Seagate Technology Holdings Plc (STX) — Q1 2026 Transcript

Thank you. Hello, everyone, and welcome to today's call. Joining me are Dave Mosley, Seagate's Chair and Chief Executive Officer; and Gianluca Romano, our Chief Financial Officer. We've posted our earnings press release and detailed supplemental information for our September quarter results on the Investors section of our website. During today's call, we will refer to GAAP and non-GAAP measures. Non-GAAP figures are reconciled to GAAP figures in the earnings press release posted on our website and included in our Form 8-K. We've not reconciled certain non-GAAP outlook measures because material items that may impact these measures are out of our control and/or cannot be reasonably predicted. Therefore, a reconciliation to the corresponding GAAP measures is not available without unreasonable efforts. Before we begin, I'd like to remind you that today's call contains forward-looking statements that reflect management's current views and assumptions based on information available to us as of today and should not be relied upon as of any subsequent date. Actual results may differ materially from those contained in or implied by these forward-looking statements and are subject to risks and uncertainties associated with our business. To learn more about the risks, uncertainties and other factors that may affect our future business results, please refer to the press release issued today and our SEC filings, including our most recent annual report on Form 10-K and quarterly report on Form 10-Q as well as the supplemental information, all of which may be found on the Investors section of our website. With that, I'll hand the call over to Dave.

Thanks, Shanye, and hello, everyone. Seagate delivered a very strong start to fiscal 2026. Revenue grew 21% year-over-year. Non-GAAP gross margin set a new company record at 40.1% and non-GAAP operating margin climbed to 29%, a level last seen in fiscal 2012. Non-GAAP EPS exceeded the high end of our guidance range, underscoring our focus on expanding profitability. Today, we announced an increase to our quarterly dividend of approximately 3%, reflecting confidence in our execution and ongoing sustainability of our cash flow generation capabilities as we leverage our leading HAMR technology and a strengthening demand environment for high-capacity hard drives. Demand strength was led by global cloud service providers, and we also saw meaningful sequential revenue growth from enterprise customers in the September quarter. The data center end market, which is comprised of nearline sales into cloud, enterprise and VIA customers represented 80% of overall revenue. Amid this improving demand backdrop, our high-capacity nearline production is largely committed under build-to-order contracts through calendar 2026. Additionally, longer-term agreements that we have with our global data center customers provide clear visibility through calendar 2027, reinforcing our view that these favorable demand conditions will persist. We remain focused on executing our HAMR-based product roadmap to support our customers' growing exabyte needs and continue working with them to transition to higher capacity drives. There is no question that AI is reshaping hard drive demand by elevating the economic value of data and data storage. This is evident by the growing demand for our high-capacity nearline drives as customers continue to ramp investments in AI applications. AI inferencing is set to inflect and scale rapidly, further increasing data's value. Inferencing consumes and generates large volumes of data, which is then stored, monitored, validated, and reintegrated into an infinite training loop. We are already seeing the positive impact of this trend as global CSPs deploy large-scale inferencing applications that rely on multimodal inputs such as text, audio, and video. Using monthly token consumption as a proxy for inferencing adoption, one major hyperscaler reported a 50-fold increase in the span of a year. This explosive growth is driving a sharp increase in unstructured