Sandisk Corporation (SNDK) Earnings Call Transcript & Summary

All right. Welcome back. I'm Joe Moore, Morgan Stanley semiconductor research team. Very happy to have with us the executive team from Sandisk, CEO, Dave Goeckeler; and EVP, CFO, Luis Visoso. Thank you, guys.

Hey, Joe, thanks for having us. We appreciate it.

Thank you. And I think, Luis, wouldn't you want to read a safe harbor first?

Yes. We will be making forward-looking statements in today's discussion based on management's current assumptions and expectations including with respect to our technology and product portfolio, our business plans and performance, market trends and opportunities and our future financial results. These forward-looking statements are subject to risks and uncertainties. We assume no obligation to update these statements. Please refer to our annual report on Form 10-K and other filings with the SEC for more information on the risks and uncertainties that could cause actual results to differ materially from expectations. We will also be making reference to non-GAAP financials and reconciliations to our GAAP to non-GAAP financials can be found on our website.

Great. So you guys saw this coming, right?

Well, I mean, we saw some of it coming. [indiscernible]

Well, celebrate for me. I mean really, I'll give you credit. You said this time last year that we would have a stronger second half. You were right in a pretty big way. And I guess just maybe kind of walk us through what did people miss. How did you guys -- you saw some of this improvement coming. Obviously, AI surpassed all of our expectations, but you know?

Yes. I mean we've had conviction for a while that the market -- yes, I think we've been saying for quite some time that we thought the market would be undersupplied through the end of '26, and that was as far as we were doing modeling at the time. And we were pretty clear about that last year when we launched the company. How do we come to that conviction? I mean we do, obviously, a lot of bottoms-up research ourselves. We have a lot of insights into the market because we play in them all the time. So we do models on demand on all the big 3 markets on smartphones, PCs, data center. We obviously do a lot of modeling of supply. Again, we have some unique insights. We have some people on our team that happen to know a lot about how you design NAND and what different people's road maps may look like and what supply is going to look like. And we would just consistently look at that picture internally, and we would just see that we thought coming out of the '23, which was kind of an industry-defining event that this market was going to get to a very different place. We have a lot of conviction that the way you have to manage the business is different. You can't just release new nodes because they're available and think they're going to drive your cost down and the market will just absorb all that supply. We thought that playbook had completely changed. And that turned out to be true. And then as we move through the second half of last year, we saw the data center keep raising CapEx spend, I think, which was very well understood. And again, we have our own models of how we translate that into calendar year '26 exabyte demand. And so it's just -- we went through 3 -- 2 forecast cycles. We went into, let's say, this time last year, late last year, we thought data center would grow mid-20s. Next forecast cycle, we thought it was going to grow mid-40s, and then we came out last -- our last earnings call, we talked about, now we see it growing mid- to high 60s. So yes, I mean, again, we just -- we do a lot of modeling bottoms-up on both sides of the equation and take a clear eye to look at it and see where we think the market is going to go.

I mean the data center storage market has been really surprising. I guess we saw huge orders in the fourth quarter for immediate delivery. How much of that do you think is because of hard drives? And I guess, I don't feel like it's a hardware replacement because I think it may be accelerated some moves to solid state that might have happened later. But the overall storage, how does that -- what role does that play?

Yes, I'm not a believer in the hard drive substitution question. I mean it's not lost on me that 2 years ago, I was...

You know a little bit about the hard drive.

Well, a little bit. But 2 years ago, 2.5 years ago, I was answering all these questions that no hard drives are going to be sold in 5 years, right? And we've pivoted 2 years later to 1 million questions about there's not enough hard drives in the world. So I think as a NAND industry, we just need to leave hard drives behind. NAND is a different market. Hard drive is great -- I love that business, too, know a little bit about it. Great business. Great technology. NAND is the same way, great business, great technology. I think I've said this very consistently for the last 6 years. These are complementary technologies in the data center. In the device, they're substitutes. But in the data center, they're complementary. And NAND is -- they're both going to grow. NAND is going to grow faster than hard drives. And I think that is borne out to be true. And I think with AI, NAND is now being fully integrated into the AI architecture. That's something that we thought was going to happen years ago. We didn't think it was going to happen years ago, but we were looking at the technology. NAND is the most scalable semiconductor technology out there. When you have an architecture that's scaling like AI is, NAND is going to come into that architecture at some point. That is happening. And that, of course, is accelerating the data center growth rate.

