WDC - NASDAQ

Good afternoon and thank you for standing by. Welcome to Western Digital's Fourth Quarter and Fiscal 2024 Conference Call. Presently, all participants are in listen-only mode. Later, we will conduct a question-and-answer session. [Operator Instructions] As a reminder, today's call is being recorded. I'd now like to turn the conference over to Mr. Peter Andrew, Vice President, Financial Planning and Analysis and Investor Relations. You may begin.

Thank you and good afternoon everyone. Joining me today are David Goeckeler, Chief Executive Officer; and Wissam Jabre, Chief Financial Officer. Before we begin, let me remind everyone that today's discussion contains forward-looking statements based upon management's current assumptions and expectations and as such, does include risks and uncertainties. These forward-looking statements include expectations for our product portfolio, our business plans and performance, the separation of our Flash and HDD businesses, ongoing market trends and our future financial results. We assume no obligation to update these statements. Please refer to our most recent financial report on Form 10-K and our 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 make references to non-GAAP financial measures today. Reconciliations between the non-GAAP and comparable GAAP financial measures are included in the press release and other materials that are being posted in the Investor Relations section of our website. With that, I'll now turn the call over to David for introductory remarks.

Thanks, Peter. Good afternoon, everyone and thank you for joining the call to discuss our fourth quarter and fiscal year 2024 performance. Western Digital delivered strong results with fiscal fourth quarter revenue of $3.8 billion, non-GAAP gross margin of 36.3% and non-GAAP earnings per share of $1.44. For the fiscal year 2024, revenue totaled $13 billion. We have maintained our strategic focus on aligning our portfolio of industry-leading products with growth opportunities across a broad range of end markets to mitigate volatility while structurally improving our through-cycle profitability for both Flash and HDD. Before we discuss our performance, I want to provide our views on where the storage market is heading and how we are well positioned to capitalize on these growth opportunities. Our diverse portfolio, coupled with the structural changes we have made to strengthen our operations, is enabling us to benefit from the broad recovery we are beginning to see across our end markets. In addition, the AI Data Cycle is increasing the need for storage and creating new demand drivers across both Flash and HDD. These AI systems process and analyze existing data, generate new data and require substantial storage for training, operating in a continuous cycle of increasing data consumption, processing and generation. As AI technologies advance, data storage systems must deliver the capacity and performance necessary to support the computational demands of large sophisticated models while managing vast volumes of data. Given this landscape, we expect Flash to benefit from both AI training and inference while HDD is poised to benefit at both the input and output stages of these AI models. To address the growing performance, power and capacity requirements, we have developed our Flash and HDD product road maps to meet our end customer storage needs across the entire AI Data Cycle. We introduced the industry-leading high-performance PCIe Gen 5 SSD to support AI training and inference. A high-capacity 64 terabyte SSD for optimizing the build-out of rapid AI data lakes in the world's highest capacity ePMR UltraSMR 32-terabyte hard drive for cost effective and deep content storage at scale. These new offerings demonstrate our continued commitment to innovation and market leadership. As we enter fiscal year 2025, we are well positioned to capture the long-term growth opportunities in data storage and believe the AI Data Cycle will be a significant incremental growth driver for the storage industry. Before I dive further into business updates, I want to update you on our separation plans. I am pleased with the progress our team has made as we continue to drive to completing the work required to separate the company at the end of the calendar year. As part of the ongoing preparation for the separation, we anticipate beginning to incur separation dis-synergy costs in the second half of the calendar year. Wissam will briefly discuss the anticipated impact of these costs. I'll now turn to business updates. Starting with Flash, the growth in revenue was driven by the recovery in cloud and a shift of our client mix to gaming and mobile, partially offset by a decline in consumer. Our focus on driving higher through-cycle profitability is reflected in our results as we proactively mix bits across our end markets. Our innovative offerings remain at the forefront of the market, reinforcing our competitive position and bolstering our growth prospects. For