Good morning, and welcome to the Seagate Technology fourth quarter and fiscal year 2019 financial results conference call. My name is Kelly and I'll be your coordinator for today. At this time, all participants are in a listen-only mode. Following the prepared remarks, there will be a question and answer session. As a reminder, this conference is being recorded for replay purposes. At this time, I would like to turn the call over to Shanye Hudson, Vice President, Investor Relations. Please proceed, Shanye.
Thank you. Good morning, everyone, and welcome to today's call. Joining me today are Dave Mosley, Seagate's Chief Executive Officer, and Gianluca Romano, our Chief Financial Officer. We've posted our earnings press release and detailed supplemental information for our June 2019 quarter 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 Form 8-K that was filed with the SEC. 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 effort.
As a reminder, this call contains forward-looking statements, including our September quarter financial outlook and expectations about our financial performance, market demand, industry growth trends, planned product introductions, ability to ramp production, future growth opportunities, and general market conditions. These statements are based on management's current views and assumptions and should not be relied upon as of any subsequent date. Actual results may vary materially from today's statements. Information concerning our risks, uncertainties, and other factors that could cause results to differ from these forward-looking statements are contained in our most recent Form 10-K filed with the SEC and the supplemental information posted on the investor section of our website. Following today's prepared remarks, we'll do our best to accommodate your questions. With that, I'll turn the call over to you, Dave.
Thanks, Shanye. Good morning, everyone, and welcome to our quarterly earnings call. I will start off by summarizing key highlights from the June quarter and sharing some perspectives on the market before outlining our progress on our key priorities. Afterwards, Gianluca will discuss details on our June quarter financial results and provide our outlook for the September quarter. Following the prepared remarks, we will open the call for questions. Seagate continues to deliver on its financial commitments, achieving June quarter results that were solidly in line with our expectations. We recorded revenue of $2.37 billion and non-GAAP EPS of $0.86, both toward the upper end of our guidance range, against a backdrop of increasing geopolitical uncertainty and regulatory hurdles. These broader macro conditions disrupted our customers' buying patterns, causing trepidation amongst some of our enterprise and OEM partners, while prompting others to accelerate demand, including a few surveillance customers.
Our ability to adapt to market volatility and intelligently manage our business enabled us to increase revenue and exabyte shipments quarter-over-quarter, supported by improving demand for our nearline drives from cloud and hyperscale customers. Additionally, our fiscal year performance demonstrates solid execution on our priorities to optimize profitability and free cash flow. We have been successfully pivoting the business towards growing markets, which include enterprise nearline drives, edge storage for surveillance and NAS, and our cloud system solutions. In fiscal 2019, we delivered annual revenue of $10.4 billion, of which approximately half was derived from these markets. These applications require reliable, cost-effective mass storage, making them well-suited to our portfolio of products. Importantly, they contribute an even higher percentage of our gross profit, providing a solid platform for margins to expand as they become a greater part of our overall revenue.
At the same time, we are continuing to supply HDDs into mature markets, which include mission-critical, edge compute, DVR, gaming, and consumer applications to support our customers' needs. These products require minimal further investment while contributing nicely to our overall operating income. We are continuing to tightly manage expenses while prioritizing investments towards areas that deliver the greatest value to our customers and strong returns for Seagate. In fiscal year 2019, we reduced our full-year non-GAAP operating expenses by 9%, while increasing our investments in next-generation technologies to improve areal density and lower cost per terabyte. We are leveraging our significant free cash flow generation to enhance shareholder value. In fiscal 2019, we delivered $1.2 billion in free cash flow and returned $1.7 billion to shareholders through dividends and buybacks, demonstrating our longstanding commitment to capital returns, as well as our confidence in sustainable cash flow generation.
We continue to advance our technology roadmap and focus on being first to market with new product solutions. This strategy enables us to provide our customers with cost and performance benefits at an attractive margin for Seagate. As we shared last quarter, we began shipping our 16-terabyte drives in late March to deliver the world's highest capacity storage solutions, and we have already introduced products for both enterprise and edge storage applications. Customer qualifications are progressing well, and we remain on track to ramp high volume shipments later in calendar year 2019. In addition to driving the next generation of high-capacity storage, we are the first to introduce dual actuator technology. This technology effectively doubles the performance at the same capacity points, making it ideal for cloud workloads and edge sequential operations, servicing large data flows such as video streaming, smart factories, AI, and machine learning.
Our MACH.2 dual actuator technology is garnering strong interest. Customers have started to qualify these drives, which we expect to begin shipping later this calendar year and become increasingly critical across the industry, starting around the 20 terabyte capacity point. Looking ahead, we expect to capture another industry first with the introduction of 20 terabyte capacity drives, which will be based on our highly scalable HAMR technology. Six years ago, I stated that HDDs would be 20 terabytes by the end of calendar 2020, we remain on track to hit that target. We are focused on making the transition to HAMR technology seamless for our customers. Our HAMR drives are built on a common platform to current 16 terabyte drives, which is helping to accelerate maturity and adoption in the market.