And then on consumer, you've seen shortages as well, and that surprised me a little bit. Is that just because there's demand coming from the enterprise side? Is it just -- what's driving that thing?

Yes. I mean I think shortage is an interesting term. I mean we just have a market -- markets have supply and they meet every day, right? And that's rationalized through pricing. And as that gets rationalized, if your business model is dependent on buying NAND from 1 supplier and turning it into something else, as your costs go up, it's going to make your business model less attractive. And I think it's natural, you're going to see business models that depend on readily available inexpensive NAND are going to be under pressure. And I think this is kind of the -- I think, still a somewhat misunderstood part of the NAND market. The NAND buyer is just fundamentally changing. This used to be a market that was more of a commodity market. The buyer wanted multiple suppliers for every point in their deployment. And that's being replaced by a buyer now that is the predominant part of the market. The largest part of the market is now a buyer that is not -- their business model is not dependent on the volume of NAND they buy. They're not reselling NAND in a different form factor. Their NAND is part of a much larger architecture that is very profitable. And their consumption continues to go up. And I think that is what's catalyzing a change in the business practices in the market. And I think that's really what's going on. And we're in the middle of that. And when you're in the middle of something like that, it's sometimes confusing, and it's hard to figure out what that's actually going to happen, but I think it's just incredibly exciting. And it is -- this technology that we've been building for 25 years, it's extraordinarily important, and the fundamental dynamics of the way the market works are changing. And it's going to be fun to see how it plays out.

Yes. Yes, it has been fun so far for sure. From the standpoint of capital spending in NAND, we've actually seen NAND capital equipment sales drop the last 2 quarters as the market's gotten really good. And I think some of that is just timing, kind of random timing stuff, but there's also a clean room space that people are allocating to DRAM that's maybe taking away. Like what do you think is the supply picture here? And what does it take for you guys to resume kind of a higher level of spending?

Well, I mean -- so first of all, I think it's important to realize we are spending more money all the time to increase supply. So we've been very transparent. We're going to grow supply mid- to high teens. We're putting billions of dollars behind that. We're putting hundreds of millions of dollars of R&D behind building the best NAND technology in the world. So we are going to grow the market, and we grow the market every single day. And we think that, that is a good long-term growth rate for the market. And we're trying to sort out as this market changes, where is the attractive demand in that market. And I think that's happening every single day. It would take a lot to get more conviction. We're not interested in what the growth rate of the market is yet next year or the year after. We're interested in the growth rate of the market a sustained basis for the next decade. I don't think it should be lost on anybody that we just spent -- we just invested over $1 billion to make sure we have supply from 2030 to 2035. That's the kind of time horizon that we're thinking about, and we continue to invest more in the business to make sure we can continue to grow this business. I think as these business practices get worked out over the next year or so, maybe faster than that, we'll find out what the real growth rate is of the attractive demand in the market.

Okay. And I guess one of your push is really even predating the spin has been that customers need to give you more visibility into what their demand is, more transparency to the long-term view. This tightness in the market, does that help you achieve that? And I guess maybe we can go into some of the long-term agreements and things like that, that are giving you that visibility now.

I mean, certainly, that's what we need to get conviction behind. Again, we're investing to grow the market, and we've talked about that. We're investing billions of dollars in CapEx. We're investing hundreds of million dollars of R&D to build incredible technology. And clearly, our time horizon for those investments are quite long. I mean if you're going to build a fab, it's a pretty big place, and you need to get 10, 15 years of return out of that. And we're clearly looking for more visibility on demand. I mean the way the market has traditionally worked is the world's NAND is kind of auctioned every quarter. That's kind of how it works. And so I think that business practice is going to change to giving more visibility and commitment to a longer-time horizon of what demand is. I think this is the data center buyer coming in is more interested in certainty of supply, right? That's extremely important. And I think our interests are highly aligned. We're going to produce NAND. They need to consume NAND. The question is, how do we put a business model and incentive scheme around that where we both have confidence, we can get what we need. And the industry hasn't done a very good job of that in the past, quite frankly.