example, our new QLC-based client SSDs which grew 50% on a sequential exabyte basis, offer significantly better performance than our previous generation TLC products. Combining this high-performance node with our in-house controller development, enable us to provide a portfolio of client SSDs that deliver unmatched performance and value. We believe these products will lead the industry's transition to QLC flash. During our New Era of NAND webinar last month, we introduced the world's highest capacity BiCS8 2-terabyte QLC memory die, specifically designed to meet growing data center and AI storage needs. Built on a chip bonded to array architecture, BiCS8 reinforces Western Digital and Kioxia's leadership in cost and capital efficiency as well as superior I/O performance by integrating wafer bonding in advanced 3D manufacturing to establish a ground-breaking foundation for future scalability of 3D NAND. In addition, our 64 terabyte enterprise SSD is now being sampled with plans for volume shipment later this calendar year. Furthermore, our PCIe Gen 5 based enterprise SSD delivers best-in-class read performance as well as power efficiency. We are seeing significant interest in this product which is currently qualifying at a hyperscaler with ramp expected in the second half of this calendar year. We'll talk more about our Flash road map at the upcoming Flash Memory Summit in early August. Turning to the Flash outlook. Throughout the fourth quarter, our product mix was dynamic as we proactively mix bits between our end markets in response to the softness we are seeing in the more transactional markets such as consumer and channel. Our success in identifying the most profitable approach to allocating bits is reflected in the growth of both our revenue and gross margin. As we look into the first quarter, in addition to the mix environment we saw in the fourth quarter, we expect the continued ramp of our new enterprise SSD offerings and seasonal strength in mobile to drive mid- to high-teens bit growth on a sequential basis. For the full fiscal year 2025, we expect enterprise SSDs to represent a double-digit percent share in our portfolio mix. The new era of NAND is driving a period of change and we are going to remain disciplined in managing our capital spending. The layers focus race is behind us. The emphasis is now shifting towards strategically timing the economic introduction of new longer-lasting nodes. Innovation now means enhancing power efficiency, performance and capacity within these nodes, while capital decisions increasingly prioritize opportunities for margin expansion and revenue growth. Turning to HDD; revenue growth was driven by strength in nearline demand and improved pricing. By leveraging our SMR leadership and lean cost structure, we have surpassed our target gross margin range, underscoring our ongoing commitment to improve future profitability. The HDD business has undergone a remarkable transformation in recent quarters, marked by strategic initiatives aimed at introducing the most innovative, high-capacity products to market. We have increased our profitability meaningfully by restructuring our manufacturing footprint and optimizing our cost structure to drive operational efficiency. All while qualifying and ramping our SMR technology. We continue to structurally change the way we are operating our HDD business. With better visibility into future demand, operational excellence and a commitment to sustaining supply-demand balance, we are poised to continue our trajectory of bringing highly innovative products to market while increasing profitability into the future. On the technology front, we ship samples of our 32-terabyte UltraSMR/ePMR nearline hard drives to select customers. These drives feature advanced triple-stage actuators and OptiNAND technology which are designed for seamless qualification, integration and deployment in hyperscale cloud and enterprise data centers while maintaining exceptional reliability. With this in mind, we are well positioned to deliver the industry's highest capacity hard drives and the best TCO. Turning to the HDD outlook. As we look to the fiscal first quarter, we expect further growth driven by greater demand and more favorable pricing. Our cloud customers continue to transition to SMR and we anticipate a third major cloud vendor to begin the ramp of adopting SMR in the fiscal first quarter. Our leading products and lean cost structure have supported ongoing profitability improvements in our HDD business. We remain focused on driving higher margins to reflect the significant innovation and TCO improvements we deliver to our customers. Our strategic approach to commercializing ePMR, OptiNAND and UltraSMR technologies has proven to be the winning strategy, enabling us to surpass our gross margin target for HDDs in the midst of AI's emergence as another pivotal growth driver for the industry. Let me now turn the call over to Wissam, who will discuss our fiscal fourth quarter results.