As we enter fiscal year 2020, we expect the macro-related uncertainties that I described earlier will continue to have some influence on near-term industry dynamics. We expect demand from global cloud and hyperscale customers will continue to improve, particularly for high-capacity drives. Seagate is well-positioned to address this growing demand with a strong technology portfolio, deep customer relationships, manufacturing expertise in precision robotics, assembly, and analytics, and the supply chain flexibility, which together all enable manufacturing cost advantages. We expect our exabyte shipments into the enterprise nearline market will be well above the long-term CAGR of 35%-40% in fiscal year 2020. We expect to deliver healthy revenue growth year-over-year. With that, I'll turn the call over to Gianluca to go into more depth on our June quarter results and share our outlook for the September quarter.
Thank you, Dave. We are focused on driving strong operational efficiency and effectively managing the business through dynamic market conditions. On a sequential basis, we grew June quarter revenue by 3% to $2.37 billion and above the midpoint of our guidance range. We increased total exabyte shipments by 10% to 84.5 exabytes, and we expanded non-GAAP operating income by 8% to $286 million. While the geopolitical situation remains uncertain, we are seeing improving demand conditions, particularly among cloud hyperscale customers for higher capacity nearline drives. Revenue for the enterprise market, which includes nearline and mission-critical HDD drives, represented 41% of total June quarter revenue, up from 39% in the March quarter, mainly due to stronger demand in nearline drives. Exabyte shipments into the enterprise market were up 15% quarter-over-quarter at 38 exabytes.
Nearline drives accounted for more than 90% of that total, with average capacity per nearline drive at nearly eight terabytes. Revenue from 12 terabyte and higher capacity drives now represents more than 50% of total nearline revenue, compared with 36% in the prior quarter. We have successfully qualified our 16 terabyte drives with a number of customers, and we expect shipment volume to increase through the fiscal year. Revenue for the edge non-compute market, which includes surveillance, NAS, gaming, DVR, and consumer applications, increased to 34% of the total June quarter revenue, compared to 32% in the prior quarter. While the first half of the calendar year is typically a weaker period for the edge non-compute market, we saw some acceleration in demand from a few surveillance and gaming customers, which led to a sequential increase in revenue and exabyte shipments.
We shipped a total of 33 exabytes into the edge non-compute market during the June quarter, compared to 29 exabytes in the prior period. Similar to nearline application, a majority of edge non-compute platforms require reliable and secure mass data storage, which is well-aligned to high-capacity disk drives. Revenue from the edge compute market, including desktop and notebook HDD drives, contributed 18% of total revenue relative to 20% of revenue in the March quarter, with exabyte shipments down approximately 6% to 14 exabytes, reflecting typical seasonality. Our non-HDD drive business, including cloud systems and SSD solutions, made up the remaining 7% of June quarter revenue, down from 8% in the prior period. The quarter-over-quarter revenue decline was mainly driven by lower demand from our enterprise SSD customers. Revenue for our cloud system business was slightly down quarter-over-quarter.
However, we improved operating profit, which reflects our ability to transition our portfolio to higher-value products. During the quarter, we jointly announced a new partnership with Cloudian to address the private cloud market opportunity across healthcare, scientific research, and media and entertainment. The Exos CORVAULT solution will be powered by Seagate 15 terabyte high-capacity drives and our new high-density storage server platform, which delivers a cost-effective solution for large-scale deployment. Non-GAAP gross margin for the June quarter was 26.8%, up 20 basis points sequentially on a more favorable product mix. Consistent with our expectations, during the June quarter, we incurred underutilization costs, which were only slightly improved from the March quarter and thus negatively impacted gross margin. We continue to proactively manage our manufacturing output to align closely with the demand environment, which is starting to improve for higher capacity drives.
Therefore, we are expanding our production capabilities to address future growth opportunities. As a result of this dynamic, underutilization cost will remain a headwind on gross margin until demand fully matches our production capacity later in the calendar year. We efficiently managed our non-GAAP operating expenses, holding them flat quarter-over-quarter at $350 million, down nearly $50 million from the year-ago period. The combination of slightly higher gross margin and flat operating expenses resulted in non-GAAP EPS of $0.86 for the June quarter at the high end of our guidance range and reflecting our ongoing operational efficiency and expense discipline. Cash flow from operation was $448 million in the June quarter. Capital expenditures were $151 million in the June quarter and about $600 million for the fiscal year, which was just below 6% of full-year revenue.
Looking ahead to fiscal 2020, we expect CapEx to be near the midpoint of our target range of between 6% and 8% of revenue to support our plans to increase our manufacturing exabyte capacity to address growing demand. Free cash flow was a healthy $297 million for the June quarter and $1.2 billion for the full year. During the quarter, we received a cash payment of $1.35 billion from Toshiba Memory Holdings Corporation for the early redemption of the outstanding preferred shares we held in the company. As a reminder, just over a year ago, Seagate made a $1.27 billion investment in TMC preferred shares. The proceeds represent our original investment as well as accrued interest income. During the June quarter, we retired $272 million in debt, including the repayment of our revolving credit facility.