Yes. So as a long-term memory analyst, you hear about long-term supply agreements and kind of triggers memories of things that didn't work in the past and things like that. You've talked about the structure of these being different. You've talked about prepayments upfront, agreements that have teeth. You don't have to see your customers to collect down the road. How should we think about that? Are you seeing those commitments? Is that desire continuing to build?

I'll talk a little bit about this, and Luis can really -- he's very thoughtful on this as well. I mean -- look, I mean, there's no shortage of people that can tell us all the stuff that didn't work in the past, right? That is like everybody has got a lot of scar tissue. There's a lot of things that didn't work. We're very well informed in all of that. I think the most important thing in all of this is we have a willing partner. Again, the customer we're talking to has significant amount of demand in the future. They're planning for '27, '28, '29, what their demand is going to look like. They're willing to share that with us. And that's very unusual, right? We usually don't get that much insight. And -- but the product is so important to them, they want that certainty of supply. They're thinking that far down the road. How am I going to be able to acquire the amount of NAND I need? And when we look at the numbers, quite frankly, they're kind of eye-popping numbers. They're like if you're going to count on showing up every quarter, you're probably not going to get that much. So then the question becomes, so our interests are aligned. The question is, how do we align the incentives to make sure everybody can do what they need to do. And maybe you can talk some about that because you're like the tip of the spear on [ on appreciating all this. ]

I think one of the key things -- one of the first things we need to do is stop talking about LTAs and really talk about this as a business transformation because that's really what it is. And there are a few components with that. The first one is we don't want to be talking about the quarter. We're going to talk about 2 years, 3 years, 5 years, right? So the agreements are really longer term. We're both committed to it. The second element is how do we ensure financials are good. And financials need to be good in good times and in bad times. And if they're good -- if we're in good times, there is upside for us and there is an ability for us to capture it. And if we're in bad times, we're protected so that the financials, our gross margins, our operating margins, our free cash flow generations are all good. And then we want a business that's growing, right? We're selecting these customers where we believe that where we are making a commitment or they're making a commitment to us that the exabyte are going to be growing at a rate faster than the market. So we're choosing very carefully so that we choose winning customers. And the last component is what David was talking about, which is there has to be enough commitment from their side that they will go all the way to the end of the contract. Exactly how that's going to work, we'll come back to you as we finish our negotiations. It's a little bit too early, but we feel very good about where the discussions are going.

But you said it's not an LTA and you sort of said take-or-pays become unenforceable beyond a certain time frame. So there has to be some kind of cash prepayment, right?

I think -- well, I wouldn't necessarily go -- look, we're not going to go into the details of what we're negotiating. But I think Luis said it extraordinarily well. This is not about just an agreement. This is about what are the business practices of the way we do business with our customers. And look, our customers are -- all of our customers are extremely enviable companies, right? They have incredible businesses, and we're very lucky to have deep relationships with them. And I think our interests are highly aligned here. As I said, they want to consume a lot of NAND, and we're going to produce a lot of NAND. The issue is how do we get a model around that where we both have more assurance of what that's going to be. And our customers -- we don't expect them to be experts in our business. Our business is a very interesting business. You build a fab, and it's very expensive and you turn it on and wafers come out. And they come out every single day. They don't stop. They just keep coming out and there's more tomorrow than there was yesterday. So getting -- that's not necessarily the way they run their business, but that's the way we run our business. And so we got to get those gears to mesh. And the way we do that is through having a conversation and figuring out how do we transform the business practices of the way we're doing business with each other. And again, I think we have a set of customers that are very interested in having that conversation. Will we get to the finish line? Stay tuned, right? What will it look like? Stay tuned. We'll say more about it when we get there. But I think there's a willingness to get this figured out.

And with those kinds of big customer deals, is there a trade-off between capturing the best near-term price and getting that long-term visibility? Do you have to give them lower prices near term to get that [indiscernible]

Again, you're getting into the details. I mean you keep...

Sorry. Just tell me everything you got. Just us.