Thanks, David. And good afternoon, everyone. In the fiscal fourth quarter, Western Digital delivered great results with gross margin and earnings per share exceeding the high end of the guidance range. Total revenue for the quarter was $3.8 billion, up 9% sequentially and 41% year-over-year. Non-GAAP earnings per share was $1.44. Looking at end markets, Cloud represented 50% of total revenue at $1.9 billion. The sequential growth of 21% is attributed to higher nearline shipments and pricing in HDD, coupled with increased bit shipments and pricing in enterprise SSD. The 89% year-over-year increase was due to higher shipments and price per unit in nearline HDDs along with higher enterprise SSD bit shipments. Nearline bit shipments were at a record level of 125 exabytes, up 16% from the previous quarter and 113% compared to fiscal fourth quarter of 2023. Client represented 32% of total revenue at $1.2 billion. The sequential increase of 3% was due to the increase in flash ASPs, offsetting a decline in flash bit shipments while HDD revenue decreased slightly. The 16% year-over-year growth was driven by higher flash ASPs. Consumer represented 18% of total revenue at $0.7 billion. Sequentially, the 7% decrease was due to lower flash and HDD bit shipments, partially offset by higher ASPs in both flash and HDD. The 5% year-over-year increase was driven by improved flash ASPs and bit shipments. For fiscal year 2024, revenue was $13 billion, up 6% from fiscal year 2023. Non-GAAP gross margin increased 7.1 percentage points to 22.8% and non-GAAP operating margin increased 8.7 percentage points to 3.9%. Non-GAAP loss per share was $0.20. Looking at end markets for fiscal year 2024, Cloud revenue increased 2% year-over-year due to higher demand for capacity enterprise HDDs and improved pricing. For the year, client and consumer revenue grew by 7% and 9%, respectively, due to higher flash bit shipments. Turning now to revenue by segment. In the fiscal fourth quarter, Flash revenue was $1.8 billion, up 3% sequentially and 28% year-over-year. Compared to last year -- last quarter, Flash ASPs were up 14% on a blended basis and 11% on a like-for-like basis. Bit shipments decreased 7% sequentially and 3% compared to last year as we proactively mixed Flash bits to maximize profitability. HDD revenue was $2 billion, up 14% from last quarter as exabyte shipments increased 12% and average price per unit increased 12% to $163. Compared to the fiscal fourth quarter of 2023, HDD revenue grew 55%, while total exabyte shipments and average price per unit were up 72% and 64%, respectively. Moving to the rest of the income statement. Please note, my comments will be related to non-GAAP results unless stated otherwise. Our focus on improving through-cycle profitability in both Flash and HDD has shown great progress. In the fiscal fourth quarter, total gross margin reached 36.3% well above the guidance range. Gross margin improved by 7 percentage points sequentially and 32.4 percentage points year-on-year due to better pricing and cost reduction as well as higher volume. Within Flash, by proactively allocating bits between end markets and executing on our cost reduction initiatives, we have improved gross margin for 4 consecutive quarters. Flash gross margin was 36.5%, up 9.1 percentage points compared to last quarter and 48.4 percentage points year-over-year. In HDD, by offering a leading product portfolio and running efficient manufacturing operations focused on cost discipline, we continue to make progress in improving profitability. We delivered a gross margin of 36.1%, exceeding the long-term target range, up 5 percentage points sequentially and 15.4 percentage points compared to fiscal fourth quarter of 2023. Operating expenses were $700 million for the quarter, above our guided range, primarily due to higher variable compensation associated with better-than-expected profitability. Operating income was $666 million, tax expense was $17 million, earnings per share was $1.44. Operating cash flow was $366 million and free cash flow was $282 million. Cash capital expenditures which include the purchase of property, plant and equipment and activity related to Flash joint ventures on the cash flow statement represented a cash outflow of $84 million. For fiscal year 2024, cash capital expenditures were $244 million or 1.9% of revenue, excluding the proceeds from the sale leaseback of our Milpitas facility. Year-over-year, this represented a 69% decline. Fourth quarter inventory was up from the prior quarter at $3.3 billion. With days of inventory increasing from 119 days to 126 days. A decline in HDD inventory was more than offset by an increase in Flash inventory. Gross debt outstanding was $7.5 billion at the end of the fiscal fourth quarter. Cash and cash equivalents were $1.9 billion and total liquidity was $4.1 billion including undrawn revolver capacity of $2.2 billion. During the quarter, we paid down the remaining $300 million of the delayed draw term loan. I'll now turn to the fiscal first quarter non-GAAP guidance. We anticipate both Flash and HDD revenue and gross margin to improve on a sequential basis as we continue to drive improvements in profitability across our businesses. We anticipate revenue to be in the range of $4 billion to $4.2 billion. Gross margin is expected to be between 37% and 39%. We expect operating expenses to increase slightly to a range of $695 million to $715 million. A decrease in variable compensation will be offset by the synergy costs as we continue to make progress executing on the separation plans. Included in this range are dis-synergy costs of $15 million to $25 million. We expect the synergy costs in the fiscal second quarter to be between $35 million and $45 million. Interest and expenses are anticipated to be approximately $110 million. Tax rate is expected to be between 15% and 17%. We expect earnings per share of $1.55 to $1.85 based on approximately 360 million shares outstanding. As shown in our guidance, we remain committed to driving higher profitability while maintaining focus on cost and capital discipline. I'll now turn the call back over to David.

