All right. We'd like to get started. Welcome, everyone. My name is Toshiya Hari. I cover the U.S. semiconductor space for Goldman Sachs. Very excited to have the team from Western Digital with us for this session. We have David Goeckeler, CEO. We also have Wissam Jabre, EVP and CFO. As always, I have a long list of questions, but would love to keep this interactive. I will pause for questions. Please have your questions ready. First of all, thank you for coming.
Well, thanks for having us.
Thank you for having us.
We appreciate it.
I think you need to read-
Oh, I need to-
I need to read the,
The most exciting part of the presentation.
So thank you. We will be making forward-looking statements in today's discussion based on management's current assumptions and expectations, including with respect to our product portfolio, business plans and performance, market trends and dynamics, and future financial results. These forward-looking statements are subject to risks and uncertainties. Please refer to our most recent financial report on Form 10-K and our filings with the SEC for more information on the risks and uncertainties that could cause actual results to differ materially from expectations. We'll also be making references to non-GAAP financials, and reconciliation of our GAAP and non-GAAP results can be found on our website.
Great. Thank you for that. Wanted to kick off high level. David, it's been a busy few years. You joined the company March 2020. We had the pandemic, ups and downs from a cycle perspective, both in NAND and HDD. I'd like to think the environment is a little more normal, whatever that means. What are some of the near-term and long-term goals you and your team are focused on now that perhaps you can be a little bit more proactive as opposed to reactive?
Yeah, it was an interesting time to start, but I think we've made an enormous amount of progress in the last four years. The goals are really the same. It starts. In our business, it starts with great products. Like, if you don't have great products, it's very hard to be in a technology business, and that's our first focus always. In our business, especially the way our markets work, you have to focus on the cost side of it and obviously the value equation you're delivering to customers. I think our technology portfolio is just in the best place it's been in a very, very long time. We have the best cost-controlled technology in both markets, whether it's NAND or HDD, and we can talk about that.
And then we've built products that give us a lot of access and deliver a very, very good value proposition to our customers. And we. On the flash side, we've spent a lot of time, you know, building a great client SSD portfolio. We're now, you know, much stronger in the enterprise SSD portfolio. We expect that to be stronger, especially as we go in the second half. We can talk more about that. Obviously, we have great consumer business, so I think that portfolio is just in really good shape. The HDD portfolio is in really good shape. We can talk more about that with our ePMR, Ultra SMR technology. Market is really resonating with that strategy. So that's always, like, the high-order goal in the organization: have the best products, best cost points, best value proposition to our customers.
Now, on top of that, we worked very hard to really put a lot of agility into the business, and that's a big focus. We're in a very dynamic business, to the point of your question. You know, we clearly didn't expect a global pandemic, but we also know we're in a cyclical industry, and our job, as we see it, is to get the best profitability out of the business, no matter where we are in the cycle, and deliver through-cycle profitability that is best-in-class. And I think we saw in the downturn, we were able to do that. We're seeing in the recovery, we're able to do that, and that's because we've built a great portfolio. We put a lot of agility into the business because we have to respond faster than the markets are responding around us.
If we don't, we'll just get carried away with wherever the market is going, and that's not our goal. So I think that strategy is what we've been driving for the last several years. You know, we can always do better, but we're going to stay focused on that as the core of the business. Obviously, on top of that, you know, we want to generate profits. We want to focus on our balance sheet. That's still a goal of ours as we emerge back into generating free cash flow out of the business, stay focused on improving our balance sheet, and that capital allocation strategy, and of course, drive to the separation by the end of the year. So that's kind of sums it up of how we're thinking about the business.
That's great. And I wanted to hit the business separation. I think you're targeting the back half of this year in terms of timing, but just curious where you are in the process. I'm sure you've had plenty of engagement with customers, partners, employees. Any new learnings as you've gone through the process or you continue to go through the process? And in terms of the new management team, I think, David, you're going to be the CEO of the flash side of the house. Irving, I believe, is taking over the HDD side of the house. How were those decisions made in terms of the leadership team?
All right. Let's-
A few parts, sorry.