I think you can have confidence that we have pretty good insights into what pricing is and things like that and where it's going. And we're trying to do multiple things at the same time. I mean, clearly, we're trying to build a business that has very enviable economics, right? But we're also trying to sustain that business, right? We want to get off the rollercoaster going up and down. And that's again, everybody wants to tell you all about this all the time. Oh, it was just a matter of time before this is going to change and everything is going to go in a different direction. We don't happen to believe that. We believe that if you have willing partners and willing customers and everybody has the same objectives, you can put a business model around that and business practices that significantly address those concerns. And I think everybody is a winner in that. There's not a loser. And so we'll see. Like I think the environment is such that there's a possibility that this could get worked out, and it would be a significantly different business model than looking backwards. And I think it's a big risk right now in our business for our company to basically just project the past on the future. That's the simplest thing to do. And again, like, oh, this is the way it's always worked, so this is the way it must always work. That's not true. It can change. And there are a lot of things changing in the environment. The demand is changing. The technology is changing, the buyer is changing. Lots of things are changing in this whole mix of ingredients to come up with to be able to transform this business to something different.

Okay. I'm sorry if I'm asking questions.

No, no, keep asking away, right? Well, keep asking away.

Maybe we shift gears to talk about the technology a little bit. How happy are you with your enterprise SSD portfolio? And how does BiCS8 kind of help you going forward?

So we're happy with the portfolio. I think we've got the right portfolio at the right time. And that portfolio is more than just the enterprise SSD. It's the whole -- you got to get a whole bunch of things, right? BiCS8 is a tremendous node, right? Wafer bonding, all the technology that's there, the performance, the power efficiency. It's just a great fundamental technology. And in our business, if you don't have a great fundamental technology, you can't really make up for it when you build the controller, right? So having that great fundamental technology is very important. We have a 2-terabit die, right? That makes a big difference because when you build an enterprise SSD, you only have so much real estate to pack these things into, so you got to have the right die. And then on top of that, now we've had an opportunity to spend the last many years building a clean sheet controller and bringing all the expertise that we have inside Sandisk around our ability to build controllers for the consumer business, the ability to bring -- builds controllers for the client business. And all of those businesses we're in, we've taken all that expertise and applied it to building a brand-new controller for our storage class enterprise SSD. That product is in qualification. And that product and the receptiveness of that product is really what's unlocking these conversations that we talked about before, right? If you're going to spend 2 years or a year going through a qualification process, that's when you're going to start getting more insight into what's your demand is years in the future, and that's what kind of unlocks this conversation. So we feel very good about where we're at.

Yes. Okay. NVIDIA talked a little bit about a new key value cash technology that if you just take the number of bits they talk about and the units that we know they ship, it's like 8% of the world's NAND. It seems like that's getting phased in maybe over time, not something that's imminent. But do you have visibility into what that opportunity may look like? It seems consistent with people giving you very high 2028 forecast.

Yes. I mean this is why I love this business. It's like people are always thinking of new things to do with our product. It's fantastic. It's fantastic technology. Look, we've looked at this architecture. We've talked a bit about this. I mean I think the bigger picture is we've had confidence for some time that NAND is the most scalable semiconductor technology. And the AI architecture is about scaling. And so models are getting bigger, context lengths are getting bigger, caches are getting bigger. You're naturally going to start pulling in that very scalable NAND technology into that architecture. I think that's now very clear to everyone, and that's driving a lot of the business. And if you look at the announcement you're talking about, we just did some -- it's early, but in our early analysis of that is if you just look at some penetration rates per device and then you look at how fast you put some assumptions around how fast they're going to be sold and ramp, you do get to a number like 75 to 100 exabytes of incremental demand, which is great, right? That's a '27 number and you go up from there. So yes, there's lots of reasons to be optimistic about demand. But again, it's about, back to the previous conversation, signing up for that demand, not just doing a paper exercise and saying this demand is there. But I think that process is happening, right? And that's where I say we have a willing partner in this conversation on how do we get our interests aligned around producing and consuming NAND.

Okay. And then you also -- you talked about high-bandwidth flash around the time of the spin. And none of us really knew exactly what you were talking about, but it's like there's a fair amount of momentum building around that.

You're so nice about that.

The deficit is me, not you. I think that opportunity starting to take shape a little bit, the agreements you have. Can you just give us an overview of where we are?