Thanks, Wissam. Today's results and guidance underscore Western Digital's strong execution with promising growth opportunities ahead. We delivered solid results as we doubled down on our strategic initiatives and product road map, capitalizing on robust growth prospects for both Flash and HDD. The emergence of the AI Data Cycle marks an incredibly exciting transformation in our industry, driving fundamental shifts across our end markets. Looking ahead to fiscal year '25 we are well positioned to leverage our leadership positions to spearhead innovative technologies and deliver unparalleled value for our customers. Peter, let's begin the Q&A.

[Operator Instructions] And today's first question comes from C.J. Muse with Cantor Fitzgerald.

I guess I would like to focus my question around HDD gross margins, truly impressive outlook in the actual results. I was hoping you could speak to your contracted supply today, your pricing visibility into the second half of the calendar year and whether that extends into calendar '25. And as part of that, how we should think about the progression of HDD gross margins as we go into September, December and beyond.

Yes. Sure. Thanks, C.J. So yes, we were very happy with where the HDD business is at. We've -- we spent a long time focused on getting our manufacturing footprint in the right spot and a lot of cost optimization there now that the volume ramps back, it puts us in a strong position plus the products, right? The continued strong adoption of UltraSMR, those drives are great for us. They're great for our customers. They're leading capacity points. So when you put it all together, we were able to deliver past our gross margin target in that business. Some of the historians around here tell me it's like the best gross margin ever. And the good news, as you said, is it's going higher. So we got good supply-demand balance that's giving us good visibility throughout the rest of the calendar year, we're -- we pretty much know where every drive is going to go. At this point, we made, I think, a really significant transition this past quarter and that we moved up our request to our customers to give us visibility 52 weeks ahead, so a 52-week lead time on HDDs. Now the reason that's so important is the cycle time underneath to build an HDD is about 50 weeks. So we've been working very hard to drive the technology, drive our manufacturing but also change the business practices of this business to give us the visibility that allows us to really align our investment with our customer demand. Our customers have responded well to that request and we now have visibility for the entire fiscal year from a number of our biggest customers. Others are still -- we're still working through it with them. But to wrap it all up, we're -- we're very happy with where the margins are. We continue to see margins going higher in this business as we continue to innovate. If we continue to innovate, deliver better TCO to our customers, we're going to continue to drive margins higher and we've got a very robust road map to do that. And we're now getting the visibility in place to -- in the business practices to make sure we can keep supply and demand balance going far into the future. So hopefully, that addresses your question.

And our next question today comes from Joe Moore at Morgan Stanley.

I wanted to ask about NAND and this kind of -- you have weaker volume in Q2 but good pricing. And we're seeing that everywhere that the volumes are sort of disappointing but the pricing has still been good. And usually, you sort of think about needing good demand for pricing to go up. So can you just talk about that dynamic? And what do you think happens over the back half of the year with supply and demand for NAND.

As you know, Joe, it's a very dynamic market. So we're still seeing -- in the negotiated markets, we're still seeing good pricing increases, especially enterprise SSD and that's something we're very excited about. We see -- we saw really good sequential growth in enterprise SSD and we're expecting very good sequential growth in the next quarter. Some of the more transactional markets where you have more players involved and frankly, some players that don't produce their own raw NAND, pricing is a little more dynamic and demand in general in consumer is a little bit weak. So in those markets, we see a little more demand headwind, a little more -- not quite as aggressive as pricing moves but in the negotiated markets, we still see good pricing. And then, you put mix all on top of that to come up with a quarter-by-quarter number where you got quite a bit of variability. But as we go through the back half of the year, we continue to see strong demand in Enterprise SSD. We see the PC market really kind of maybe a little bit of an inventory rebalancing here going on in the next quarter but we see that improving as we go throughout the year. I think that's a normal process given a couple of really strong quarters here in the first part of the recovery of NAND pricing. So we still see demand outstripping supply through the second half of the year. And quite frankly, our modeling shows throughout next calendar year as well.