data generation that creates demand for hard drive storage. Video content is a major contributor of unstructured data and is driving considerable demand for hard drives today from social media platforms to content delivery networks and online marketing. AI-generated videos promise to further fuel demand growth. There are already numerous text-to-video tools that democratize creativity by letting anyone generate professional quality videos from text, images, or sketches. We see this trend already taking hold. For example, Google reports over 275 million videos were generated on its Veo platform within the first five months. With a 1-minute AI video being up to 20,000 times larger than a 1,000-word text file, the data storage implications are clear. The rapid adoption and growing capability of these tools are already having a positive impact on the demand for storage. Beyond the application space, we have discussed new storage use cases from emerging trends around hybrid cloud environments that enhance data security and compliance with data sovereignty regulations. Recently Seagate partnered with a global CSP to develop a sovereign cloud solution for managing massive volumes of sensitive telemetry and sensor data collected from a fleet of autonomous vehicles. These types of data sets are subject to strict requirements and stipulate such data must be processed, stored, and managed locally. As in any other data center, hard drives provide the ideal solution by meeting customer requirements for throughput, durability, and cost-efficient long-term data retention. As data generation explodes and new use cases emerge, Seagate is answering the call with a clear long-term roadmap to capture demand. Momentum continues to build for our HAMR-based Mozaic platforms, and we achieved several important milestones in the quarter consistent with what we discussed during our analyst event. We now have five global CSPs qualified on Mozaic 3+ terabyte per disk products, which can deliver capacities up to 36 terabytes per drive. We remain on track to qualify the remaining three global CSPs within the first half of calendar 2026. Additionally, we shipped over 1 million Mozaic drives in the September quarter. These products are performing well in live production environments, and we are on pace to achieve 50% exabyte crossover on nearline HAMR drives in the second half of calendar 2026, and we started qualification with a second major CSP on the Mozaic 4+ terabyte per disk platform, with initial volume ramp starting in the first half of next calendar year. This platform will offer capacities of up to 44 terabytes. Advancing aerial density is a key competitive advantage, not just for Seagate, but for the hard drive industry overall. We are leveraging our manufacturing expertise and advancements in technologies, including silicon photonics to pave the path to 10 terabytes per disk. Our aerial density roadmap delivers a superior and sustainable TCO advantage for hard drives compared to alternative technologies well into the future. Customers clearly see the value of transitioning to higher capacity HAMR products as the most efficient way to support their rapidly expanding data storage needs in an AI-driven world. Wrapping up, the Seagate team continues to execute at an exceptional level. We are delivering on our target financial framework supported by a structurally improved business and a strong sustainable demand environment. We are advancing our HAMR led technology roadmap, which creates significant value for our customers and positions Seagate for long-term success. With the strength of our technology roadmap and the transformative impact of AI, we believe the best years are still ahead of us. I am proud of how our teams are rising to meet the opportunities ahead as we remain focused on delivering profitable revenue growth and expanding cash flow generation in fiscal '26 and beyond. I'd like to thank our employees, supply partners, and customers for their many contributions to our performance and to Seagate's ongoing success. Let me now turn the call over to Gianluca.