Unpack that a little bit. That's so fine. That's, that's great. So, the separation process, you know, we announced this at the end of October, very committed to it. We have a large team inside the company focused on this. We're making a lot of very good progress. It's not, you know, tremendously, externally super interesting work, but we're doing things like creating legal entities around the world, so we have the right place to... You know, we're a global company. We can, you know, pay all of our employees and provide them benefits. We've got tens of thousands of contracts between customers, suppliers, partners... that we have to go through, and we have to sign those to the businesses, and in many cases, duplicate them and get customers and partners to agree to separate contracts.
So all that work is going on, organization structures, all those kinds of things are happening. We're getting to the point where we'll, we'll be able to to start focusing on maybe some of the more, externally interesting stuff, like the balance sheet and the capital structure of each business as we go through the second half of the year. But we've learned a lot. You know, we're not just spinning off a small division. We're actually almost completely splitting a major company in half and creating two separate publicly traded companies. So, it's a big effort. It's well underway, where, you know, we got work in front of us. We're targeting the end of the year for the separation, and we're making good, good progress on that. Now, management teams, you know, with a lot of conversation, first of all, about Irving.
I've known Irving quite some time. He joined Western Digital a little over two years now. He's led operations. He's just done a phenomenal job of really all this work we talk about, of we've, you know, optimized our cost structure, our manufacturing footprint. You know, really, Irving and Wissam have been leading all of that work and putting us just in a much better position to generate better profitability, and we're seeing that as we come out of the downturn. So Irving is a tremendous executive, very deep operations, very deep sales experience, global perspective, so he's gonna be a great leader of that business. How were the decisions made? You know, look, it's... For me, personally, it's hard to leave any one of these businesses behind. I have an enormous amount of conviction in both businesses.
I think all of the relationships that I've built around the NAND business, it's a more complicated business, kind of on a global geopolitical level. I think myself and the board thought that I was well-positioned to lead that business, and I'm super excited about doing that. So I think that, you know, the separation, a tremendous amount of focus. By the way, we're not losing focus on executing the business during this time. That's also a big part of the focus, but very good progress, and as I said, targeting at end of the year kind of time frame. We do have a little bit of incremental news on kind of who the CFO is gonna be, at least to one of the businesses, and maybe I'll let Wissam share that.
Thanks, David. So, I'm pleased to announce that I'll be the CFO for the hard drive business after the spin. Look, I like both businesses, Flash and HDD. I think they'll be great companies with great leaders. They both are great businesses. They've been performing very, very well. They're very well-positioned for great success. I'm very excited to be part of the hard drive business after the spin, to continue to add value, contribute to creating shareholder value and future success for the company.
And meanwhile, as David said, this is a bit ways out, so we still have a lot to do with respect to continuing to execute on the business, as we go through the cyclical recovery and continue to focus on the spin.
Congratulations.
Thank you.
There's no tension between you two guys or anything like that?
There'll never be any tension-
Very, very, very friendly.
I think this is a great outcome for everybody. Irving and Wissam are gonna be an awesome team.
That's great. Yeah. Congratulations.
Thank you.
David, you were kind enough to attend our September conference last year, and we talked a little bit about AI back then. At the time, you know, the compute space had taken off. I think networking, your peers in networking, had seen, you know, good demand. Memory was kind of moving. Storage, not so much. It's been eight months since then?
Mm-hmm.
Um, it-
Time flies.
Yeah, time flies. It feels like we're starting to see green shoots, whether it be Enterprise SSD or nearline business, we'll get into later, but both sides of the house, demand is recovering. The big question we get from investors is: Is this simply a inventory restocking cycle, or is it attached to AI and therefore sustainable? What's your view on that?
I think it's different in each business, and I'm optimistic about AI being a kind of secular tailwind for both businesses. I think it's showing up at different rates in each business, and there's, you know, there's some very good reasons for that. So first of all, in the HDD business, I think primarily what we're seeing right now is a cyclical recovery in ordering in the business. It's difficult to tell if, you know, some of that ordering is AI-driven. We just don't have that much visibility, but I think that we'll figure that out as the use cases get more deeply deployed into all of our daily lives, whether it's in the enterprise or what we do personally. But, you know, we believe that AI is just gonna generate more data in the future.
It also says that any data that you have is probably valuable in ways that you didn't think about before. So for me, it just means the incremental value of storing data is gonna go up, the ability to generate data is going to be automated in many ways, and, you know, the vast majority of that data is gonna be stored on HDDs. Right? All of the models that are being trained, all of that data lives out there predominantly on HDDs. Now, in the Flash business, I think there's been quite a bit of movement in the last eight months. It's very clear. It's becoming much, much clearer how AI is gonna impact the Flash business.