And our next question today comes from Tim Arcuri with UBS.

This is Grant Joslin on for Tim Arcuri. I was hoping you could talk a little bit about how you view the balance sheet and how you want to prioritize the free cash flow generated between now and the separation? And any updated thoughts on the appropriate leverage for each of the 2 segments?

Yes, sure. As I noted in the prepared remarks, in Q4, we paid down the remaining $300 million of the delayed draw term loan. And so as we generate cash, our focus is to continue to strengthen the balance sheet. The -- with respect to the second part of the question related to question related to the capital structure. Obviously, we're making good progress as we prepare for the separation, we will be as -- we get closer to the time of the separation, will be hosting Capital Market Days or Investor Days to discuss that in more detail. But as we've always said, our aim is to create 2 world-class companies that are competitive in their own spaces. And so their cap structures would be reflecting their -- the profile from a profitability, cash generation and market dynamics as well.

And our next question today comes from Aaron Rakers with Wells Fargo.

Yes. One clarification I just want to throw out there. Is the dis-synergy costs all in operating expenses? Or is there anything in COGS? And then from a question perspective. When I look at the NAND business and I think you guys talked about in the prepared remarks of growing kind of mid- to high teens sequentially on a capacity ship basis this next quarter. I guess where my math leads me is to think that your underlying assumption would be maybe more of a flattish ASP trend in the September quarter. I guess my question is, is that kind of how you're thinking about that? And if so, why would we think that pricing would be flattish? Is there a mix dynamic, something that we should consider in that?

I'll take the second part of the question. Kind of remember the first part now.

Let me take the first part related to the dis-synergy costs. The question was whether they were split -- they were all in operating expenses or there's anything in COGS. So Aaron, the numbers that I quoted, the $15 million to $25 million are all in operating expenses. There will be some dis-synergy costs in the COGS but they're really not material. And so that's why we didn't call them out. And they're reflected also in our guidance for Q1.

Yes. And Aaron, on the NAND pricing, I think you kind of answered your question as you went through there, just to confirm it. There's a huge mix dynamic. So we've got -- like everybody else, we're seeing some weakness in the consumer. And so that's -- consumer is 1/3 of our Flash portfolio. So that's a big business and that has very we've always talked about higher than average through-cycle margins. So that's a bit of a headwind. And then we -- it's a seasonally strong quarter for mobile. So that's -- there's a big mix dynamic there as well. So that -- but you're thinking about it the right way.

And our next question today comes from Karl Ackerman with BNP Paribas.

One clarification question and a question, if I may. For my clarification question, is the PCIe Gen 5 Enterprise SSD at a hyperscaler separate or in addition to your sampling of 64 terabyte SSDs in the second half of this calendar year? I guess as you address that question, could you speak to the breadth of customer design wins you have on enterprise SSDs which are indicative of your growing share of SSD relative to your overall portfolio mix.

Yes. They are 2 separate products targeted to different parts of the AI Data Cycle. So the compute product that you mentioned, what we call a compute product because it's designed to feed data into GPUs and keep the pipeline full. So very high performance, very high-performance NAND. That product is a qualification in a hyperscaler and we're getting a lot of interest in that product across the AI ecosystem of anybody that's building an AI training infrastructure because we believe that product is the best read performance and really good power efficiency and that's what everybody is looking for in that market. The 64 terabyte drive is a separate product that's targeted at building these high-performance data lakes for AI training and we're looking at that across the kind of same sort of customers. Anybody that's building an AI training infrastructure is looking for those kind of products. And that's in addition to the products we had qualified enterprise SSD, we had qualified 2 or 3 of the hyperscalers before the downturn and those products are picking back up now as well. So we have a much -- we were -- we were happy with the portfolio and now we keep adding to it. And those products are really well targeted to where the demand is in enterprise SSD which is a lot of pull on people building AI training infrastructure.

Our next question today comes from Krish Sankar with TD Cowen.

This is Eddy [ph] for Krish. Two-part question, if I may. On nearline HDD congrats on great progress there. This is the second quarter of above 50% market share for you guys versus historically your share being in the low 40s. I wonder if you think this has anything to do with your main peer transitioning from PMR to HAMR? Or do you think this is more tied to specific customer exposure? And also, I wonder if you think you'd have that capacity within fiscal '25, given that your exabyte shipments are at all-time high.