Thank you, Dave. Our September quarter performance demonstrates strong operational execution and underscores the enhanced structural economics of our business model. We delivered revenue of $2.63 billion, up 8% sequentially and up 21% year-over-year. We achieved a record non-GAAP gross margin of 40.1%, up 220 basis points sequentially. And we expanded non-GAAP operating margin by 280 basis points to 29% sequentially. Our result in non-GAAP EPS was $2.61, exceeding the high end of our guided range. We have continued to execute our technology roadmap to support ongoing demand momentum for our higher capacity products. In the September quarter, we shipped 182 exabytes, up 32% year-over-year, with the vast majority of that volume delivered to global data center customers. As we shared last quarter, we will be discussing the business across two key markets: data center, which is comprised of nearline products and systems that are sold into cloud, enterprise, and VIA customers, and edge IoT, which includes consumer and client-centric markets, along with network-attached storage. In the September quarter, data center revenue represented 80% of our total revenue at $2.1 billion, up 13% sequentially and 34% year-on-year. Demand from global cloud customers continues to grow, and we also saw a notable improvement in the enterprise OEM markets. We project these positive trends to continue with cloud growth expected to outpace enterprise demand. Whether data is stored in public cloud, private cloud, or on-premises, the shift from AI model training to inferencing is driving the need for large capacity drive storage. This includes everything from saving checkpoints to maintain model accuracy and integrity to storing the vast data sets required for effective inference results. In the September quarter, we shipped 159 exabytes into data center customers, up from 137 exabytes in the prior period. Cloud exabyte demand increased for the ninth consecutive quarter, resulting in close to 80% of nearline volume on drive capacities at or above 24 terabytes as customers continue to move up to higher capacity drives. Over the past year, average nearline drive capacity has increased by 26%, which is a primary contributor to our exabyte volume growth. Amid tight supply conditions, we are partnering closely with data center customers to support and where possible, accelerate their qualification timeline on our high-capacity Mozaic products. As Dave highlighted earlier, a majority of the largest cloud customers in the world are now qualified on our HAMR-based Mozaic drives, and we are continuing to ramp these products to support customer demand. The strong data center growth that I just described more than offset lower sequential sales in the edge IoT market, which made up the remaining 20% of revenue at $515 million. We are expecting some seasonal improvement in Edge IoT revenue in the December quarter from both VIA, Edge, and consumer products. Moving on to the rest of the income statement. Non-GAAP gross profit increased to $1.1 billion, up 14% quarter-over-quarter and 46% compared with the prior year period. We expanded non-GAAP gross margin to 40.1%, which represents an incremental margin of nearly 70%. This margin growth reflects the benefit of increased adoption of our latest generation products and ongoing execution of our pricing strategy. Non-GAAP operating expenses were $291 million, up 2% quarter-over-quarter and in line with our expectations. The combination of strong top-line growth and significant financial leverage drove a 19% improvement in operating profit to $763 million. Other income and expense were $74 million, and we are currently projecting OI&E to be essentially flat in the December quarter. We grew non-GAAP net income to $583 million with corresponding non-GAAP EPS of $2.61 per share based on tax expenses of $106 million and a diluted share count of approximately 223 million shares, including the net impact of our 2028 convertible notes of approximately 7 million shares. Turning now to cash flow and the balance sheet. We invested $105 million in capital expenditures for the September quarter or roughly 4% of revenue. For fiscal '26, we anticipate capital expenditures to be inside our target range of 4% to 6% of revenue while we continue maintaining capital discipline. Free cash flow generation was flat quarter-over-quarter at $427 million, including the substantial variable compensation payout we discussed on our July earnings call. Looking ahead, we expect free cash flow generation to expand in the December quarter. We returned $153 million to shareholders through dividend. As Dave noted earlier, we are increasing our quarterly dividend by approximately 3% to $0.74 per share. We deployed $29 million to repurchase shares of our common stock at an average price of $187 per share. We will continue to opportunistically repurchase shares and anticipate share repurchase activities to vary from quarter to quarter. We remain committed to returning at least 75% of free cash flow to shareholders over time. Cash and cash equivalents increased 25% sequentially to close the September quarter with ample liquidity of $2.4 billion, including our undrawn revolving credit facility of $1.3 billion. We exited the quarter with gross debt of approximately $5 billion and net leverage ratio of 1.5 times based on adjusted EBITDA of $831 million for the September quarter, up 19% quarter-over-quarter and up 67% year-on-year. We are pleased that our strong execution is being recognized with S&P upgrading our credit rating earlier this month. Looking ahead, we expect net leverage ratio will continue to trend lower as profitability increases in the coming quarters. Additionally, we are exploring opportunities to further reduce debt, supporting the positive leverage ratio trajectory. Turning now to the December quarter outlook. The demand environment remains strong, particularly among global cloud data centers. We expect to increase revenue and expand margins as these customers continue to shift to our next-generation storage solutions to support their increasing demand. We expect December quarter revenue to be in a range of $2.7 billion, plus or minus $100 million, which represents a 16% year-over-year improvement at the midpoint. Non-GAAP operating expenses are expected to remain relatively flat at approximately $290 million. Based on the midpoint of our revenue guidance, non-GAAP operating margin is expected to expand to around 30%. Non-GAAP EPS is expected to be $2.75, plus or minus $0.20 based on a tax rate of about 16% and non-GAAP diluted share count of 227 million shares, including estimated dilution from our 2028 convertible notes of 10 million shares. As demonstrated by our September quarter results, Seagate is delivering on our financial commitments, reinforcing our track record of operational execution. Our performance is underpinned by a strong product roadmap that offers enterprise exabyte scale storage solutions, enabling them to maximize the potential of their data. This strength positions Seagate to drive meaningful value for both customers and shareholders.