Yes. I'll say, Joe, we were just chatting before we started. It's been an exciting year. We're like almost exactly 1 year from when we launched the company, and a lot has happened in that year. And if I look back on where we were a year ago and where we are today, I would say this is one of the things I'm personally happiest about. We had been working on this idea that NAND needs to play a larger role in the AI architecture, especially inference, where NAND brings an enormous amount of density. When you talk about something is 10x more than something else, that is a very good sign that you should spend time trying to figure out how to make that product fit into that architecture. And so we started working on that many years ago. And I think it was some insight by some very clever people in our organization, which is if you're a NAND designer, you've been told your whole life, figure out how to get me more bits. That's all I care about. I want more bits at the lowest cost possible. But now you need more bandwidth. For inference, you need more bandwidth. And so if you go back to that NAND designer and you say, hey -- okay -- now that -- we're not declaring victory on density. We still -- but we have a long road map on density, right? The technology is very productive. Why don't you start thinking about how to increase the bandwidth? And it turns out very clever people come up with ideas how to do that and enough ideas where this technology can move into the inference architecture. And that's what high-bandwidth flash is all about, right? Can you build a NAND product? Can you use NAND in that inference architecture? It's not training. We're not replacing HBM. We know that. That would be crazy. But a read-optimized, very deterministic. I know what I'm reading. I'm reading the same model over and over again, loading it into a processor. We believe that, that's a memory constrained problem, not a CPU or a GPU constrained problem. So we started working on that. And then we were going to launch the company, we had a long discussion about should we talk about this? It's very early. And we decided, hey, let's talk about it, right? Because we're asking people to invest in the company. And we think -- we like this idea, and we talked about it and everybody looked at us like, what are you guys talking about right? And maybe was meant with a little bit of skepticism. And one of the things I'm happiest about in this year that's transpired, and everything that's transpired is people are not -- they're not looking at it that way anymore, right? Others have come along. Other people in the industry are now saying, hey, storage needs to be redesigned for AI. They're not talking necessarily about high-bandwidth flash, but it's the same kind of concept. How do we bring this very scalable semiconductor technology to bear to help this architecture, which is phenomenal, scale better. And we're working on it. We're talking to potential customers, both on devices and cloud infrastructure of how they could use this. It's not a plug-compatible component for the current architecture. So you got to change the system a little bit to accommodate it. If you're going to do that, you got to know what use case you're trying to optimize around. So there's a lot of work that has to go on, and we're having those conversations, and we're optimistic that's going to lead to a good outcome. We're not ready to declare a victory. And at the same time, we're doing the work. To the end of this year, we'll have a die and NAND die, right? It's a derivative of BiCS8. It's not like we have to build a whole new node. And we're also working on the controller. And maybe a year from now, we'll have -- what we would call a device that we could put in customers' hands that they could start using. And then along the way, one of our peers decided they wanted to work with us on the standardization of this. I think that was a big step forward. It's hard to create a market all by yourself. And so one of our peers step forward, SK, and they wanted to standardize it together. And I think it was just last week that we announced that we're going to do that at OCP. Now we're not working on the NAND die. We're not working on the device together. We're just working on the specification of the system, so people would know how to use this. So I feel very good about how far we've come in a year. And we still have more work to do, and we're going to stay very focused on that. And we think that -- I think that original insight that this very scalable semiconductor technology has got something to add to the AI architecture and will be central to that architecture, I think, has proven to be a very good insight, and I think generally accepted now. And so there's different ways of doing that, primarily enterprise SSDs today and I think we're going to continue to see innovation in that space. And I think this is what's super exciting about the world right now. I think we have this awesome technology. We have this -- we've been working on this technology for 25 years. We have like 25 years of accumulated R&D. We have 25 years of accumulated CapEx investment on fabs that cost tens of billions of dollars. And here, we have this just incredible innovation -- in addition to all the other great markets we're in, which is like every possible device you can imagine, smartphones, PCs, tablets, just goes on and on and on. And now we have this entire new area of AI to innovate in and figure out how we can bring all of this intellectual property and all of this accumulated investment that we've done over 25 years to bear on one of the most exciting technology developments in a very long time, if ever.

Yes. That's very cool. So you don't worry about demand destruction in consumer markets. We've talked about this. I guess why not? If you see prices in a consumer solid-state drive double or triple, isn't the natural thing to cut the content in half?