At the end of the quarter, the company debt balance was $4.25 billion, with a gross debt to last 12-month non-GAAP EBITDA ratio of just below two times. We repurchased 7.8 million ordinary shares for $350 million, illustrating our view that Seagate shares represent an attractive investment. We exit the June quarter with 269 million ordinary shares outstanding, down 6% from the prior year. At the end of the quarter, we had $2.2 billion remaining on our authorization. Our board has again approved a quarterly dividend payment of $0.63 per share, which will be payable on October 9, 2019. Through the combination of dividends and share buyback, Seagate returned approximately $1.7 billion to shareholders in fiscal year 2019, or about 145% of free cash flow, which reflects our focus on enhancing shareholder value.
As of the end of June, cash and cash equivalents were $2.2 billion, up $832 million from the prior quarter, with an additional $1.5 billion available through our revolver. As we enter fiscal year 2020, the industry landscape is improving, and we remain focused on executing plans to expand our manufacturing capacity to address growing demand for mass capacity storage. While these actions have a near-term impact on gross margin, we believe they position Seagate well to capitalize on future growth opportunities. Prior to sharing our quarterly outlook, I'd like to outline a change to our financial reporting. Starting in the September quarter, we will begin excluding share-based compensation expense from our non-GAAP results because companies utilize different factors and methodologies to calculate the expense, as well as to be more consistent with the majority of our industry peers.
This expense is approximately $30 million per quarter, of which a majority is included in operating expenses. I would also point out that the September quarter is a 14-week period, and we expect to incur additional operating expenses due to higher variable compensation and the extra week in the quarter. We expect the net impact to operating expense to be an additional $20 million in the September quarter. Taking these factors into account, our outlook for the September quarter is as follows. We expect total revenue to be in the range of $2.55 billion, ±5%. Non-GAAP gross margin to be relatively flat sequentially. Non-GAAP EPS of $0.90, ±5%. While the macro environment continues to disrupt near-term demand, we expect exabyte volumes to meaningfully grow.
Over the long term, our focus on cash generation and a solid balance sheet will provide us the financial strength to capitalize on future storage growth opportunities and enhance shareholder value. I will now turn the call back to Dave for final comments.
Thanks, Gianluca. To summarize, Seagate is executing well on its strategic priorities to optimize profit and free cash flow. We are continuing to manage the business intelligently through industry-related cycles and the current market dynamics. Over the longer term, we believe the fundamental demand for data is driving the need for mass storage capacity. Seagate is creating solutions to help customers manage the exponential volumes of data securely, efficiently, and cost effectively. As we enter fiscal 2020, I am confident that we have the financial foundation, manufacturing expertise, and technology portfolio to capitalize on these future growth opportunities. We will be hosting an analyst event on September 19th in New York City, where we plan to outline our strategy in more detail.
Before opening the call for questions, I would like to take a moment to thank our customers, suppliers, business partners, and employees for all of their contributions to the success of our business. Gianluca and I will now take your questions.
At this time, I'd like to remind everyone, in order to ask a question, please press star, then the number 1 on your telephone keypad. If you would like to withdraw your question, please press the pound key. Your first question comes from the line of Katy Huberty from Morgan Stanley. Please go ahead. Your line is open.
Thank you. Good morning. I'm surprised gross margin didn't recover more in the June quarter, given the 10% increase in exabytes shipped and the improvement in nearline and in surveillance. It sounds like underutilization is still an issue. Does that tie entirely to the 16 terabyte investment, or are there other areas of the business where you're seeing underutilization? Then just how should we think about the progression of gross margins over the next couple of quarters as 16 terabyte ramps? Thank you.
Good. Thanks, Katy. At a very high level, no, it doesn't tie to the ramp of the 16 terabyte. I'll describe the market dynamics and then let Gianluca do a walk on specific impacts if that's more helpful. I don't think we should underestimate the disruptions in the market from a demand perspective that we saw in Q4 and continue to see ramifications of in Q1. There's a lot of supply that was pushed into the market, I think, into channels that are not necessarily tied to revenue quickly. That's partly because people were buying things in anticipation of maybe supply disruptions that didn't happen. I think those factors have actually played into the quarter-over-quarter compare, and may even be a little bit part of underutilization as well, just as we try to tie those things back together.
As you know, we like to build only what the customers absolutely need, and those disruptions have impacted us. I think if I step to the very high level of the industry and look at revenue per terabyte, you can see the revenue per terabyte's going down quite a bit. Some of that's the transition to higher capacity drives. Some of it's the fact that the cloud is still not fully turned on, you can see that competitive progress there. How do we get out of it and when, as to your question, that's when we can go drive cost per terabyte. The biggest product that we have coming there that's going to be impactful is 16 as well. I'll let Gianluca do the walk as well.