So first of all, there, you know, I think it's fairly well understood at this point, there's an emerging market for kind of the AI model training, data infrastructure business, and that uses enterprise SSDs, depending on the part of the training. You know, there's kind of four parts of the training or the AI life cycle. There's, like, all of the data that exists in the big data pool that you're gonna pull data from. That is kind of mainly HDDs. You've got then that information is pulled into a set of enterprise SSDs that are high-capacity drives, like 30 TB, 60 TB. I think customers would take even higher terabyte densities if they could get them to kind of prepare all of that data for the training cycle.
There's the actual training itself, where you tend to have enterprise SSDs that are more compute-focused, so maybe less density, but higher interface speed. So when we talk about these PCIe Gen 5 compute SSDs, that's really where it's showing up in that phase. And of course, then once you get the models built, you have the inference phase. And we're seeing these models now, people preparing for the models being deployed on the device, whether it's smartphone... Of course, there's a lot of discussion of the AI PC. We were talking about that out in the hallway a little bit. It's clear that there's going to be use cases that drive more demand for NAND on the endpoint.
We can't quite put our finger on exactly how much of the TAM is it, how fast is it growing, but we are definitely seeing that, especially in the enterprise SSD market right now, a pickup in demand as we go into the second half of the year for enterprise SSDs, that is, you can tie it very clearly to AI infrastructure being built.
Okay, that's great. We'll definitely come back to Enterprise SSDs. But before we go there, taking a step back, just on the broader NAND cycle, you know, it's clear that you're seeing a recovery in pricing, which is great. You're coming off the lows, though, and I think profit margins for you guys and for the broader industry are not quite there, right? Relative to what you would expect, on a through-cycle basis. I know you typically or you always avoid giving guidance beyond the current quarter, but as you think about the NAND cycle overall, pricing and things like that, into the second half, what are your current thoughts?
I think there's... you know, we're seeing a, you know, fairly rapid recovery, and I think that recovery kind of has three phases to it. We had a phase of the recovery where basically we're working through the inventory that all the suppliers had on their balance sheets, and I think we worked through that phase probably faster than everybody expected. And, you know, I think if you look at the results of different players, you saw, like, big step-ups in bit shipments on a sequential basis, and that's just, to me, that's just a sign that we were able to move that inventory quickly out into the market. That is followed by a phase of utilization of existing fabs, like returning to full utilization to supply the market. I think we got into that phase probably faster than most were expecting.
I think that's healthy. You know, I think there was a little bit of a, you know, maybe a little bit of discussion in the market about, you know, "Oh, my God, fab utilization is coming back. This is a sign. This is a bad sign." I think actually it's a good sign. It's as we move through the inventory phase more quickly, and now we're gonna need that supply out of the fabs. That phase is going to... You know, we're working our way through that phase, and then you're, you know, eventually you're gonna need more capacity, right? And then that's a much, much longer-term-
Right
... investment cycle to bring more capacity online. So as we look at the setup, we're moving through those three phases. I think we're moving through them faster than probably anybody thought we would, and I think we're gonna run up against this phase where, you know, it's gonna take more CapEx investment to bring more capacity online. And if you look at the amount of CapEx that's been invested in the business over the last several years, it's been severely depressed-
Mm-hmm
... which is not surprising, given profitability. So for our particular investment, we're gonna be looking. You know, we have a through-cycle model, to your point. We've just kind of are getting to the point of getting into that model-
Mm-hmm
... for the first time in many, many quarters. So we're gonna wanna see profitability above that model for a significant amount of time and get conviction that this is sustained demand before we're gonna be willing to invest more, in more capital in the business.
Okay, that makes sense. Focusing in on the demand side of the equation, in your NAND business, I think, you know, a couple of years ago or several years ago, bit growth expectations were in the thirties, or, you know, if you go-
Right
... way before that, 40s, 50s. I think today, you know, we're thinking teens, right? High teens, maybe 20%. You know, law of large numbers, some of the classic end markets, like phones and PCs, growth rates have come in a little bit. As you think about sort of the puts and takes, from a demand perspective, we talked about enterprise SSDs, but what are you sort of excited about, and how do you think about overall NAND bit growth, on a through-cycle basis?