We know -- we think that the share in our products is because they are great products, right? We're delivering the highest TCO products. We've had this road map for a long time. None of our customers are surprised by our road map. They know about our road map, quite frankly, years in advance and we work with them very closely. So we believe we're providing the highest capacity -- well, we are providing the highest capacity drives at scale. They're easily qualified. The architecture is well known to our customers. and they deliver absolutely the best TCO. And if you're building a data center architecture at the scale our customers do, TCO is incredibly important. And now we've delivered a 32-terabyte drive. First time anybody has crossed the 30 terabyte at scale, that's sampling really great TCO. And then the UltraSMR side of it, these drives can deliver 20% more capacity with UltraSMR versus a typical SMR drive which is 10% more capacity that's great for our customers. They get more capacity per unit. It's great for us because we get 20% more capacity or the incremental 10% without adding any COGS to the system, so that's margin accretive. That's very nice. And we're seeing now the third hyperscaler starting to adopt our UltraSMR technology starting in the second half of the year. So we just feel really good where the portfolio is. We think this portfolio has been in development for decades to get to this point. These were decisions that were made a long time ago given the lead times in this business. And so we believe we're -- customers are just responding to great products. And the TCO increases in the supply-demand balance allow us to also work on the pricing side of it as well. So as far as capacity, look, I mentioned it earlier, we got to this 52-week lead time that was really important because now we're starting to get forecast from our customers a year out. It gives us the ability to plan for capacity. So as we get -- as those forecasts come in and we roll them up and we work with our customers and we look at that, what that picture looks like, then we'll be in a position to address this issue of do we add capacity. But we don't want -- we're not going to add capacity for a 2-quarter increase in volume. We think this is much more sustained than that. As I talked about, AI is going to be a big driver -- the AI Data Cycle will be a big driver for HDDs as well. We see sustained demand into the future. But we want to see the commitments from our customers and them to give us that visibility and we now have that in place. And as we get the whole market, all of our customers looking at that over the next quarter or so, we'll have a better I think a better way to answer that question of do we plan to add capacity. But for right now, we're happy with where our capacity is. and what we're able to ship and mix keeps going up and we're able to ship more exabytes per quarter based on mix.

Our next question today comes from Amit Daryanani with Evercore.

I guess maybe just focusing on the HDD side again. I think the fear everyone is going to have is, are you sitting at close to peak performance at this point, both from how much exabyte you are shipping versus the last week which I think within 10 exabytes of that? And where the gross margins are. So maybe if you can just focus on gross margins up. Can you just talk about what is the right way to think about HDD gross margins assuming exabytes keep growing and your conviction that pricing can remain favorable on the HDD side or through the back of this year and potentially into '25?

Yes. I think the whole key to that question is to continue to innovate. It's just as simple as that. If we continue to innovate and build a better product and deliver a better TCO to our customers, we're able to continue to participate in the value of that R&D that we're spending. We just announced a new drive that's 32 terabytes which is 13% more capacity than next biggest drive on the market which is our 28-terabyte drive. So as long as we continue to do that, we go to our customers and they get a better value proposition and we're able to participate in that. And as long as we continue to drive the road map forward and I have a tremendous amount of conviction that's the case, we're going to do that. We will continue to drive margins higher in this business.

Maybe just help me clarify this. Is the performance on the HDD side, more a reflection of your peer has not had HAMR come out yet. And once that comes out, perhaps market share and gross margins normalize, I guess, the feel would be, is this all durable? Or does it go away from a margin perspective once Seagate has HAMR in volume production with the hyperscalers.

No. Like I said, I don't think it has -- it has to do of continuing to deliver greater TCO. The underlying technology that you're delivering on the product I don't think matters that much, right? It's like customers want more density in the same slot and we're able to deliver that. We have a road map to deliver that. Other people in the industry have a road map to deliver that and we're going to continue to drive TCO down. We get better visibility into ordering. I've talked a lot about, I think, hard drives have been an industry that's been persistently oversupplied given the client to cloud transition. We're past all of that now. And I think we're seeing what the dynamics of the industry are like when we get the supply-demand balance right. We get our manufacturing footprint the way we want it. We deliver world-class leading products that provide enormous amounts of value to our customers. And we're -- that drives margin higher because it's a great value proposition for our customers. It's a good deal for them. It's a good deal for us. And I think that is incredibly durable.

