NetApp - Earnings Call - Q2 2026

Transcript

Speaker 2

Good day, and welcome to the NetApp second quarter of fiscal year 2026 earnings call. All participants will be in listen-only mode. Should you need assistance, please signal a conference specialist by pressing the star key followed by zero. After today's presentation, there will be an opportunity to ask questions. Please note, this event is being recorded. I would now like to turn the conference over to Kris Newton, Vice President, Investor Relations. Please go ahead.

Speaker 1

Hi everyone, thanks for joining us. With me today are our CEO, George Kurian, and CFO, Wissam Jabre. This call is being webcast live and will be available for replay on our website at netapp.com. During today's call, we will make forward-looking statements and projections with respect to our financial outlook and future prospects, including without limitation, our guidance for the third quarter and fiscal year 2026, our expectations regarding future revenue, profitability, and shareholder returns, and other growth initiatives and strategies. These statements are subject to various risks and uncertainties, which may cause our actual results to differ materially. For more information, please refer to the documents we file from time to time with the SEC and on our website, including our most recent Form 10-K and Form 10-Q. We disclaim any obligation to update our forward-looking statements and projections.

During the call, all financial measures presented will be non-GAAP unless otherwise indicated. Reconciliations of GAAP to non-GAAP estimates are available on our website. I'll now turn the call over to George.

Speaker 0

Thanks, Kris. Good afternoon, everyone. Thank you for joining us. We delivered a strong Q2 with revenue of $1.71 billion, up 3% year over year. Excluding the divested Spot business, total revenue was up 4%. All flash and public cloud, which address growth markets and carry higher gross margins, made up 70% of Q2 revenue. This shift, combined with our continued operational discipline, has enabled us to drive profitability metrics higher. Our gross margin set a Q2 record and exceeded our guidance range. Both operating margin and EPS surpassed expectations and marked all-time highs. Expected softness in U.S. public sector revenue was offset by growth in all other geographies. We saw strong demand for our AI solutions, first-party and marketplace cloud storage services, and all-flash offerings.

In the age of data and intelligence, customers are choosing NetApp for our unified data platform that delivers exceptional value and operational efficiencies, fueling our success in the face of the ongoing uncertain macro environment. In October, we hosted our annual customer conference, NetApp Insight, where we unveiled major advancements to our enterprise-grade data platform, including enhanced AI workload capabilities, stronger cyber resilience, and deeper AI integrations and data solutions with our hyperscaler partners. Customers and partners shared how NetApp is driving their success in the age of data-enabled intelligence. Their feedback and achievements underscore our commitment to innovation and delivering value in a rapidly evolving landscape. We launched AFX, an ultra-scalable, extreme-performance, disaggregated storage platform certified for NVIDIA SuperPOD, designed to power demanding AI workloads and AI service providers.

AFX seamlessly integrates into an organization's hybrid multi-cloud data estate with the proven enterprise-grade data management and security features of ONTAP. We also introduced the NetApp AI Data Engine, an end-to-end AI data service integrated into ONTAP. The AI Data Engine, referred to as AIDE, simplifies data discovery, querying, searching, and analysis. It helps operationalize and scale data pipelines for AI with integrated data discovery, curation, policy-driven guardrails, and real-time vectorization. This enables fast data access, efficient transformation, and trusted governance. Together, AFX and AIDE transform how enterprise customers achieve positive AI outcomes by accelerating data discovery and simplifying data pipelines while maintaining security, access controls, and data integrity. Native integration with leading AI platforms, including Domino, NVIDIA, and Informatica, enables compatibility with enterprise workflows.

Our zero-copy caching and native cloud connectivity help organizations unify data and apply advanced AI capabilities across any site, cloud, or model, speeding time to insight. We also enhanced our rapidly growing Keystone Storage as a Service for enterprise AI, offering AFX and AIDE under a single subscription for elastic scaling and usage-based billing, encouraging broader enterprise AI adoption. These innovations build on our growing success in AI workloads. In Q2, we closed approximately 200 AI infrastructure and data lake modernization deals across diverse geographies, industries, and use cases. Our massive install base of unstructured data, advanced data and metadata movement services, industry-leading data security, and unique hybrid multi-cloud capabilities make us the clear choice for enterprise AI. Here is an example of why we are winning in enterprise AI deployments. A global semiconductor capital equipment manufacturer selected NetApp to unify its enterprise AI data foundation across on-premises and cloud environments.