Congratulations on a successful quarter. You mentioned at the start about the strong orders and it seems you have an order backlog extending into 2027. The supply appears to be quite tight in the market. I'm curious if you have any plans to increase capacity or if there are specific supply chain bottlenecks that might ease over time. Clearly, there are advantages and disadvantages to that, so I would appreciate your thoughts on the supply-demand balance and any capacity additions you might be considering. Additionally, congratulations on qualifying the fifth hyperscaler for HAMR and solid execution thus far. I'm interested in what feedback you are receiving from customers regarding HAMR adoption. Is there any change in the projections you previously provided for the HAMR rollout moving forward?

Thanks, Mark. Your two questions kind of go hand in glove. What I would say is that our strategy for adding capacity, if you will, is to go through product transitions. We're not really adding unit capacity. Through some of these product transitions, we actually lose a little bit of capacity because the process content is a little higher as we go through, but we add exabyte capacity. And to your question about HAMR, a lot of the reason that customers are engaging with us on these long-term agreements is because we have visibility into higher and higher capacity points. So the demand that most of the hyperscalers certainly are feeling is for more exabytes. More efficient exabytes in their data center allows their space and power and all their other metrics to get the best returns of TCO, if you will. And that's what we're really trying to answer the call for. We're very focused on going through the product transitions, getting our yields up on the product transitions and then ultimately transitioning to the point where we get 40s and 50s and so on and add exabyte capacity that way.

Any update on the HAMR rollout, sorry?

Nothing more than what we said in the script, I think the qualification you made reference to. The ramp continues on and part of this is predictability of those transitions that we're going through. We have to make sure we're staged for that so that we get the customers what they need.

Yes, we are happy to have achieved another call during the quarter. So now we have 5 customers qualified on 5 big customers, cloud customers qualified on HAMR. Of course, this is contributing to our delivery in the quarter and how we guided December quarter. So we are achieving those level of revenue and profitability a little bit faster than what we were thinking. And of course, this is related to the transition of the mix to higher capacity drives mainly to HAMR.

And congrats on your results tonight. Gianluca, since your May Analyst Day, you've reported 2 quarters where your incremental margins have been 60% to 70%. Is it fair to say that in the new demand environment that we're in and the need for higher capacity drives, we should be thinking about your incremental margins just being higher, consistently higher than that 50% incremental margin you outlined at your Analyst Day? Or why would we not see these level of incremental margins sustain? I realize not literally every quarter. But generally speaking, why would we not see 60% to 70% incremental margins sustain from here?

Thank you, Erik. Well, you're right. No, we are executing a little bit better than what we were expecting. And as I said before, this is mainly due to the move to better mix in terms of profitability. Not every quarter is the same. So of course, now that we are qualifying more customers and moving more customers to the HAMR drives. We have a little bit of a higher support in terms of profitability. I would say this will be different every quarter, even the pricing strategy that we have is very consistent, but not every quarter, we have the same number of new negotiations going on. So it's not easy to estimate exactly what will be the profitability and the mix of a specific quarter. I think in the short term, we are delivering very good results. Longer term, I think the model that we presented at the Analyst Day is a strong model. But you're right, in the short term, we are doing a little bit better. And you have seen how we got in December, it basically implies a higher margin than the 50% incremental from September.

I was wondering if you could maybe address the level of cost reduction you expect to achieve on a blended basis as we look out into calendar 2026. Is there anything that would kind of prevent you from achieving that kind of mid-teens cost down on a blended basis, even as you ramp HAMR more aggressively? Could that actually be better than that? Or is there a reason it would be worse than that? And then just to clarify, the prior statement you made on not adding unit capacity, is there any kind of benefit you get in terms of sort of dual capacity tracking HAMR versus conventional technologies that would add a little bit of unit capacity into the overall mix? Or is that just an overall dilutive event to overall units?

I'll address the second part of your question first and then let Gianluca respond to the cost inquiry. My understanding is that some of our PMR technologies are currently in high demand, and as we shift towards HAMR and the newer products with capacities of 30 to 40 terabytes, we will need to make significant changes. We don’t see this as reverting to building more PMR technology; rather, we view it as freeing up PMR technology to facilitate this transition, which is how we plan to increase exabyte capacity. Regarding unit capacity, much of this depends on the requests from customers who are qualified and our plans in collaboration with them. These plans require considerable time to develop. As Gianluca mentioned earlier, if we experience any marginal upside, it will be due to pulling future results closer. Gianluca, would you like to address the cost question?

Yes. On the cost, of course, we don't guide cost for calendar '26. I would say the good improvement in our cost per terabyte is coming from the transition to higher capacity drives. So the mix is very important and every quarter is different. I would say you will see more and more transition to HAMR. We have two customers that are qualifying 40-terabyte drive. That, for sure, will be a good boost to our reduction of cost per terabyte during calendar '26.