Yes. Okay. In terms of underutilization cost, in SQ4, we still had about 100 basis points of gross margin that we lost due to underutilization. The overall gross margin was fairly well aligned to our guidance, and as you know, even in EPS, we are actually higher than the midpoint of our guidance. I don't know if the expectation was to have a much higher gross margin for SQ4. In SQ1, you will still have some underutilization cost impacting the gross margin. Less than in SQ4, still probably 50, 60 basis points. As Dave said, it's not really related to the 16 terabyte specifically, but we are adding capacity to our manufacturing capabilities because of our expectation of much higher volume coming in the next few quarters. Until we ramp all our production and fulfill the factories, we will have some underutilization costs.
As Dave said, there is also some pricing pressure in the market that is keeping gross margin maybe a bit lower than what you were expecting. Maybe let me take the opportunity to talk about also EPS for SQ1. We guided $0.90 ±5%. There are several items that will impact EPS in Q1 in different directions. First of all, we will have the positive impact of higher revenue at a similar gross margin as we guided. We also have a positive impact from excluding share-based compensation starting SQ1 in order to be better aligned to our competitors and the normal practice in the industry. Then we have a couple of negative impact. One is OpEx. Because SQ1 is a 14-week quarter, we will have about $20 million-$25 million higher OpEx in the quarter.
We also have higher OpEx due to the variable compensation, so that is probably another $20 million. Finally, we also have lower interest income because we redeemed our investment in Toshiba. As you know, that investment was generating about $20 million of interest income that we will not have in FQ1. This is how probably you can reconcile the $0.86 in FQ4 to the $0.90 in FQ1.
That's really helpful. Thank you.
Yeah, thanks.
Your next question comes from Steven Fox on Cross Research. Please go ahead, your line is open.
Hi. Good morning. Sorry for the background noise. Just one question. As you ramp 16 terabytes in the interim, is there a meaningful share loss at the cloud that we should consider? If so, can you sort of give us a sense for how that sort of plays out over the next few quarters? Thank you.
Thanks, Steven. The way I think about it is we're ramping 16s and communicating to our customers what we want to do on 16s. That's the platform. We didn't just stage this platform last month. We've been working on it for years. That's the platform we've been out selling to our customers, getting them to align to, getting them aligned on the ramps and so on. To your point, especially when the market is relatively soft for nearline, we've been down. It's actually starting to pick back up as we talked about, but it's still not up to full speed. You can see that versus where we were, say, a year ago, when things were really hot. We want to make sure we don't push the wrong drives out there then.
From our perspective, let's not overbuild, say for example, on 12s or 10s and push those into slots that ultimately the customers maybe they don't want for their long-term TCO proposition. They're going to be putting these data centers up and running drives for five, seven years. The TCO proposition for the 16 is huge. We don't want to be temporarily going after that. If you call that share loss or something like that's fine. That's not a metric we're really managing on, to your point.
Thank you.
Yep.
Your next question comes to the line of Aaron Rakers from Wells Fargo. Please go ahead, your line is open.
Yeah, thanks. Two real quick questions if I can. First of all, just kind of trying to understand the gross margin trajectory from here. Can you help us appreciate at what level of capacity shipments on a quarterly basis you think that you kind of fully absorb the underutilization of your fabs? I mean, is that north of 100 exabytes? I'm just trying to understand or frame that as we kind of build the model, and I, again, do have a follow-up.
Yeah, that's good, Aaron. I think it's a good way to think about it. This time last year, we were north of 100 exabytes and then we've dipped down into the 80s in the last couple of quarters when things have been soft. I think we need to get up over 100 now and we're installing capacity for that. Largely a portion, it's not just 16 terabytes, right? It's some of the low cap stuff moving to eight terabytes and some of the surveillance markets, the edge storage markets moving to four and eight terabytes and so on. Exactly to your point. Those are the utilization targets that we've got set, and we think it's going to come, so we're staging for it. Yep.
Okay. Just not to read too much between the lines, but last quarter, I think you said that the nearline capacity shipments for fiscal 2020 may exceed 35 to 40. Now you're saying, quote unquote, "well above" that range, that long-term margin. Can you help us understand or maybe define what well above means?
Yeah, I think that goes back to deep collaborative work with our customers and talking about what exactly they need, when they need it, and making sure that our ramps are big enough and flexible enough to be able to accommodate their needs. It goes without saying that 16 terabytes and above, 18 terabytes when we get there, 20 terabytes when we get there, are very meaningful TCO improvements for the customers. Depending on which ones, they may be cycling out old equipment, they may be building new data centers as part of their plans. All of that improves their capability and they're going to be running the gear for a long time. We think we're up against a fairly big bubble this time in exabytes. Yeah.
Okay. Thank you.
Your next question comes from the line of Christopher Muse from Evercore. Please go ahead, your line is open.
Hello?
C.J., are you there?
CJ?
Yeah. Can you hear me?
Oh, now we can.
Sorry about that. I guess to the September model, I'm having difficulty hitting the numbers. I guess, could you provide guide for OpEx and gross margins, including stock-based comp?