Our hybrid multi-cloud data visibility and secure governance drove confidence in the compliance and operational efficiency of their AI workloads. We enabled them to create a single searchable view of corporate knowledge across millions of documents, emails, and engineering data sets, providing employees with faster, more accurate access to institutional knowledge. A large amount of AI innovation takes place in the public cloud. ONTAP is the only unified data platform natively integrated in the public cloud, putting us in a unique position to enable customers to leverage any of the major AI models without the complexities of moving data. In Q2, we expanded our native AI capabilities in Azure and Google Cloud, adding to what is already available in Amazon Web Services, giving customers the flexibility to run AI workloads wherever they choose.

Adding to existing multi-protocol support in AWS, we launched support for block storage capabilities in Google Cloud NetApp Volumes in Q2. This brings the full power of ONTAP to Google Cloud with high performance, unified storage, integrated data management and protection, and a common cloud control plane. We introduced new capabilities in Azure NetApp Files, including single file restore and a flexible service level for independent scaling of throughput and capacity. In Amazon FSx for NetApp ONTAP, we announced support for Amazon Elastic VMware Service, enabling secure, efficient migration of VMware workloads to AWS. By continuously adding new functionalities to our public cloud storage services, we are broadening our addressable market, driving new customer acquisition, and positioning ourselves for continued growth. This strategy has yielded rapid expansion in our highly differentiated first-party and marketplace cloud storage services, with revenue increasing approximately 32% from Q2 a year ago.

In Q2, a leading cloud-based media production company selected FSxN as its standard for file and block storage. In addition to multi-protocol support, FSxN delivered cost savings through storage efficiency, high availability, superior multi-tenancy, and intelligent caching to put data closer to its users. The company believes that FSxN gives it a competitive advantage in optimizing cloud storage and enhancing performance for its customers, crediting these capabilities for a key customer win. Customers are choosing NetApp to provide a unified, cyber-resilient, and efficient way to manage their entire data estate. Built for the age of data-enabled intelligence, the NetApp data platform redefines what a modern enterprise foundation should be: unified, enterprise-grade, intelligent, cloud-connected, and ecosystem-ready. We help organizations modernize, secure, transform, and use AI with confidence.

Continued strong customer engagement and interest in our unified and block-optimized all-flash storage portfolio delivered 9% year-over-year growth in all-flash array revenue to $1 billion in Q2, or an annualized run rate of $4.1 billion. Exiting the quarter, approximately 46% of install-based systems under active support contracts are all-flash. NetApp helps customers confidently safeguard their data with built-in security through real-time threat detection, protection, and recovery. In Q2, we enhanced the NetApp data platform's industry-leading cyber resilience by launching the NetApp Ransomware Resilience Service for both structured and unstructured data. This service is designed to stop cyber threats before they cause extensive damage by proactively detecting data breaches in real time and providing isolated environments for safe, clean data recovery. Our industry-leading cyber resilience capabilities are helping us win new customers and displace competitors.

In Q2, a major Asian life insurance company chose NetApp for its mission-critical private cloud environment, replacing its long-standing storage vendor. Ransomware protection was a top priority, and our ability to provide strong cyber resiliency for critical workloads was a key factor in the decision to choose NetApp. We also announced the latest version of Storage Grid, with new capabilities designed to enhance AI initiatives, improve data security, and modernize organizations' data infrastructure. Many customers begin their AI journey by updating data lake environments, and Storage Grid object storage delivers the optimized performance, intelligent data management, and modern cloud integrations needed to manage massive data sets. In Q2, a leading financial services company selected Storage Grid to modernize its legacy Hadoop environment with capabilities for a hybrid architecture featuring data durability, a global namespace, robust security, and automated zero-intervention backup and disaster recovery.