There was commentary earlier about inference demand. If you could talk a little bit about the visibility of that demand. What gives you confidence that this one continues to grow maybe in terms of applications that you're seeing and that's kind of giving you the confidence of demand through calendar year '27? And how should we think about seasonality here? Typically, March does have a seasonal drop? How should we think about that given AI inferencing demand is pretty strong here?

It's very interesting to observe. I believe that much of our demand in recent years has been driven by the shift to video, with over 80% of Internet traffic now consisting of video content. Every day, there are tens to hundreds of millions of new videos uploaded. To address your question, the quicker people can generate and access these videos through inferencing, the better it is for our storage needs. As for how this will evolve over the next nine to twelve months, it’s somewhat uncertain, but we're optimistic with the introduction of new applications that enable video generation. The videos are becoming longer and richer in content, with more embedded video elements. However, predicting this is challenging. Our customers seem to be facing similar uncertainties, which is contributing to the strong demand.

Yes. On the seasonality in the March quarter, of course, it's a bit early to discuss about March. We just guided December. I would say considering that data center revenue is 80% of our total revenue, the impact of seasonality is probably lower than what you have seen in the past. I think every year will be a little bit lower. But of course, we are not guiding much, but we expect a good quarter.

Can you talk a little bit about how you're managing pricing in this very constrained environment? How much is contractually locked in going into a quarter? How much flexibility do you have in intra-quarter basis? And when we look at just the reported quarter, right, like the dollars per terabyte decline was consistent with the prior quarter. But obviously, the demand environment is very strong. So hoping you could unpack that a little bit. Is there a differential there between HAMR and non-HAMR that's contributing to that? Or what's causing that dollar per terabyte decline to be relatively consistent, given just like this very strong demand backdrop?

Yes, thank you, Wamsi. Much of our current focus is quite predictable. As mentioned earlier, if we can enhance our execution slightly beyond expectations, we can reduce costs or expedite the qualification of certain products, though we are somewhat limited in supply. Moreover, we are ensuring that our customers benefit from predictable economic conditions through our contracts. As we navigate these transitions more swiftly over time, we can position ourselves to secure the appropriate economic returns aligned with market needs for upcoming contracts. This is our long-term strategy. In the near term, everything is progressing according to plan, albeit with slight adjustments.

Yes, our pricing strategy has remained consistent for about 10 consecutive quarters. When we renegotiate a contract, we make a slight increase in pricing for the same product. Additionally, when customers transition to higher capacity products, they receive a lower price per terabyte. This is why, during the major shift of customers to the HAMR product with higher capacity, you may notice a slight decrease in the average price per terabyte. However, the profitability reflects the like-for-like price increase and the reduction in cost per terabyte due to the product mix.

I want to clarify the March seasonality question earlier. Curious, if you were to assume that consumer were seasonal, could you pivot supply more to the cloud? And then I guess as the main question, your customers are turning to SSDs, given the tremendous tightness on the HDD side. Curious your thoughts around that cannibalization. And I guess what would maybe change your mind in terms of the vision for this sustainably higher demand and then potentially add capacity to support that.

Yes. Thanks, C.J. So I don't think that customers are really changing their architectures because of what they're seeing right now. What everybody is driving us to do is getting more predictable over time and if anything, be more aggressive on the product transition. So I don't really think there's any 'cannibalization.' As a matter of fact, I don't think it's in anybody's economic benefit to do so. And the architectures are pretty well set going out for a couple of years.

Yes. We are seeing that the gap between supply and demand is increasing each quarter. This indicates that demand is shifting more toward the future and is not being addressed by other technologies.

Your observation on seasonality is quite intriguing. In certain edge IoT markets, we do see some seasonality. We are gradually reallocating some supply from Edge IoT products to cloud products as demand shifts. This transition is happening organically and relates to the product changes we've discussed. We cannot move quickly on this until we reach products like the 4 terabyte per platter, where there is consistency across platforms. Despite the slight decline in Edge IoT revenue, profitability is actually improving because we are prioritizing these products differently than when there was excess supply. I believe this will result in somewhat muted seasonality, as that market remains robust, with last quarter's Edge IoT revenue surpassing $0.5 billion.