Well, look, if we look at NAND, you know, we look at NAND demand this year, we're looking in the high teens. To your point, if we look at NAND demand next year, we're talking maybe low twenties. We look at NAND supply this year, we're talking high single digits. And then coming off of that number, we're gonna see a higher number next year. But we still see an undersupplied market for quite some time. On a through-cycle basis, you know, in the NAND business, we see in that twenties range of through-cycle demand. And I do think you're right, the investment thesis for how you think about the business is changing. I think in the 2D era of NAND, you could just roll nodes forward.
There was enough elasticity to continually absorb all of that supply that was coming on the market. I think one of the things that, that we learned in the downturn is you got to be much more careful. In the 3D era, the nodes are very productive. You can't roll forward a node and just put all that, all that supply out there, because technologically, you can do it. You gotta, you gotta be a little closer to what the demand equation is, and I think that's the adjustment that, that is happening right now.
Got it. And with the 20%+ bit growth expectation, is the 10%-12% revenue CAGR that you put up a couple of years ago at the Investor Day, is that still the right range?
Sure.
Or is that still the relevant range in your view?
Yeah.
Okay. SSD versus HDD, and I think you sort of answered the question before, but I'll ask it anyways. You're in a very unique position, at least before the separation, with both NAND and HDD-
Right
... under one roof. Based on input from your, your customers, do you see any risk to the, the nearline HDD business, or do you kind of see them as complementary technologies?
I think they're-
In the data center.
Yeah, I think they're highly complementary technologies. I mean, clearly, there's, like, a whole new use case that's been developed for Enterprise SSD, called AI model training, that we just talked about.
Mm-hmm.
Nobody used to do AI model training on HDDs, so it's not a cannibalization story; it's just a growth story for enterprise SSD. You know, we still see the vast majority of data being stored on HDDs for a very long period of time because we have a lot of conviction that we can drive the better. We can continue to drive TCO down in HDDs for a long time going forward. I mean, we're just talking about the industry is just now grappling with, like, jumping onto a whole new cost curve around HAMR. Does it happen this year, next year, the year after? It's sometime in that timeframe. You know, we think it happens in the 40 TB timeframe for us, is when it makes economic sense, given our roadmap.
But that's gonna allow us to drive the cost of storage on HDDs down, you know, commensurate with the levels you can drive cost downs on NAND. So both of these businesses are gonna continue to be growth businesses. You know, there's, they're both very dynamic markets. You know, we have a lot of conviction on the growth of both businesses, and there's not a cannibalization story happening here.
Got it. Very clear. Enterprise SSD market share goals. At the time of the Investor Day, I think you guys were hovering around 8%, and I think the medium-term or long-term goal was to double that to 16, give or take. Obviously, since then, you've had to go through the downturn. I think you had relatively high customer concentration. On the flip side, you did talk about multiple quals ongoing-
Right
... potential wins. Maybe talk about your aspirations and how we should be thinking about the next, you know, call it 12, 18 months.
Yeah, we made a lot of progress on Enterprise SSD just in time for the market to stop buying them. The timing was fine. You know, it's great to have the product done. Customers, it turns out, built an enormous amount of inventory. It's really the last market that's recovering from the inventory digestion. And we're seeing that as we go into the second half. Not only do we have this demand for the AI training infrastructure that we talked about on Enterprise SSD, we just have normal web scale enterprise SSD demand returning in the second half. That has been... You know, that market, like, capacity enterprise hard drive, is dormant for quite some time, and we're seeing that inventory correction end and more normal ordering happen as we move into the second half.
So that gives us a lot of conviction on the enterprise SSD market and the flash market in general. But we feel good about where the portfolio is. Those drives are still qualified. As that ordering comes back, the ordering of those, those drives will come back. We're qualifying new drives at new capacity points for these AI use cases we talked about, and we're just now launching a new product that's this PCIe Gen 5, compute-focused, high-bandwidth-type product into the market. It's being qualified at a major hyperscaler right now, and so we feel good about... You know, not only do we have the portfolio we had going into the downturn, we're now expanding it with new, new swim lanes, as we call it, or new use cases that we're able to address as part of that market.