Storage Grid addressed the company's next-gen AI workload requirements. In summary, strong execution and operational discipline delivered an outstanding second quarter. Our focus on growing markets, all-flash, public cloud, and AI continues to yield top-line growth. The substantial innovation we introduced this quarter extends our differentiation and helps solidify our leadership position as the intelligent data infrastructure company. Looking ahead, we are focused on leveraging our alignment to customers' top data initiatives and pressing our significant competitive advantage. Despite the unsettled macro environment and near-term US public sector headwinds, we remain confident that our visionary approach to a data-driven future will enable us to outgrow the market and capture additional share. I'll now turn it over to Wissam. Thanks, George, and good afternoon, everyone.

As George mentioned, in the fiscal second quarter, we delivered strong results, exceeding both the midpoint of the revenue guidance range and the high end of the EPS guidance range. Total revenue for the quarter was $1.71 billion, up 3% year-over-year. Non-GAAP earnings per share was $2.05. Excluding the divested Spot business, which generated $23 million of revenue in the year-ago quarter, total revenue was up 4% year-over-year. The effect of foreign currency exchange rates was favorable to revenue growth by approximately one percentage point year-on-year, while it was immaterial relative to guidance. Looking at revenue by segment, hybrid cloud revenue of $1.53 billion was up 3% year-over-year, driven by product, support, and Keystone. Keystone continues to show great progress with growth of 76% year-over-year. Public cloud revenue of $171 million increased by 2% year-over-year.

Excluding spot, public cloud revenue was up 18% year-over-year, driven by strong demand in first-party and marketplace storage services. At the end of the second quarter, our deferred revenue balance was $4.45 billion, up 8% year-over-year and 7% year-over-year in constant currency. Remaining performance obligations were $4.9 billion, growing 11% year-over-year. Unbilled RPO, a key indicator of future Keystone revenue, was $456 million, up 39% year-over-year. Moving to the rest of the income statement, please note my comments will be related to Non-GAAP results unless stated otherwise. Gross margin for the fiscal second quarter was 72.6%, above our guidance range. Sequentially, gross margin was up 1.5 percentage points. Gross profit was $1.24 billion, up 4% compared to Q2 2025. Hybrid cloud gross margin was 71.4%, up 1.4 percentage points sequentially due to product gross margin improving by 5.5 percentage points to 59.5%.

Our support business continues to be highly profitable at 92.1%. Professional services gross margin was 30.3%, improving 40 basis points sequentially, driven by higher Keystone revenue mix. Public cloud gross margin was 83%, up nearly 3 percentage points sequentially and over 9 percentage points year-over-year. Operating expenses of $707 million were flat sequentially and down 2% year-over-year, despite the unfavorable effect of foreign currency exchange rates. Operating income was $530 million, up 12% compared to Q2 2025. Operating margin was 31.1%, up 2.4 percentage points year-over-year, driven by higher revenue and gross margin combined with lower operating expenses. Earnings per share was $2.05, growing 10% year-on-year. Both operating margin and EPS exceeded the high end of our guidance ranges. Our results demonstrate strong execution on key growth opportunities in all-flash, public cloud, and AI, as well as continued focus on operational discipline.

Cash flow from operations was $127 million, and free cash flow was $78 million. During the second quarter, we returned $353 million of capital to our shareholders, with $250 million in share repurchases and $103 million paid in dividends of $0.52 per share. Q2 diluted share count of 202 million decreased by 8 million shares, or 4% year-over-year. At the end of the quarter, cash and short-term investments were $3 billion, and gross debt outstanding was $2.5 billion, resulting in a net cash position of approximately $528 million. I'll now turn to Non-GAAP guidance, starting with Q3. We expect revenue of $1.69 billion, plus or minus $75 million. At the midpoint, this implies a growth of 3% year-over-year. Excluding the divested spot business from the year-ago comparison, our revenue guidance implies a 5% growth. We expect Q3 gross margin of 72.3%-73.3%.