No, I don't think we will do that, but in my script I said that the impact is about $30 million, so you can calculate that fairly quickly.
I should be thinking 30 plus 20 plus 25, so $75 million higher OpEx Q on Q, including stock-based comp?
No. The OpEx part of the stock-based compensation is about $20. You have $20 million that is not included in our guidance because of the stock-based compensation. The 14 week, so the extra week in the quarter, is a similar amount, so $20 million. Because we enter into a new fiscal year, in the guidance, there is an assumption for variable compensation that is higher than what we had in the prior quarter. You need to add all those items.
Okay, thank you. As a follow-up, can you speak to, as you think about gross margins, and I know you're focusing on underutilization, but is the 16 terabyte transition having any impact there? At what point should that be mitigated?
We are ending the 16 TB. As you know, the expectation for demand is really strong, but is a time lag between when we start the capacity and when we can really run the production. Which is why we still have some underutilization costs. Well, depending from how much additional capacity we will add in the next few quarters, but we expect to be close to being at full capacity, fairly quick, probably a couple of quarters.
Yeah. I think you can start to see that, C.J., in our CapEx numbers a little bit, if you look year-over-year, as we're staging the right technology to be able to get to that. That's one of the reasons why we reinforced the expectation for revenue growth in FY 2020 as well, is because we're getting to the point where we believe that those TCO propositions are so advantageous that people will stretch for them.
Thank you.
Your next question comes from the line of Ananda Baruah from Loop Capital. Please go ahead, your line is open.
Hi, good morning, guys. Thanks for taking the questions. Dave and Gianluca, I have two, if I could. First is just sticking with gross margin. Dave, the metrics around utilization, how to get the margins going, what you're shooting for, the 100 terabytes, that's really helpful. My question in that regard is, last cycle, when you hit that 100, the gross margins, I believe, were close to 32.5%. Could you just give us a little more context on, is that the ultimate ceiling again? I guess just around the sliding scale and the 100 terabytes, and how we should think about what the ceiling could be this time. I have a quick follow-up.
I'm glad you asked that. If I think about gross margin %, we don't manage the business on a day-to-day perspective for gross margin %. It's a long-term planning item. When we say last cycle, gosh, it was only a year ago. It feels like these cycles are going very, very quickly. We're investing to be able to hit the peaks of those cycles better, to your point. I think there's competition as well, which is to my commentary about revenue per terabyte. We need to get cost per terabyte down, but we also have to realize that revenue per terabyte is coming down fairly aggressively as we move. All these things factor in.
Over a longer period of time, the margin range serves as a guide for how much we want to invest, where we think we're going to go. I think if you look over the entire fiscal year, to your point, you saw a peak and a valley, if you will. We think there's another peak coming as well, exactly to your point. I think if we get to the top end of the range again, we earned it. I would also always ask the team, though, kind of end quarter, if we have the opportunity to go grab dollars, even if they're dilutive to gross margin %, we'll take it tactically. Does that make sense?
Yeah, it does. It does Dave. That's really helpful. I appreciate it. Then the follow-up is just with regards to 16 terabyte kind of quals and progression. I believe sort of 90 days ago, or at least 60 days ago, let's say, you guys were expecting to get the volume September quarter, in some context in 16 terabytes, and then to really sort of see things kick up in the December quarter. It sounds like you're still expecting that in the December quarter. How is the progression relative to prior expectations through the September quarter? Appreciate it.
Yeah, we're on a fairly aggressive ramp. Remember that the lead times for things like heads and disks and drives are getting longer, especially on these big capacity drives. We're still driving it. The qualifications really don't have any significant technical hitches at this point. Some have timed out. There's a few that are, for various reasons, customers have pushed a few weeks because their tools weren't ready or because they're not ready to intercept where they want to be able to take a 16 terabyte. A lot of that is about where they want to go. As far as I'm concerned, we're pretty happy and we're definitely staging materials.
Appreciate the feedback.
Ananda, you are correct. You will see a big ramp in the December quarter. We are very active already in this quarter, but you will see much more volume starting next quarter.
Thank you.
Your next question comes from the line of Mark Delaney from Goldman Sachs. Please go ahead, your line is open.
Yes, good morning. Thanks for taking the questions. First, I was hoping to follow up more on the commentary around revenue per bit. Dave, you spoke about dual actuators and improved performance. How do you think that translates into your ability to improve pricing? What sort of price premium do you think you can get for that type of technology? I assume there's some sort of additional cost, so kind of related to that, what would the gross margin implications be as well?
It's an interesting space, Mark. I don't think it's near term. Just to be quite frank, I think there are some smaller customers who have very high-performance workloads who are really pushing for this technology. There are some smaller divisions of cloud service providers, because everything, all cloud service providers are not created equal, and many have different workloads. Where this technology is immediately applicable is a subset. I think the technology over time becomes much more important. I would think about it as above 20 terabytes. You just can't continue to have bigger drives all behind one actuator at a relative IOPS per terabyte streaming speed that's less, if that makes sense. We need to go to this technology. I think it'll be competitive, and I think we're being driven very hard for it.