As far as the share gain goals, like, look, we obviously wanna... What I talked about in the business is we want to build this dynamic business where we can allocate our supply to where we're gonna get the best return. Now, Enterprise SSD has traditionally been a place where pricing is good. It's a place where you want to participate. To the extent that continues to be the case, as we go through, you know, the recovery here in a more sustained Enterprise SSD market, if that's the place where it's the best home for our bits, that we can get the best profitability, that's where we're gonna put them. If we can get better profitability in a different market, we'll put them in a different market. So we want to be careful with the market share goals.
I know we were the ones who put the goal out there. We still stand by it.
Mm-hmm.
But what we're really going for is the best profitability at any given time in the market. And the whole portfolio strategy we've been executing over the last several years is give ourselves the most optionality possible of where are we gonna place our bets. And we start from a very good position because we have a great consumer business, and we have a great consumer brand in SanDisk. We've built a great consumer brand in gaming. We have the number 2 client SSD portfolio in the industry, and by the way, we think we're gonna lead the QLC transition in client. We just launched a new drive that's QLC-based, based on BiCS6, which is getting very, very good reviews. We have gaming, we have automotive, we have mobile, and now we have enterprise SSD.
The idea is to mix across all those markets to get the best profitability, no matter where we are in the cycle.
I know you don't typically talk about profitability by application, but do Enterprise SSDs stand out in either direction from a margin standpoint, or are they kind of in the center, if you will? To the extent you're willing to-
Yeah.
- to share.
Typically, don't talk about it down to that level, but generally speaking, they would be, well, they would be at or potentially accretive to-
Got it.
Where we're at.
Okay.
But remember, ultimately, on the flash side, pricing is mostly determined by supply and demand.
Right.
That's the biggest lever, typically-
Right
-for profitability.
Yep, of course.
So maybe one other way to answer that question, to give a little more insight into at least how I think the market works, is any part of the portfolio where it takes a lot of qualification, and you have to build something on top of the NAND, means it's most likely to have better pricing. One, because you have more non-memory component-
Mm-hmm.
and you're more likely to be using the NAND on the trailing part of the node.
Mm.
Like, the cheapest bits or the most cost-effective bits are the newest ones, 'cause they're on the newest node. But if once you get that NAND out of the fab, you gotta build a controller, and then you gotta go through a 9-month qualification cycle and all that kind of stuff, you find yourself, you know, a year or two down the road, on the technology you're using. And that, in some ways, is a natural limiter to the amount of NAND that's available for that market, and also the amount of investment required to actually build a product for that market. So it's more likely to be a better pricing environment. Doesn't necessarily mean it's the best profit.
Mm-hmm.
'Cause you have to invest a lot to get there.
Right.
So, you gotta balance all of that.
Right.
Each product requires a different amount of investment on our side, and it uses product that's a different part of the evolution of the nodes coming out of the fab.
Got it. That makes sense.
Just to make it more complicated.
Yeah. No, no, I appreciate that. So you talked about the three phases of the recovery, right? I think you said, getting sort of excess inventory off your balance sheets, you and your peers, take up utilization rates, and then the third phase would be you start thinking about, investing in tools. And you mentioned how, you know, that bar is really high, which makes sense. And I think to your point, based on what the tool companies have reported, CapEx across the industry is still very, very depressed, which is a good thing, I think. But as you think about, you know, 2025, 2026, longer term, at some point you will be investing.
What's sort of the, you know, what's the internal debate, as you think about the timing of, of sort of pulling that trigger, if you will? I know it's not zero one. It's a continuum, right? But-
It's not a switch on the wall.
Yeah.
To start.
But, yeah, like, what are some of the things that you take into consideration? What do you need to see for you to start spending on tools again, in an aggressive way?
Yeah, I mean, look, capital investment decisions are based on return on investment.
Yeah.
And so, for the short term, we'd like to see our business being more profitable. So we're aiming for a 35%-37% gross margin through cycle-
Mm-hmm
... which means, we still have a ways to go. That's really a through cycle measurement. So it needs to... I mean, margins have to exceed that quite a bit before we're able to see potential return on our capital investments. Until then, I would say, we're still going to continue with the status quo.
Got it. Thank you. My last question on the NAND side is just on industry consolidation, the competitive setup. Yourselves, your JV partner, Micron, the two companies over in Korea, you've got five very competent, credible players in the NAND market. Not you, I don't think, David, but a couple of your peers historically have said, "We need consolidation for margins to improve in a meaningful way." Curious to get your view, do you think the 35%-37% through cycle gross margin target is attainable given the current setup, or do you think you need some sort of, some form of consolidation, whether it be organic or inorganic?