Operating margin is anticipated to be in the range of 30.5%-31.5%. We expect EPS to be between $2.01-$2.11, with a midpoint of $2.06. Turning now to full year 2026, we continue to expect fiscal year 2026 revenue to be between $6.625 billion and $6.875 billion, which at the $6.75 billion midpoint reflects 3% growth year-over-year. Excluding spot, our revenue guidance implies a growth of 5% year-over-year. Based on our Q2 performance and the confidence in our outlook for the second half, we are raising gross margin, operating margin, and EPS ranges for the fiscal year. We now expect our fiscal year 2026 gross margin to be in the range of 71.7%-72.7% and are increasing our operating margin to 29.5%-30.5%. We expect other income and expenses to be approximately negative $15 million.

For the year, we expect a tax rate in the range of 20.2% to 21.2%. We are raising our EPS range to $7.75-$8.05, with a midpoint of $7.90. In closing, as we look ahead to the rest of fiscal year 2026, our commitment to executing our strategy remains strong. We are poised to seize the expanding opportunities in all-flash, cloud, and AI and remain focused on consistently delivering exceptional value to our customers and shareholders. I'll now turn the call over to Kris for Q&A. Thanks, Wissam. Operator, let's begin the Q&A. At this time, if you would like to ask a question, press star, then the number one on your telephone keypad. To withdraw your question, simply press star one again. We will pause for just a moment to compile the Q&A roster. Your first question comes from the line of Aaron Rakers with Wells Fargo.

Please go ahead. Yeah, thanks for taking the question. I guess the first question would be, as we think about the component environment, both from a potential pricing perspective as well as whether or not you might be seeing any kind of constraints, I'm curious how the company's managing that. Have you leaned in on any kind of strategic purchases and any thoughts on the duration of those strategic purchases as far as the next couple of quarters? How much visibility do you have on the pricing dynamics underpinning the gross margin outlook? Thank you. Hey, Aaron, thanks for the question. On the components, we did mention last quarter that we did lock in some prices. Based on that, we do have visibility for a couple more quarters, I would say, at least until the end of this fiscal year.

When you look at where we are, obviously, we did Q2 product margin was slightly better than our long-term model, which is in the mid to high 50%. Then when we look at the rest of the year, we expect product margin or product gross margin to be relatively stable to where we ended in Q2. Looking ahead, if I think of the component pricing, look, this is an environment that could be volatile. We are not going to make a call on that. What we do is we look at various scenarios. We have a very capable supply chain team that has been through many of these past cycles of tight supply and some of the commodities that we buy with rising cost environments, for instance. We have been able to manage very efficiently over the years, in fact, while also in some cases growing EPS.

We would continue to manage our input cost and maintain our supply continuity. We have not seen any disruptions so far. We have not heard of any as such. Ultimately, our goal is really to focus on our total gross margin. This comes down to the various components of our revenue and the mix. There is the rate there and the mix. If we continue to see current levels, let's say, of some of the commodities, in particular, I know you probably have in mind one of the things, for instance, is NAND. If we continue to see similar levels, we will probably have a bit of headwind into fiscal 2027 from a product gross margin perspective.

When we look at the mix of the business going forward, we continue to see high growth in the cloud business, which is now operating between 80-85%. I mean, Q2, it was at 83% gross margin. We continue to see good growth in Keystone. The mix is very much favorable to us. The last couple of points, I would say, as we think through all of this, we're focused on driving growth in gross profit dollars, which is foundational to the profitability engine of our business. Lastly, if we are faced with higher commodity prices relative to where we are today, we will always reconsider our pricing. Commodity prices are typically passed through for us, and we do not have an issue passing them through. Yeah. Wissam, I appreciate that. That was a very thorough answer.

Looking back historically, how quickly could you pass those through? Is that within a given quarter? Does it take a couple of quarters to see that flow through? Look, this is probably more of a hypothetical, obviously. As I said, where we are today, we have good visibility till the end of the fiscal year. We do not necessarily need a lot of time to pass them through. We can always adjust prices as needed. Thank you, Wissam. Thank you. Thanks, Aaron. Next question. Your next question comes from the line of Eric Woodring with Morgan Stanley. Please go ahead. Great. Thank you, guys, for taking my question. Maybe I am going to ask a similar question to Aaron and Wissam. That is, you are effectively at 60% product gross margins. I know you said we should expect them to be relatively unchanged through the rest of the year.