I also think since we're providing so much more value, if we happen to out-compete, it should be good for us. I think the other interesting thing about the technology is everything I just said about the cloud, it's very applicable back into the edge data centers, which are right now starved for that value, I think, from a lower cost per terabyte perspective, but also from a performance perspective. The rest of the edge is also. If we talk about surveillance drives or something, we're getting driven for multiple streams in surveillance drives. The technology is probably relevant there too, but I just want to be very upfront and say that the first instantiation now that we're shipping these products is going to be the early adopters, if you will, are going to be more niche-y for a while.
That's helpful. My follow-up question, during the past quarter, the company announced the EVP of sales will be leaving later this year. Can you talk about how Seagate plans to fill that sales position, and are you contemplating making changes in how you're going to market going forward? Thank you.
Yeah. Thanks for the question. It's interesting. No, I won't talk about any individual staffing that we'll do. I will say that are we planning on changing how we go to market? It's really interesting. The markets are changing very quickly. The customer types are changing. Look at the customers we just even talked about on this call versus two or three years ago. Exactly to your point, there's a lot of changes going on. The Seagate team is pretty deep, I think as everybody knows, been together for a long time, and I have a ton of faith in the rest of the team. I really think we've made great transitions in the last three years, and some with Jim's help. I think going forward, we're going to have to rely on the growth of the Seagate team.
From my perspective, I am very focused on who are the new customers, how are they going to buy, how do we adapt to them, and then some investments we're probably going to have to make.
Your next question comes from the line of Karl Ackerman from Cowen. Please go ahead, your line is open.
Hey, good morning. Dave or Gianluca, clearly your PC exposed drives have decelerated the last few quarters, presumably following the normalization of NAND ASPs that I think make it a little bit more economical for SSDs in those environments. In the past, you've counteracted that headwind by raising the density per drive while using only one platter. I guess from here, how should we think about the levers you can pull on the cost side to stabilize that business? Thank you.
Yeah, Karl. Thanks. It's interesting. We said in the script that, I don't know if you caught this, we talked about how we're doing minimal investments in some of these spaces. I think what's important is we're managing the business for free cash flow over the long term, not over the short term, right? Some of these markets are still, with minimal investment, the free cash flow is still quite good. Operating margin's still quite good. As a matter of fact, I would argue that today, the competitor, if you will, as you just talked about, being flash drives or something like that, are not as good. There's a reality of the market space. We're not really investing a lot in those spaces. We'll continue to run them over the long haul and think about them as how do we generate free cash flow?
There will be disruptions in some of those spaces, and we forecast that over time, but I think we have to be careful because the tail is actually quite long as well. I'll just pick on PCs for a second because a lot of people like to talk about it. From my perspective, from a hard drive perspective, PCs, the interesting one for hard drives already have dual drives in them. It's actually a longer tail, and there's a reason why the HDD is in there and the SSDs in there. I don't really think of it as per se an either/or. The other thing to keep in mind is we're out to service our customers, and our customers kind of dictate the demand. We don't make demand by our strategy. We don't think of it that way. If that helps you.
If I can add something on free cash flow. I think it's very important to keep in mind that even during a quarter that is a down cycle quarter, Seagate was able to generate $300 million in free cash flow, and was very similar last quarter. I think it's a change compared to what was in the past. Even during a down cycle time, the focus on free cash flow is giving very good results.
Appreciate the color. Thank you.
Your next question comes from the line of Mehdi Hosseini from SIG. Please go ahead, your line is open.
Yes. Thanks for taking my question. Dave, from a big picture, it's very interesting and supportive, you having confidence in your enterprise exabyte shipment for FY 2020. Can you put that in context and give us some framework as to how the overall exabyte would grow FY 2020 versus FY 2019? I have a follow-up.
Okay. Sorry, I didn't catch the very last part of that, Mehdi. Could you just repeat just the last sentence?
Yes. Sure. I was just going back to your enterprise exabyte growth of well over 35%-40% for FY 2020. How does that impact your overall exabyte shipment? Can you put the overall exabyte shipment in the context, given how comfortable you are with the enterprise segment?
Oh, I see. Okay. It's becoming a bigger and bigger portion. I think in the script we talked about it already being 50%. From my perspective, it's going to be a bigger and bigger portion. The leverage that we get, because we're doing fewer drive types than before, I think we'll get relatively better payoff. If I think about the large part of the driver for exabyte growth, it's the fact that we're getting higher capacity points, 16 versus 10 in the last peak of the last cycle. I'm not saying that the hard drive size is the primary driver of that. I think there are many drivers, but I think that's the biggest thing driving us near term. Longer term, to your point, if I think about enterprise, the cloud will continue to grow.