So I would rather focus on. I think he and I would both rather focus on the incredible demand creation that's going on, right? And we're looking for the -- what we consider the attractive demand for our products, which is something that recognizes that value of all this accumulated intellectual property, the value of all this accumulated CapEx that we -- most technology businesses are either very IP-intensive or they're very CapEx-intensive, kind of one or the other. We're like both. And so this technology is extremely valuable and I think this is extremely exciting time because that value -- I think it's fair to say for a long time, that value wasn't really recognized. I think that's -- you just have to look back a year at our value of our stock and like people did not recognize that value.

Well, the income for the last 5 years hasn't been what you should have for assets like that.

So the good news of that is we have all this accumulated R&D. We have all this accumulated CapEx. It's like this gigantic coiled spring that's all uncoiling all at once. And it's uncoiling in this creation of incredibly attractive demand. So other demand that's not as attractive, again, we -- all of our customers are fantastic. But there's supply. There's demand. They're going to be rationalized through what is the price. And I think we're covering that the world is valuing this technology a lot higher than it has in the past.

Great. So we only have 4 minutes left. Let me see if we have questions from the audience. One in the front.

What would you need to see to supply to, let's say, a mid-20s bit growth rate? And how does that conversation with your JV partner go?

You want to take that one?

Yes. I mean, remember, we invest on a 10-year basis, right? That's our CapEx. And we need to see -- we need to have conviction that for the next 10 years, we're going to see that type of growth. And therefore, we're not there yet. We're very far and we're sticking to our plans, which is to invest to the mid- to high teens rate, and that's what we'll continue to do. We work very closely with Kioxia to define the investment plans, and we do that together for the JV portion of their spending, and it has worked very well for 25 years.

I guess just a follow up on that. Is there a hypothetical world where your LTAs necessitate CapEx given the prepayments?

So I think this is one of the things that the more people I talk to, I think, is a bit misunderstood about what we're doing. So we're not trying to create a world where we can spend more CapEx. What we're trying to do is create a world where people commit to buying what we know we're already going to produce. That's really what we're trying to do. And so we know we're going to produce in the future. In fact, we're going to produce more in the future than we do now. And we're committing billions of dollars to make that happen. And what we're looking to do is to work with our customers and our partners to align ourselves around how that supply is going to be consumed. And like I said, we have, I think, it's super interesting because I think the main buyer -- everybody in the market is extremely important. We have incredible enviable customers there, like some of the most fantastic technology companies ever across all domains, whether it's smartphones, PCs, gaming, data centers, all of it. We're just trying to get this business model better aligned of our investment horizons and their consumption horizons. That's what we're trying to do. And I think as we get that better aligned, we'll have a basis to answer this question. Right now, there's no basis to answer this question. Like my view is we're already investing mid- to high-teens bit growth. We're doing we're investing more to grow this industry. It's now about getting the consumption of that aligned to that same level of investment.

Maybe just to wrap up then. I mean the conventional wisdom for the last few years has been that this is a space that needs to consolidate. All of this goodness has happened without anything consolidating.

Isn't it amazing?

Yes. So it doesn't really seem to me that we do need that, but just what's your perspective?

I think any time you're in a high fixed cost industry like scale matters, but we get scale through the JV, right? So there's always benefits of if you can do that. I just don't think it's a precondition for the economics being incredible in this industry. I mean the way I look at it is there's like this entire spectrum of technology that the world needs. NAND is a key part of all of that, and there's only just a few companies in the world that can do that. And if you want to create a NAND company, best of luck. Find yourself an R&D team is to start with, that's really hard, and we'll see you in 10 years. So -- but the world doesn't need that, right? It's just like there's -- this is -- I mean this is why I chose to come to this industry a year ago, and I think a lot of people told me I was crazy. I found out that a lot of people put it in writing, they thought I was crazy because I think this is just a spectacular industry. I think it's a spectacular technology. I think all the elements are there. We just need to change the business model a little bit and everybody is a winner.

I didn't think you're crazy, but I probably underestimated WD a little bit.

You were probably one of the few people who didn't think I was crazy.

Anyway, look, guys, we're out of time. Thank you very much.

Thank you. Appreciate it.

Thank you, Joe.

Thank you, guys.