First part of that is just, is this a function of mix to all-flash? Is this still pricing tailwinds? Is this kind of bomb cost down? I'd love if you could just maybe contextualize what the real drivers are of that product gross margin expansion. Second, I'll just ask both questions at once. I realize you don't really need to talk to fiscal year 2027. It's too early, and the commodity cost environment is quite volatile. Are we kind of at peak product gross margins? Can product gross margins expand from here? I'd just love to understand how you think about the sustainability of where we are, even regardless of the memory cycle. Thank you so much, guys. Yeah. Thanks, Eric. Hey, look, what I said, for the rest of the year, we expect product gross margin to be more or less where we are now.

We're in the 59%, I think, and change. When we think about the drivers, it's a combination of things. If I sort of look at where we are year-on-year, for instance, cost is now roughly flattish. We are seeing differences mostly driven by mix and pricing. That's pretty much what drives the margin. As I look forward, obviously, as you mentioned, it's too early to talk about 2027. I'm going to talk about theoretically how we think of the long-term product margin of our business. We are targeting mid to high 50% margin. That's really we are operating a little bit better than that now. Our target is the mid to high 50% because ultimately, to drive better gross margins for the company, it's really the mix that's going to be also a tailwind.

As our fast-growing cloud business continues growing at the high and delivering higher gross margin than the corporate average, this should help us improve on our margins. I would not put a cap on our product margin because ultimately, we continue to be very operationally disciplined in how we conduct the business. I hope this answers your question. All right. Thanks, Eric. Your next question comes from the line of Samit Chatterjee with JP Morgan. Please go ahead. Hi. Thanks for taking my questions. Maybe if I can pivot a bit more to the revenue side. George, you talked about sort of the AI-related transactions or deals that you closed. Looks like 200 versus 125 last quarter. Wondering sort of what you are seeing in terms of trends there. Looks like an acceleration on the sort of headline number of deals you are seeing.

Is the AFX solution that you launched at Insights, what sort of customer feedback are you getting? Is that playing into those deals that you're now seeing? I have a quick follow-up after that. Thank you. We are seeing a fairly stable mix of transactions. The volume is growing. The mix has stayed roughly the same, which is data prep and data lake modernization, about 45% of the mix, training and fine-tuning, about 25-30% of the mix. The rest is kind of RAG and inferencing. I think you can see the pace of kind of wins have grown. A year ago, it was about greater than 100. Now it's close to 200 or 200 approximately. We are seeing the growth in the number of wins and the mix staying roughly stable.

With regard to the AFX platform, listen, we have had a huge amount of interest in it. It's going through customer qualifications. It's too early to say that it's a big cause of the results in the quarter. It takes a few quarters for a new system and architecture to get qualified, but we're seeing strong early interest in it. Got it. Got it. Maybe a quick follow-up for Wissam here. Wissam, when I look at the three-Q revenue guide, it's roughly sort of flat, modestly maybe down quarter over quarter. I realized last year you had the same seasonality, but for most of the previous years, prior years, you typically sort of have a seasonal growth from Q2 to Q3. Obviously, it's a dynamic macro, and you have sort of public sector headwinds.

Anything specifically sort of driving this seasonality to be a bit below average relative to your prior years into Q3? Wissam, really, the couple of reasons are pretty much what you mentioned. There is a bit of a dynamic macro. We are expecting our US public sector business to be slightly below seasonality given, obviously, the most recent shutdown, and it takes a little bit of time for government to reopen. Having said that, we view these as temporary, and long-term, we expect that business to get back to normal levels. Got it. Got it. Thank you. Thanks for taking my question. Thank you, Samit. Your next question comes from the line of David Vogt with UBS. Please go ahead. Great. Thanks, guys, for taking my question. I am going to roll them into one.