No, I absolutely believe we can meet that through cycle target on the current way the way the industry is structured today. I mean, I think if you look from before the, maybe before the downturn, the market, the industry has actually changed quite a bit. I mean, there's been incremental consolidation with, obviously, what SK-
Mm-hmm
... did with Solidigm.
Yep.
I think that it's much—Let's just say it's much more difficult for an insurgent's company to start up in the NAND business than it was two years ago. But I think the industry, the setup, I like to talk about our business a lot more than I do the industry, but I think that, you know, I think from what I see, in our ability to use the JV to build tremendous technology. I mean, don't forget, we are able to invest in our core technology as if we are the largest provider in the industry. That is, that is a huge advantage. And in an industry where you have commodity-like pricing. You have to be the low-cost provider. If you're the low-cost provider, you're in a much better position.
If you look at the output of the JV over the last decade, we have been able to produce incremental supply at one-third less CapEx than the industry average, right? That tells you that our R&D is extremely good, and we're able to build a very, very high-quality product at the lowest cost, and that puts us in a great position. So, you know, what we need to do in our portfolio, there's two things we have to get right. We've got to get the fundamental technology right, because the fundamental NAND technology is in the wrong spot. If it's not cost-competitive, it doesn't have the right performance, it doesn't have the right quality, very hard to overcome that.
So you got to get that right, and because of the JV, we're able to invest in that at a level that is equal to or better than anybody in the industry. Then we need to pair that with a great portfolio. After those wafers come out of the fab, what are we going to do with them? Are we going to sell the wafers? Are we going to, you know, cut them up and put them into components and sell them into the mobile business to somebody else's controller? Are we going to build our own controller and a client SSD? Are we going to build an enterprise SSD? What are we going to do with that? And that part of the equation is what we have dramatically changed in the last four years.
We, like, built an entire team and put an organization structure in place that is able to make the absolute best decision for how do we spend our money to build the best portfolio in the industry. And like I said, we've got a really, really great starting point in that we have the SanDisk brand, and we have the global SanDisk franchise. And so we add on top of that, you know, what I said earlier before, number two client SSD portfolio in the industry, a great portfolio in gaming, a great portfolio in mobile. Now we have an ascendant portfolio in enterprise SSD.
We get those two things right, we have the right agility between them to make sure that we can be more agile, move faster than the market's moving around us, and I think that's what allows us to deliver the profitability goals we set.
Got it. Very clear. I'll pause here and see if anyone from the audience has questions. We'll just transition to hard drives. Okay. Similar questions, supply, demand. From a supply perspective, I think both you guys and your nearest competitor, you've taken capacity offline. Supply, demand is in a much better place. I think on the demand side of the equation, you've seen a shift in demand away from what I would consider as secularly declining businesses, and the growth parts are a significant portion of the market in your business today. So the setup is really good. How should we think about supply-demand being tight going forward? I know pricing has been a tailwind for you in the market.
How sustainable is that, and how should we think about your, your margin profile going forward?
So, look, I think you got the setup right, which is this is an industry. My observation coming to this industry four years ago, this is an industry that's gone through an incredible transformation, which is hard drives used to be a client business, where the industry generated hundreds of millions of units a year to supply that industry, and now it's transitioned to a cloud business, where you need 60 million units or something like that. The industry had all of this embedded infrastructure to build all of that capacity. You know, we don't call up an ODM to build our products. We have like - we call up our team in Thailand and Malaysia and all the other great places where we have a team.
So you had this, like, decline of client and the rise in the cloud, and this transition happening over this 15-year period. And that led to, quite frankly, hard drives were just persistently oversupplied in the market. There was just so much capacity available that was being reused to, like, fuel this cloud market. And I think the downturn we went through, the inventory correction was so severe, it really drew an end to this era of the market. And basically, for us, we just decided to remove an enormous amount of capacity from the system, as opposed to playing this substitution game over a long period of time. So here we are today, now demand is coming back.