The cloud will continue to cycle through some of their existing footprint and upgrade as well. I think data is still going to grow in the cloud. I think the other interesting topic that we have going on right now is that in the on-prem data centers, data is being repatriated, but there's a marked difference in the cost between the cloud and the on-prem. What we're seeing is a lot of people in the on-prem arena have focused on high-performance storage, and that's fine. There's going to be a lot of needs for compute there. The high-performance storage or high-performance memory, if you will, needs to be very close to that compute. With the extreme data growth that's going on, in order to have enterprise growth in those places, you're going to need cost-effective solutions vis-a-vis the cloud. We see a great opportunity there.
We talked about it a little bit in the script. I think that's also an opportunity for us to get to market a little bit faster with these same technologies, that will drive exabyte growth.
If I may refine my question. If your enterprise exabyte is growing over 35%, 40%, your surveillance consumer electronic exabyte growing at a faster rate, would those two help you with double-digit total exabyte growth?
Yeah. Maybe back to Karl's question, eventually they'll take over from some of the more legacy systems from an exabyte perspective, exactly to your point. What we've seen about some of that edge storage, like surveillance, is it's been a little choppy like the cloud has, and unfortunately, sometimes they phase up, and you don't see it. I don't know it's easy on a rolling one, two, three-quarter basis to draw any trends. Certainly over the last three or four years, if you start drawing the cloud trend and the edge storage trends that are around surveillance, then it presents a pretty good story.
Got it. Thank you.
Your next question comes from the line of Jim Suva from Citi. Please go ahead, your line is open.
Thank you very much. Thus far, your answers have been very useful. I just wanted to make sure I heard and understand correctly on the gross margin and underutilization. The June quarter, it was about 100 basis points, and the September outlook about 50 to 60 basis points. You're going to be adding and filling the more capacity with more volume, so it should be normalizing pretty quickly after the September quarter. Is that the way to think about the impact past and forward?
Yes, I think that's the right way. It all depends how much capacity we will add in the December quarter and the following quarters. I think from a modeling standpoint, you are correct.
My last question is just knowing the cycle time or the throughput of your production, I would assume the materiality of the revenues kind of come in probably after the December quarter because you simply don't turn on the fabs and they come out perfect right away. Probably more like the March quarter as opposed to December quarter? Do you think December quarter will be the full run rate of the revenues coming out of your increased capacity?
Well, it is difficult to guide revenue right now for December or later quarters. Of course, it depends from how much we ramp in 16 terabyte, and as we said before, we will continue to ramp into the December quarter and after that. You will have probably more impact in the next few quarters.
It's a good way to think about, though, with lead times, because we don't know exactly what's going to happen right now on when people will pull. Relative to what we're staging from a materials perspective, we're staging those parts with that, if that makes sense.
It does. Thank you so much for your details and questions. It's greatly appreciated.
Your next question comes from the line of Sidney Ho from Deutsche Bank. Please go ahead, your line is open.
Great. Thanks for taking my question. You talked about improving demand condition, especially for hyperscale customers. Can you give us some color as to how broad-based that strength is? Given your exabyte for Nearline is still about 20% below a year ago, when do you think that on the exabyte basis that Nearline drives could get back to year-over-year growth? Thanks.
Year-over-year growth, I think is coming certainly in FY 2020. Some of it depends if we talk about just on exabyte growth, some of it depends on the specifics of the ramp of the 16, and I don't want to get any further ahead than next quarter. What I would say is you're on the right point, which is last year, the drives, the factories were full Q4 to Q1. This year, as we're ramping right now, we're definitely staging to be able to capture the peak of that cycle again. I think we make that available to guide the eye, if you will, on our investor relations website to show those cloud cycles. What we've seen over time is that the peaks and valleys, if you will, of the climb in the cloud, have really been fairly predictable.
Things can always get thrown off just a little bit, but we believe there's another one coming, and it certainly consistent with the discussions we're having with our customers.
Yeah. Yeah, we expect a very good increase in exabyte already in FQ1 for the Nearline.
Okay, great. Thanks.
Your next question comes from the line of Munjal Shah from UBS. Please go ahead. Your line is open.
Yes. Hi. Thanks for taking my question. Last quarter you mentioned that the Nearline ramp would be wider and higher. Do you still expect that, or do you think it's going to be coming back much stronger as we go through next cycle?
Yeah. Let me say it this way, the data center buildouts that we've typically heard of, and even all the way extending back to last August and September, some of the data centers we'd hear about being built geographically are still in the plans in a lot of places. They've just been postponed for various reasons. There are other people who you can tell that data is growing against their application, and they want to continue to invest, but they wait for the right architectural decisions. Sometimes it's the hard drive capacity point. Sometimes it has to do with other architectures that are going on. It's hard to paint the cloud with a uniform brush because there's so many different applications and strategies that are going on. I do think the overall data growth is very consistent, and that's what drives that periodicity that we're referring to.
A little wider and deeper this time, maybe. I think it certainly has felt like that in the last six months. I think what I said a year and a half ago was that geographically, we were starting to see enough diversity that maybe it wouldn't be as deep. Clearly, the markets have been fairly disturbed in the last six months. I think that overall data growth, the demand for places to put the data is still there and driving that trend.