Maybe George and Wissam, when I think about kind of the demand drivers going into the second half of this year, and I think you said you still expect federal to be subseasonal, how do we marry that sort of outlook and then going into 2027 with kind of how you're thinking about inventory and purchase commitments? Because your inventory on the balance sheet is still relatively modest. Doesn't seem like you took in a lot of finished goods from my quick look. I'm just trying to think how we should think about marrying maybe a recovery in some of the verticals like US federal next year. Hopefully, that's behind us. And how you're thinking about adding purchase commitments and working capital to support the business next year. Thanks. I think maybe I can start, and then Wissam can add color.

I think, first of all, if you look at the second half implied growth rate, it is up if you look at export relative to the first half. We are seeing some acceleration despite the U.S. public sector headwinds in the second half of the year. We see the non-U.S. public sector segments had a strong result in Q2. Our flash business accelerated from 6% to 9% in the quarter despite a tough compare a year ago. We feel good about momentum. We are aligned to what customers are spending. Despite the choppy macro, customers are spending on AI projects, on data infrastructure modernization to get ready for AI. They are implementing additional cyber resilience protection. We feel good about our alignment to customer spending.

I think the third thing is that the faster growing parts of our business are now a bigger part of the overall number. Cloud and flash are now 70% of the overall number. The slower parts of the business that are not growing are both smaller and also stabilizing more. That gives us belief that second half, hopefully, once US public sector gets clarified, we should have a strong set of results going forward. Yeah. David, with respect to purchasing commitments, as I mentioned, we have good visibility till the end of this fiscal year, and we would be opportunistic as we go forward. We will be obviously looking at fiscal 2027. Before we enter the year, we would want to be able to have good visibility and, in many cases, secure whatever supply we need.

We will not hesitate to take the right actions to make sure we protect our business and we have the supply that's needed. Wissam, can I take a quick follow-up to you? Yeah. Is this a couple of quarters or a full year kind of purchase commitments given the environment? What's your philosophy going into next year? I would say it all depends on what we're buying and when. There isn't really one answer that fits all. We will be very dynamic. We'll make decisions very fast. As I said in my earlier remarks, look, we have a very capable supply chain team. We monitor the commodities, and we don't hesitate to take action as needed. We do have a strong balance sheet.

If we need to take any specific action to protect our supply or to sort of lock in good prices, we will not hesitate to do that. Great. Thank you. All right. Thanks, David. Your next question comes from the line of Chris Venkara with TD Cowen. Please go ahead. Yeah. Hi. Thanks for taking my question. The first question, George, is for you. You kind of spoke about the 200 AI deals that you closed in the quarter. What is the average size of these deals? Also, where are we in the pilot-to-production journey for enterprise adoption of AI? Is there a way to think about that into 2026? I have a quick follow-up. Listen, the deal sizes are all over the map.

I think that there are some that are smaller in proof of concept, and there are others that are in scale deployments that are much larger. I would not share a particular number related to the average deal size. It would misrepresent the answer. With regard to the broad themes, I think, first of all, data lakes and data prep are usually prior to scale deployments. They are in the get ready to use AI, bring all of my data together. I think fine-tuning and inferencing are at various stages of the kind of proof of concept to production deployment pipelines. If you look at the industries, life sciences, financial services, some parts of manufacturing, and public sector are areas where we see AI being broadly adopted with the most advanced use cases in life sciences. I think that the adoption use cases depend on industry.

Got it. And then there's a quick follow-up for you, Wissam. You spoke about the first-party hyperscale services demand was very strong last quarter. Can you just quantify it? How much did that grow last quarter, that services business? The first-party and marketplace services business grew 32% year-on-year. Great. Thanks, George. Yep. You're welcome. Thanks. Thanks, Chris. Your next question comes from the line of Stephen Fox with Fox Advisors. Please go ahead. Hi. Good afternoon. I just had one question. Given all the supply shortages that you're talking about, which seem to have accelerated, I guess, in the last several months, and I understand that you've taken some tactical moves to secure supply, why do we think, or is there a risk, I guess, that you run into next year sometime?