I had a question and a clarification. Dave or Gianluca, in the past, you spoke about sharing the cost benefit of HAMR with your customers. But now with cloud becoming a bigger portion in the AI tailwind, I'm wondering if you're rethinking your pricing strategy for HAMR, i.e., can you increase it further? And then a clarification, Gianluca, historically, your revenue guide had a range of $150 million, now it's more like $100 million. So is that because better visibility build-to-order helping you tighten that range?

Yes. Thanks, Chris. I believe what you mentioned is very accurate. As we navigate these product transitions, we are striving to be as predictable as possible for our customers. However, some contracts are quite lengthy. As demand increases, the market will adapt. It's important to note that most of the advantages our customers are experiencing come from the total cost of ownership proposition with higher capacity drives. Pricing plays a role in that, but the benefits they see in their total cost of ownership are significant as well. Therefore, adjustments will occur gradually as the market evolves.

On the HAMR pricing, in the past, we discussed only with one customer to giving them a slightly lower price, but this customer, of course, helps us with the first qualification of the product. And so for a certain volume, they have a lower price than other customers. But of course, this is transitioning fairly quickly. It's just a matter of a few more quarters.

I guess, Dave, as you see an uptick in video creation, you've been talking about this, but with offerings like OpenAI, Sora. And given your nearline capacity is fairly committed through calendar '26, are you seeing customers looking to potentially fund or co-invest CapEx dollars for Seagate to get access to more units? And is that something you'd be open to? I would love to just understand like if this sort of keeps playing out the way you outlined, is aerial density going to be enough? Or would you or your customers have to eventually add some capacity to get more units? I'd love to understand if you'd be open to that co-investing angle. And then if you just qualify this a little bit. When you talked about shipping 1 million-plus Mosaic drives this quarter, does that imply that about 36 exabytes of the total units were HAMR-driven this quarter? Is that fair?

I'll let Gianluca answer the question about exabyte capacity because he'll calculate it, but you're close. We believe that going through these transitions opens up greater capacity opportunities. From a customer perspective, there's limited visibility on what some of these new tools will achieve, but there's a lot of optimism regarding demand, particularly in video applications. Consequently, no one wants to be too late to these developments. I don't think there are any major architectural changes, but what we see as driving factors is stability. Customers want predictability in what they will receive. The demand levels exceeding our supply are reported by someone else, and that's not necessarily how each customer views it. What customers are noticing is a temporal shift, asking if they can pull in capacity a bit because they recognize that we are increasing exabyte capacity over time, particularly through product transitions. That's what contributed a bit more to last quarter's performance. As we push through product transitions, we will be adding more exabytes. Completing these qualifications and ramping up production is a top priority for customers as it's how they plan to increase capacity.

Yes. No. First of all, I think it's very important that we keep the current balance between supply and demand and not taking action to oversupply in the future, the industry. In terms of exabyte, it's a nice question. I would say you know that our HAMR product is between 30 and 36 terabytes per unit. So with 1 million units sold in the quarter, we are into that range.

Yes, first regarding housekeeping, I’m interested in how we should view the systems business in the data center sector, considering its performance over the last several quarters and how it may trend moving forward. Additionally, in terms of nearline demand, given the growth we're experiencing and looking back at the Analyst Day where a mid-20% CAGR was mentioned, have your thoughts changed about whether that projection could be higher than what you initially expected a few months ago?

Thank you, Aaron. From the systems business, we package the drives into racks to assist some customers with their workload. This is a fairly small group of customers, but they are valuable to us, and we don't expect significant changes in this area. The customers who are currently purchasing their drives and using external integration services are not likely to shift towards our systems. We are closely monitoring demand, especially concerning the mid-20% growth target, which remains a key topic of discussion regarding supply levels. I want to reiterate that increasing our exabyte supply hinges on releasing larger drives, and that is our current focus.

Yes, Aaron. And we report system as part of data center, and this is very similar to what the rest of the industry is doing. So we try to align, so it's easier to look at the different players in the same way.