And our next question comes from Vijay Rakesh with Mizuho.

Just -- I mean, obviously, very solid execution on the gross margin side, I think, is the highest in 5 years. But as you look at the road map forward, can you level set where you think hard disk drive and NAND margins could get to? Because I think in the past, obviously, NAND margins have gone like 45%, 50%. And I think we see hard disk drive margins at 38%, 39%. Could you help us like how to walk to the margin ARPU going out?

I think in the -- look, so next quarter, margins in both business are going up which is what we want. And I think as long as we continue -- in the NAND business, we -- again, we have to continue to innovate and bring great products to market and I think our portfolio is as strong as it's ever been. And we've always -- we've been strong for a long time in consumer. I mean that's been a stalwart of the business for a very, very long time. We have a very strong client portfolio. We have a very strong gaming portfolio. We've always been good in mobile and now we have enterprise SSD, where we're building out the portfolio right at the time where the AI Data Cycle is really driving a significant amount of demand. So I think from a technology perspective and where the portfolio is positioned, we're in very, very good shape. At that point, we look at long-term supply-demand balance of the business. We all know it's a cyclical business. You have to add CapEx in very large increments to bring on capacity, nodal transitions give you more capacity. We talked about this in our new era of NAND. I think this idea that the layer race and let's just put bigger and bigger nodes on the market and count on elasticity to soak up all the bits. Like I think that era of NAND is in the past, the era of NAND in the future is more strategic introduction of new nodes that add capacity, more innovation within the node on performance, power, capacity and so that, I think, is something that says the way we're going to manage our business is different than it has been in the past. And then, we look at just the overall supply-demand balance. And there's been not a lot of CapEx spending in NAND coming out of the downturn. And so we see a good supply-demand balance. So we expect to see continued positive momentum in the portfolio. On HDD, this is a business that is, like I said in the prepared remarks, we think it's gone through a remarkable transformation over the last year, really. We're starting to see the -- we're starting to see the results of this transformation that's been going on, quite frankly, for a long time. One is about getting through this client to cloud transition. Industry has been in that position for 15 years probably. So we're kind of through that now. Client is just a much, much smaller percent of the portfolio. Even if you look back to the past peak of our business, client was a much bigger percent just 3, 4, 5 years ago than it is today. So it's a cloud business now. We continue to deliver really strong TCO gains with our technology road map. And I think the business has a long road map on continuing to be able to drive better TCO for our customers. And then on the demand side, you've got -- again, you've got AI now the AI Data Cycle where HDD is going to be a big part of that on the input side, all of this data is stored on HDDs. And on the output side, people producing a lot more content to be stored on HDD. So we see a good demand environment, good business practices of getting the right visibility into how we should add capacity, so we don't end up in an oversupplied market. So I think the HDD business has structurally changed. I see the margins going higher and going higher in a structural way that will be maintained at a higher point. It's not a value proposition question. The value is really there. And like I said, we continue to drive TCO lower. We're going to continue to drive margins higher.

And our next question today comes from Wamsi Mohan with Bank of America.

I was wondering maybe you could square your comments a little bit on proactively mixing bits across end markets in a quarter where your bit shipment declined below what you had originally expected. But at the same time, you also had inventory build in the quarter. And as a clarification from a prior question, could you just help us think through in the September quarter, what is the like-for-like ASP trajectory for NAND as well.

Okay. So first of all, the mix, I think, the first one was the mix question. So was the question in the quarter?

Yes, in the fourth quarter, it's how do we proactively -- how we proactively mix the bits to generate better profitability.

When bits were declining. Okay. So yes. I mean, look, I mean, I guess, in a big picture, we're always just looking at every market that we're in and what demand is on a week-over-week basis and what our customers are telling us and we're trying to put the bits to where we're going to get the highest return. We saw some headwinds in consumer. So we mixed into other parts of client business. And we also saw really good growth in enterprise SSD. I think we saw 60% sequential growth in enterprise SSD. So that provided a floor on kind of how we think about the mix side of it. And the second part of the question?