I'm not saying first quarter or second quarter, but is there a risk that at some point that the lack of supply hurts overall demand, whether it's because of overall price of product or just your availability to ship? Thanks. I'll just clarify the comment by saying we did not suggest that we have any supply shortages. We said that we have secured supply commitments and pricing through the end of the fiscal year and will take appropriate actions to guarantee supply across a broad ecosystem of suppliers through the next fiscal year, depending on what we see. Okay. So you're confident that whatever price you're talking about a year from now, George, is sustainable in the marketplace, I guess? Listen, there's a lot of puts and takes in the bill of materials in our products. And we have a broad ecosystem of suppliers.

We work with them both to look at a variety of different price points and offerings, and we'll have the right mix. As we said, the historic pattern of this industry is that when component prices, commodity prices go up, there are pricing actions taken to share some of those price increases with customers. If it reaches that point, we have the experience to do that. Great. Thank you for the color. Thank you. Thank you. Your next question comes from the line of Tim Long with Barclays. Please go ahead. Thank you. Two, if I could as well. Hate to kill the commodities here, but just curious. A lot of talk about NAND, but HDDs are tight and going up as well. What's the impact on the business of QLC being used more than hybrid or maybe more than TLC?

Any moving parts that might actually benefit the move towards all-flash, number one? Number two, the cloud revenues have been pretty consistent, high teens, ex spot. George, you mentioned a lot of new offerings on the big cloud players. Could you just talk about that kind of ex spot growth rate? Do you think that high teens is sustainable, or can some of these newer offerings help accelerate that number even though the base is fairly large? Thank you. Yeah. On the first question, listen, HDDs actually performed well for us this quarter. It performed, in my expectation, better than we had planned. We are seeing a broad range of use cases for these products. We do not see that one type of technology replaces the other. There are use cases for TLC, QLC, and HDDs, and we are seeing that pattern in our business.

I think with regard to your question on cloud, listen, I think broadly speaking, we are super positive about the growth of our first-party and marketplace cloud services business. They have been ticking along nicely north of 30%, representing strong customer demand for our differentiated offerings. I think that the scale of the business, if you look at the Q2 quarter-on-quarter growth, it was the highest quarter-on-quarter incremental revenue in the history of the cloud business. We feel really strong about that. Broadly speaking, we have a really good set of tools for enterprise workloads on the cloud. The next big push is to capture the AI and data-intensive workloads on the cloud. You will see us bring innovations in the market. We already talked about some of them, and you will see a lot more of that over the next 6-12 months.

Super excited. This is really about scaling the go-to-market. It is almost we can bring on as many clients as we want. It's really bringing an effectively scaling go-to-market motion for our cloud offerings. Okay. Thank you. Thank you. Your next question comes from the line of Simon Leopold with Raymond James. Please go ahead. Great. Thanks for taking my question. I wanted to see how you thought about your opportunities to win business with some of the Sovereign or NeoClouds who are not necessarily hyperscale and building their own storage solutions. How do you see that fitting into your strategy, and what kind of opportunities have you seen so far there? Yeah. We are part of a large number of Sovereign clouds across the globe for many, many years. With regard to their AI landscapes, we have mentioned Sovereign AI wins.

We have expanded the range of offerings with the AFX solution that allows us to participate in a broader range of footprints within those Sovereign AI clouds. We are bringing more innovation to the market over the next few months. Stay tuned. How do you think about the competitive landscape in that application and broadly the AI opportunities? I think that as those Sovereign providers look to serve enterprise clients with AI workloads, all of the value that we have, both in terms of the differentiated capabilities around security, multi-tenancy, data protection, hybrid landscapes, plays to our advantage. We are seeing some of that momentum with the introduction of the AFX. People have been super interested in our ability to serve those new use cases that those AI cloud providers are trying to go after. Thanks. Thank you, Simon.

