You can scream it and then I'll repeat it.

Much faster than your fundamentals, can you sort of parse the 2 together and -- from a capital allocation standpoint, what share owners look like for next [indiscernible] the trends at SPC shares outstanding and buybacks?

Sure. Thank you. So I think most of the folks got the question. So in fiscal '23, just to make sure we have the same facts. Fully diluted shares outstanding actually will decline by 4%. When you look at fully diluted shares this year in '23, we front-end loaded the buybacks in fiscal '23. So we're down pretty significantly. From -- going into next year, we haven't guided next year yet, but what we tell you is, hey, our goal is to keep it at least flat. When we look at capital allocation, we're actually spending now more than 100% of free cash flow this year, and that includes -- not accelerate, we would even do accelerated, but higher buybacks early on. You'll likely see us do that next year as well because we want to keep share count at least flat. Thanks for the question on SPC because I know I got this question on the call. The bump you saw in Q3 up to $93 million in share-based comp, we have an ESP program, Employee Stock Plan, that looks back 2 years. Every other quarter, you're going to see us have to do a look back to say, what did we actually buy those at? And there's a catch up. In the quarter, that was about $12 million of a onetime hit. It will drop down a little bit in Q4, and then you should expect to see it increase with -- basically with the rest of OpEx. But that's why it bumps up in Q3. I was reluctant to guide on the call when I got the question just because, hey, it happens every 2 quarters, and we have to look back, and I don't know what we're going to buy those at. So that was the bump in Q3. Hopefully, that helps. But again, fiscal '23, fully diluted share is down 4%. You should expect to see it at least flat next year. From a capital allocation perspective, the dividends, it's $2 a share per year. That's about $420 million. And then the play is acquisitions versus share buybacks, and we've over-indexed on share buybacks during the last several quarters.

All right. Well, that takes care of my capital allocation. That's very helpful. And so maybe just to end with, you've talked a lot about AI, ML workloads and kind of the opportunity there. I mean just how does NetApp better position itself to capitalize on the AI opportunity that's much of the theme of the next 4 days or kind of take advantage of AI within NetApp?

Yes. So thanks for the question. So we feel like we have a really good position in AI. And we've actually been, from our external customers, that's been a big workload for us already. Not only on-prem and as customers burst to the cloud and we talked about it. It's a great workload to move to the cloud because you don't have to do all the hardware. We have our ONTAP as a fully -- we have an AI solution that's a fully converged solution. We partner with other folks that are obviously talking about this a lot, the NVIDIAs of the world, to deploy that with our customers. It's such a massive amount of data from edge to core to cloud, and we really help optimize that. So we do think it's going to continue to be a growth driver. In the next couple of quarters, maybe, maybe not, but especially long-term in the cloud. And this is where, I think, especially ANF, because of its high performance, will really do well in the cloud.

Perfect. All right. Well, with that, we're basically out of time. So Mike, thanks so much for being here today and letting us know a little bit more about NetApp.

Thanks for having us, Meta.