Just to follow up. When we see the next ramp, are we seeing from existing applications, or do you think those build outs that were pushed out are starting to happen, like the new data center build outs?
Do you have visibility there?
That's certainly true. There are new applications coming online. I wouldn't talk about any specific customers of course, but yeah, there are new applications.
All right, great. Thanks a lot.
Your next question comes from the line of Mitch Steves from RBC Capital Markets. Please go ahead. Your line is open.
Hey, guys. Thanks for taking my question. I just had one just to talk about a little bit of FY 2020. I know you guys aren't going to give exact guidance, but given that this year has been a little bit of a strange year in terms of the first half being a little weaker in terms of calendar year, when we look at 2020, should we assume that it's going to be more of a typical seasonal year for you guys? Or is there anything else we should be aware of? It sounds like maybe Q4 would be more seasonally strong. Just looking for any sort of high level comment in terms of that seasonality next year.
Yeah, interesting. I think, Mitch, there's still seasonality in some parts. Like for example, consumer is still very seasonal. As we have less and less exposure to things like PC, some of the traditional seasonal spikes that we would see that were very predictable are not there as much, and as we talked about earlier, some of the PC design points are changing a little bit. The seasonality may be changing. The cloud and surveillance markets. Surveillance used to have a little bit of seasonality. I think it's been a little disrupted. I would say it's aseasonal, and the cloud certainly is aseasonal. It goes through different patterns that cause the spikes that we talk about.
With all these things considered, though, just looking at the total data growth over the last few years, we think FY 2019, exactly to your point, people have been very conservative and in FY 2020 they're going to have to go invest in data, and that's why we have confidence in revenue growth.
Perfect. Just one last follow-up just on the enterprise side, and we saw what NetApp posted in terms of negative pre-announcement. I guess, is there anything there that was surprising to you guys in terms of their comments, or do you guys think that they may be more company specific? Anything you'd give in terms of why their miss was so much more significant versus what you're talking to in terms of demand?
I wouldn't talk about a specific person. I would say it's super interesting to me what's going on in on-prem enterprise. I think if I look over the last five to 10 years, there have been people very focused on high-performance rigs. That's important. We see it in our own data centers that we have to build. We also see a lot of people want to grow the data on-prem for themselves, whether it's their own control, their own application control. Some people talk about repatriation. I don't think it's a good way to think about it, because I think the cloud will grow substantially. The applications of the cloud have a great value proposition as well.
I do think that on-prem, low cost, efficient storage to cover the entire life cycle of data, not just the compute, but the life cycle of the data, is super important. I think some companies have been very focused, and that's their business model, to be very focused on high performance. I think there's an opportunity for all of us, for everybody, in this more economical on-prem stuff. We pointed to that a little bit in our script, and I think it's a space to watch in the next five years.
Very helpful. Thank you.
Thanks.
Your next question comes on the line of Vijay Rakesh from Mizuho. Please go ahead. Your line is open.
Hi, guys. Just between, on the hyperscale side, there's been some confusion in terms of, as you look at the back half, is there a difference in demand pickup between enterprise and hyperscale? What you're seeing geographically in terms of nearline demand picking up between U.S. or your Chinese customers? Do you see any color there?
Yeah, it has been a little choppy this year, that's for sure. I think there's various reasons for that, but overall, most of the discussions we're having with our customers are, there's a lot more planning involved. Is there a data center going to be built, or are you going to be transitioning some of your old gear into new gear or new applications coming online to that point? We do feel that the last six months or nine months, geographically, there have been a lot of people just say, "I'm on hold. We'll come back to this." Some of those business models are still desired in place, and then there may even be new ones coming up, which I think causes some of this choppiness that we see in the nearline exabyte demand.
All right. I know this is very recent, but in terms of some of these tariffs that are again going back in place in a month's time, any thoughts? I know this is very preliminary, but thanks.
Sorry, I didn't catch that, Vijay.
Just in terms of some of the tariffs that were announced yesterday.
Oh.
Oh, yeah.
Yeah. Okay, thanks.
We don't expect any impact from the new tariffs.
From the new ones.
Yeah.
I think from our perspective, there's a lot of things that we're obviously working with our customers through. We tend to focus on, do we have the right stuff in the right place at the right time? We react to these things just like everyone else does. I think we have pretty robust supply chain that we can react quickly. From the new tariffs, I think there's minimal impact. Everybody's analyzing the same things in the world, and going through the same things. Seagate's markedly similar to everyone else, and I've heard their earnings calls. You can tell other people are struggling with it a little bit more, but I think we're dealing with it.
There are no further questions at this time. I'll now turn the call back to Dave Mosley for closing remarks.
Okay. Thanks everyone for joining us today in Dublin, thanks for your interest in Seagate. I'd once again like to thank our customers and our suppliers and business partners and all of our employees for their contributions to our fourth quarter performance. I'd also like to thank our shareholders for your ongoing support. We look forward to seeing you all at an analyst event in New York on September 19th. Thank you, Kelly, also for hosting the call.
This concludes today's conference call. You may now disconnect.