I wanted to understand the timing of the crossover between the Mosaic 3 platform and the Mosaic 4 platform. I think you guys have historically talked about the first generation is what takes a long time. That's what was the struggle with your first big customer, but now you're really seeing that adoption kind of accelerate across other global CSPs. So at Mosaic 4, it sounds like it's starting to ramp in volume in the second half fiscal year '26. Could you get to a crossover point in the first half '27? Like should we be thinking about 15% or 20% contribution? I'm just trying to understand how much supply you can actually add to the industry with that transition in the short period of time.

Thanks, Tom. Yes, we are also considering this question because as we progress, yields and scrap rates on the new product are what determine our results. We have a good understanding of the old product since there is significant commonality between the two technologies. Therefore, this isn't completely unfamiliar as we ramp up. From my viewpoint, I anticipate a slightly quicker ramp-up. We need to execute successfully, and that's where the team's focus lies. The qualifications are progressing well so far, so there’s no reason for disappointment at this point. I will continue encouraging the teams to increase the yields as quickly as possible to expedite the transition.

Dave, you spoke about longer-term agreements offering visibility into 2027. Presumably, that visibility is on exabytes for HAMR drives. But since you didn't specify what that visibility is, does exabyte slowdown in the context of that 25% exabyte growth over time? Or do you expect demand actually accelerates into '27 from what you see today?

There are two questions to consider: one about supply and another about demand. I believe the supply will increase significantly as we transition to the 4+ product family, which will help us generate much more data. We are pushing hard to achieve this. We are collaborating with customers, especially major hyperscalers, on the five products we have qualified on the 3+ version. Our goal over the next year is to qualify as many products as possible on the 4+ version, which will result in a substantial increase in data output. That's our focus.

I also had a question about the HAMR crossover. So five of the CSPs are qualified now and the other three are going to get qualified in the first half of next year. It sounds like it's about high teens of exabytes now, maybe pushing 20%. I would have thought you could get to exabyte crossover before a year from now if more than half the CSPs are called now and everyone is pushing to get these higher-cap drives. So why would it take another year to go from 15% to 20% of exabytes to more than half of exabytes?

Yes, there's a delay in the supply chain. We have started wafers for various products, and we will be selling those products. However, simply getting the qualifications does not mean we can instantly switch from the old product to the new product. We need to manage that transition in our factories ourselves, and we will do this as aggressively as possible. As I mentioned in response to Tom's question, there is some commonality, allowing us to adapt to some extent, but not as easily as we would prefer. The progress we can make will also depend on our yield ramp with the new products. Overall, things are going well, and we will be as proactive as we can.

It takes time to ramp. Now as you know, the cycle time for HAMR is not short. So we need to qualify first and then to ramp the product. So it takes a certain number of quarters to really go up in capacity.

Just listening to all the questions and thinking back to the presentation in May where you lined up sort of the roadmap and the time it takes. It seems like hard disk drives is going to become more of a bottleneck to some of the expansion plans for the cloud guys as we get into next year. How legitimate is that of a concern? And what else could be done at the cloud guys to just sort of leverage existing capacity to sort of get through that period and keep you guys on track?

Yes, Steve, I don't look at it as something that's immediate or going to be solved in a period of 3 months, the industry is not going to bring on more supply in 3 months. So really what's coming out of all of this is customers are getting very predictable on long-term plans. And that's what they need to do, and that's that we need them to do as well because we've got long cycle times as we've been saying. So it's building industry health and that's good. We can't react with supply to everyone's wish list, and there's probably wish list that aren't real at some point. But the customers that we're working with are being very predictable for us telling us exactly what they're going to need, and we're answering, call it, to the extent that we can. And if they need more exabytes usually, we're off book. It's a timing problem, a temporal problem. We're up by a few months or 6 months or something like that. And so ultimately, as we go through these transitions, we're going to bring on more exabyte capacity and they're going to be happy with it. Again, we don't see any evidence architectures changing or anything like that.

Dave, how should we think about the pace at which customers will look to transition to higher capacity points given the shortages and density? Do you think customers will mix up faster than they have historically? I understand there is a significant increase in capacity when switching to HAMR, but once HAMR normalizes, let's say in about 12 months, could we see an even quicker transition within HAMR? If not, what factors would hinder that? Thank you.