Yes, the -- so on the -- maybe on the comments on the inventory, Wamsi, it's not unusual for exiting the June quarter for us to have inventory builds as we get prepared for the second half that tends to be more consumer oriented and sort of there's more shipments that typically take place. And so we're comfortable with that. Yes. So on the like-for-like for the September quarter, we're expecting the ASPs in NAND to be slightly up in the sort of low single-digit percentage range.

And our next question today comes from -- excuse me, Toshiya Hari with Goldman Sachs.

I had one quick clarification and one question. On the clarification, the dis-synergies, we saw the $35 million to $45 million you guided to for the December quarter, is it fair to assume that's kind of a steady state for the combined business going forward post separation? Or do those costs go away? Or do those costs potentially go up beyond separation. That's my clarification question. And then my question for Dave on the NAND side, -- are your fabs fully loaded at this point? I know at the trough of the cycle, you had cut pretty significantly along with your peers. But where are utilization rates today? And how should we think about fiscal '25 CapEx?

Yes. Let me -- thanks, Toshiya. Let me take the first part of the question. So with respect to the synergies costs that I discussed for the December quarter, they range in and you noted $35 million to $45 million. This is -- it's fair to assume that this is where steady state will be until the separation and going forward.

And then on the NAND -- the fabs are -- we're running at full utilization. We don't have underutilization costs on NAND at this point. And then the third part of the question.

On the CapEx.

FY '25 CapEx.

Yes. So on the CapEx side, look, we're -- what I'm -- what I think, we will share is where we see CapEx in this current quarter. And the -- from a cash CapEx perspective, for this quarter, I expect it to be more or less in line with the fiscal fourth quarter which was around 2% of revenue. And our thoughts on CapEx with respect to basically future investments haven't changed, we'd like to continue to focus on the profitability of the business. And at this point in time, that's really our main focus.

The next question comes from Steven Fox at Fox Research.

I was just wondering how -- given experience in prior cycles on the HDD side and you mentioned the 52-week lead times, how do you sort of protect yourself against customers being overzealous with orders and then maybe disappointing you down the road? What are you doing to make sure customers live up to their promises.

I mean, these are very big relationships with very big customers. So we're very close to them on a day-to-day basis. So that is clearly something we look for. But I think there's a lot of relationship value here as well. So we'll be on the lookout for that. But we expect our customers to be fairly rational about this whole process. I mean, look, we're in an area of HDD where almost everybody that we talk to would like to have more HDDs right now. So I think everybody is understanding that giving us more visibility will allow us to set our manufacturing infrastructure in a way where we can satisfy the market without ending up in the situation we are in, in the down cycle where we were having to really, really cut costs and quite frankly, move a lot of people off the payroll which is an unpleasant exercise. So we'll stay very close to our customers. And we have also have a lot of information on kind of ordering patterns and all that kind of stuff. So we'll keep an eye out for that.

[Operator Instructions] Our next question today comes from Asiya Merchant with Citigroup.

Great If I may, just if you can talk a little bit about the industry structure that has changed and maybe how it reflects in your cost declines across both your Flash and HDD business. how we should think about that going forward, given that you see a favorable pricing environment and if you can balance that with maybe improved cost decline for both NAND and HDD.

Yes. So on the HDD side, there's been a couple of things that are -- have been contributing to better margins getting here. One is just we reset the manufacturing base and we've been coming off a very low number. So as we produce more and we get to full utilization, we get better cost per unit out of that. So that's been a good tailwind. But we're -- as I talked about, we're at the point where we know where all the drives over the next couple of quarters are going. And then the other part of it is as you said, technology-driven cost improvements, continuing to innovate, deliver better TCO, better cost dynamics. And we continue to drive down the cost per terabyte on the HDD business. It's going down like high single digits a year or something like that. On NAND, of course, we continue to drive cost down in the 15% range for this year, we're still good with that number. So that's a big part of the storage industry, has continued to drive the cost down through technology innovation and we feel very good about where we are in the R&D teams we have that can continue to do that.

Thank you. This concludes our question-and-answer session. I'd like to turn the conference back over to David Goeckeler for closing remarks.

All right. Thanks, everyone. You appreciate your time today. We look forward to talking to everybody throughout the quarter. We're very happy with where the business is and where we're driving, we think the portfolio is in a great shape and things are going in the right direction. We look forward to talking to you about it more as we move throughout the quarter. Thanks for your time today.