We have fundamentally restructured our business and taken a lot of cost out of it, and that's led to now much better supply-demand balance in the market, at least for us. And when you have better supply-demand balance, you get better business practices. You know, you can't just show up in any given quarter and get hard drives whenever you want them. You actually have to tell us ahead of time. And, you know, we've gone to six-month lead time, and that's, you know, we may even extend that. And, you know, it's real because we don't have the capacity to produce what we had before, and the demand is now coming back. We're getting much better balance. As that becomes in balance, then we're also able to continue to deliver a great product. Where I started, we have a great product.
We have the best TCO in the industry, right? You can go to customers and say: If you deploy my new product, you're going to be able to operate your infrastructure in a more cost-effective way. That's exactly what they want, and it allows us to participate in the value creation that we're driving through our R&D. That's what supply-demand balance brings, and that's what we're able to do. So you're seeing as the demand comes back, because we've taken cost out of the system, we're more profitable at a lower volume level, and because we have supply-demand balance, we're able to basically drive pricing in, you know, the right direction. Instead of, like, pricing always going down, now pricing is, you know, flat to going up.
And by the way, it's still a great value proposition for our customers because we're delivering a better TCO equation as we go through that. So much, much healthier dynamics for the industry, and I think this is just the start of a much more healthier business for us. As long as we continue to deliver great products, right? Not a demand side issue. We're all pretty sure the cloud's going to continue to grow. Data is going to continue to grow. As we talked about earlier, data is more valuable than it's ever been. People figuring out new ways to, like, monetize stored data, we're the place where that's stored. We have a long roadmap of being able to deliver a tremendous value proposition to our customers, and I think we have a much healthier business.
The shift in sort of the way of conducting business, if you will, customers not showing up at the end of the quarter, expecting, you know, thousands and thousands of drives. Has that shift taken place across the board, or is it still kind of in motion? Do customers understand that things have changed, you know, in drives?
It happened very rapidly, quite frankly, because I think everybody... Look, we're not, you know, we're not inventing something here, right?
Yeah.
This is like every other technology that is part of, like, going into the same infrastructure that we're going into, operates as we're talking about. You can't just show up at the last minute and buy whatever you want at the beginning of the quarter. It's not quite that bad, but, you know, like, more planning, more visibility into what's going on. You deliver a better value equation. You expect to get paid for that R&D you invested to deliver that value equation. These are not unique concepts. This is what everybody else in the technology world does. We're just joining the party. So we're not plowing new ground here. We're just bringing these business practices that exist across the whole industry into the drive business.
Got it. Just for context, from, you know, the semiconductor process with their heads all the way through final test-
Yes.
How long is that today?
It's over a year.
Over a year? Okay. Yeah, that's pretty serious stuff.
It's pretty serious.
Yeah.
Yeah, it takes a long time to build a hard drive, and I think, you know, it's extraordinarily complicated technology.
Right.
I think the whole world is maybe finding out right now.
Right. Right. So as the world finds out how complex things are, want to go to margins again. As you know, Wall Street is very greedy. I think 12, 18 months ago, people were questioning if you could ever get to 31%-34%. Now, the question is: Why are you so conservative? How should we think about sort of the puts and takes? You're, you're in that range today. Based on everything you just said, it feels like any bias is to the upside. I certainly don't want to put words in your mouth, but-
No, I mean, look, we put out a model, and we hold ourselves accountable to that model, right? We work very hard to meet that model, and we've been below it for quite some time now. We've—we had a great quarter. The, you know, the products are performing extremely well. You know, maybe we'll talk about this a little bit. Our ePMR, UltraSMR drive is the thing that customers are standardizing on. It's clear. We now have the largest, fastest-growing, and most profitable drive business in the industry. It's hard to do all three of those things at once, right? The only way you can do all three of those at once is if you have great products, and your customers want them. So that's leading to great dynamics in the business. Like I said, supply-demand is more balanced.
We've now entered the range of 31-34. That's great. We're going to work as hard as we can to work through that range. You know, we hold ourselves accountable to that range. Am I going to change that today? No, we're not going to change that today. Am I satisfied with it? No, I'm not satisfied with it. Do I think it's the end of the road? Of course, it's not, but it's what we hold ourselves accountable to. Let's execute in that range, see how fast we move through it, and then we'll talk about what a new range could be.