Thanks, Ananda. It doesn't feel like it was too long ago when we discussed moving from 16s to 18s or from 18s to 20s, which offered diminishing returns on PMR products. Now as we shift from 30 to 40, we’re witnessing a significant demand for 40s. I believe this trend will continue with 50s as we work to facilitate that transition, especially since the total cost of ownership is much more favorable for those building data centers who need these products. This change reflects our customers’ behaviors. In the past, when transitioning from 16 to 18 or 18 to 20, we didn’t see much momentum. Currently, on the real demand side, the global context is also a factor; we mentioned video in the script. Certain video sectors are currently booming, and the ability to produce and diversify video content is substantial, with people successfully monetizing it. This surge is fundamentally driven by human creativity, which I don’t anticipate will slow down.

Could you elaborate on the duration of the long-term agreements for HAMR? Is there a pricing component to it? And what percentage of total revenue is currently based on those agreements? How does that compare with what the mix of those agreements was last year?

Well, every customer has a different agreement or build to order. So the duration is different, the volume is different, the mix is different. But what we said in the prepared remarks, the vast majority of our nearline exabyte have already been committed for entire calendar '26. So we have a fairly long build to order in place.

Yes, it's a good question, Tristan. I think they'll remain like they are today. Getting through the HAMR transition, you have to add a significant amount of content to wafer, in particular, which is one of the longest lead times. But the next generation and the next generation after that, you don't have to go through that amount of transition again. So I do think it's a step up in lead times. We'll work it to get it back down as quickly as we can, but I think there's a substantial amount of process content. So we've gone through that step up, at least on those products. And then after that, we don't have to get to 4 terabytes per platter, 5 terabytes per platter and so on.

Congratulations on another good quarter. I'm curious if your margins and yields on your HAMR drives are comparable to those of the legacy PMR drives. If they are not, when do you anticipate that happening?

Yes, Mark, I would say we don't have ePMR. Our latest generation PMR drive is excellent and performs well with great yields, which I am very pleased about. We will continue to focus on developing the next generation drives as diligently as possible. Our current emphasis is on achieving 4 terabytes per platter, where we aim to improve yields, although it's still early in the process. We will make every effort at Seagate to accomplish this within the year.

So just I saw good numbers on the gross margin side, up pretty nicely sequentially. Just wondering with the drop-through on HAMR, would you expect to hit the mid-45% to 45% margins exiting fiscal '26? Or how should we look at that margin progression? And I have a follow-up.

Yes. As I said before, from Erik's question, no, we are very pleased with the progression in gross margin and also with the increase in revenue. So we have achieved the $2.6 billion in revenue and the 40% gross margin a little bit earlier than what we were thinking. And we have guided December with an incremental margin that is higher than the 50% that we discussed in our financial model. Every quarter is a little bit different. Right now, we have a fairly quick transition from the PMR drive to HAMR drive. And with higher capacity drive, we get better profitability. So in the short term, we are progressing well. And as I said, every quarter will be a little bit different. I think the model that we presented in May is a strong model for the next three years, but it doesn't mean we cannot execute even better.

Right, Vijay. And as I reflect on a lot of the questions today, it's about how well do we don't know the demand further out and some of it's predictable, but the demand may keep on coming based on what we see in some of the end applications and then our ability to continue to work the yield issues around the new 4 terabyte per platter and get through that transition as quickly as we can, and that will help the cost side and supply side from an exabyte perspective. So that's what we're all focused on.

Got it. And Dave, regarding the video side with the increased traffic, do you see potential for higher long-term exabyte growth than the 20% to 25% you previously outlined, considering the significant rise in video traffic on storage?

We're currently examining the situation closely. We've observed a significant interest in the new AI content generation capabilities we've mentioned earlier, and we're looking into how quickly people are adopting these innovative features.

Thanks, Gary, and thanks to everyone for joining us today. As you can see, fiscal 2026 is off to a great start. It's marked by strong operational execution and outstanding financial results. I want to once again express my gratitude to our employees, our suppliers, our customers, and our shareholders for their contributions to Seagate's ongoing success. Together, we are advancing innovation to serve growing data storage demand and position Seagate for long-term value creation. Thanks for your continued support. We look forward to the significant opportunities ahead.