Got it. That's great. You talked a little bit about HAMR previously. I think you guys have been super consistent in your views, in your commentary, as it pertains to HAMR and its insertion, and I think it's fair to say you've been very right on it, too, frankly. I know you're working on HAMR behind the scenes. What are some of the hurdles that at the industry level, you need to overcome?
I would say, just in general, for hard drives, this is very, very difficult technology. You know, this is not... Like, I'm a professional software developer, so if I say anything about software, I'm not trying to be offensive, but this is not writing software. Like, I know a ton about that, and it's not easy on its own, but this is like - this is serious, serious stuff. I mean, this is material science, this is physics. This is like trying to change the properties of how things work, applying heat and all this kind of stuff. Like, this is very difficult, and it takes a very long time for this technology to mature. And because you're, you know, you're literally, you know, at HAMR, you're using something to heat up the media to get it to change its properties, and then not just do that once.
Like, you can, you can, you can do anything once, but you need to get that to work for, like, five years. ... Put that in a data center and make it work for five years, and have high reliability and consistent performance the entire life cycle of that. That is a very hard problem, and it takes a long time for that technology to mature. And I think when our team looked at this many, many years ago, in fact, after our last earnings call, when I was talking about this, our CTO sent me a slide deck of when they made the decision to focus on UltraSMR and ePMR, and it was in 2013. I've never seen anything like this. A technology team made the call on when a technology was going to be ready a decade ahead of time.
They basically said: "Look, we think we're going to need another set of technologies to complement our HAMR roadmap, because when we get around this 30-terabyte time frame, we don't think it's going to be ready. Not going to have cooked long enough. We're not going to be able to get the reliability where we need it to be, all these kind of things I just said. And so we're going to develop these other technologies that give us more runway. Doesn't mean HAMR is not important. It just means it's our calculus, and it's not going to be ready. That turned out to be an incredibly important decision, and we built ePMR, then we built UltraSMR, and that's what the world's data center operators are now standardizing on, right?
They've looked at the same equation, they have very deep insight into the whole technology roadmap, and they've decided this is where we're going to go, and that's why you're seeing the results in the business. Now, I will say UltraSMR is independent of the recording technology, so once HAMR is there, UltraSMR is still going to provide that extra 20% of capacity. So getting customers to adopt that technology, which is what's happening now, and, you know, it's been happening over the last couple of years, that is a persistent advantage that will be there for years and years to come, even once we move from ePMR to HAMR. So look, I think that, as I said, we're able to deliver our customers...
You know, the R&D teams have done a tremendous job, the development teams, the operations teams, to operationalize all this technology, produce it at scale, build the right product at the right time that delivers the right value proposition for the customers. That's what we've developed. You know, we feel really good about where it is and our ability to continue to maintain that position going forward.
Right. And to your prior point, 40 TB is sort of the potential insertion point for you?
40, 40 TB is an economic issue, which is, if you're going to, you know, what I said earlier, you, in my opinion, in our markets, we've got to stay very, very focused on cost. So when you think about it from a product management perspective, you do not want to add cost to your product unless you get a benefit for it, right? Otherwise, you have margin compression. So for us, given our technology roadmap, you have to get 40 TB to justify the additional cost you're adding to the product to deliver HAMR, right? So at that point, the economics flip, and you get equivalent margins versus the technology we have today. It's not really a technology question, it's more of an economic question. And that's why we've been so focused that we're not going to release technology just to release technology, right?
Why, why would we add cost to our product if we don't need to? The whole idea is to deliver the right TCO, the right value proposition to the customers at the most advantageous cost to us. That's how all of our shareholders make money.
Yeah.
And so that's what we're very focused on. We've built a roadmap to deliver that, and we feel very good about how it's being accepted by the market.
Okay. We're out of time. Anything else that we should have touched on that we didn't have time for or any-
Anything from you?
Key messages from you guys?
Nothing in particular. I would say, the one thing is, we'll continue to focus on our balance sheet as we're generating cash now, and we will basically focus on free cash flow to pay down debt as the opportunity comes.
And then the only thing I'll add to that is, well, we're going to stay very focused on the separation. We believe that the company is still significantly mispriced in the market for the value we're delivering, and, you know, we think the separation is a big part of getting that recognized. We're going to continue to stay focused on executing the business over the next several quarters, get the best result we possibly can, and drive to the separation so we can get the true value creation out of this asset.
Great. Well said. Thank you so much.
Thank you very much.
Thank you.