Your next question comes from the line of Jason Adder with William Blair. Please go ahead. Yeah. Thank you. Good afternoon. I wanted to first ask just on the performance in the quarter. You had a nice roughly $20 million beat, and then you reiterated for the year. I just wanted to understand why you did not flow through the beat. Is it just a little bit extra conservatism based on public sector or supply chain or any other factors? Yeah. I mean, I think we saw in the quarter some really strong, we talked about large deals. We had a good set of those that helped offset the really steep decline in the U.S. public sector business. We had not anticipated when we guided the quarter that U.S. public sector would suffer such a long shutdown.

I think as we look out to the second half of the year, if you look at the overall guide for the second half of the year ex spot, it actually is an acceleration year-on-year from the first half. Clearly, we want to have more line of sight into how US public sector plays out. As Wissam mentioned, it takes a while for the governmental agencies to get back to procurement and spending, and so we're being a tad cautious about that. There are no implications from the supply chain in our guidance. Okay. Great. Just a quick follow-up on Keystone. I don't think that's come up in the Q&A yet. You've been in this space for a long time, George. I mean, do you feel like we're hitting some type of an inflection in the storage as a service model? Maybe it's not an inflection.

Maybe it's just sort of every year it's going to get bigger and sort of be more of a slow burn. It does seem like that part of the business is incredibly strong, and we've seen strength across other suppliers as well. Do you think that this is just going to be more of the norm in the storage market over the next decade or so? I think this is a new way for customers to buy infrastructure. I think it is driven both by maturity of the offerings as well as customers' kind of maturity in using these business models coming from the cloud, right? They've learned how to buy infrastructure on a consumption model from the cloud. We see that growing as a part of the overall business.

Like all things, enterprise infrastructure is a very large market, and so it will grow as a part of that market. We're excited to lean into that trend. I don't think the whole market switches overnight. Nothing ever happens that way, but it'll grow faster than the rest of the enterprise infrastructure buying models. Thank you. Thank you, Jason. Your next question comes from the line of Anuta Berua with Loop Capital. Please go ahead. Yes. Thanks, guys, for taking the question. George, with video, such a big theme at Insight, let me just since we have you, take the opportunity to ask you anything notable or interesting on video that you saw over the last 90 days that you're seeing emerge? I guess video specifically, but unstructured generally in your customer base or in some of the proof of concepts. Thanks.

I have a quick follow-up too, if I could, thanks. Yeah. Unstructured is clearly the place that customers are spending time organizing. We talked about one of the use cases where a large semiconductor equipment manufacturer brought together millions of documents across a variety of applications to better understand what their engineers and their operations teams were working on so that they could kind of continue to optimize yield and efficiency in their business. We see that more and more across organizations. I think with video itself, we talked about the content production house that is using AI to build better, more enriched experiences for their digital clients. You are seeing that in certain use cases. We are also seeing the use of multimodal applications starting to grow, and then video analytics, for example, in weather forecasting, weather analysis, things like that. It is growing.

I wouldn't say it's been a dramatic inflection, but multimodal is growing. We are seeing demand from customers for support for more complex multimodal language models, for example. That's helpful. Thanks. Just a quick follow-up on memory. The hard drive makers seem, as we go through next year, like they're pretty intent on mixing up pretty quickly from 20 TB drives to 30 TB drives. Is that consistent, relatively speaking, with what your customer base is looking for also? I guess I'm wondering if that's organically in tandem with what your customers also would want. Thanks. That's it for me. I think broadly speaking, customers are looking for aerial density improvements from all of the silicon. We work with our supplier ecosystem to qualify different media types and density points, and we'll do the same. We follow customer demand. We look at what the supplier ecosystem is telling us.

Broadly speaking, performance, cost, and density are the key drivers of new silicon adoption. Thanks, George. Thanks for that. Your next question comes from the line of Luis Miciosilla with D.A. Davidson Capital Markets America. Please go ahead. Hey, George. On the latest earnings call, you seem to be a lot more encouraged about AI, and obviously, you have a lot more wins this quarter. I guess you sort of were asked this question before, but let me just try again. Obviously, there has been a massive growth in many areas, mostly in servers, but storage in general has not really seen a big, big pickup from AI. When do you think inference applications really start to create more data that would then drive material, new revenue for you? Do you see it as maybe first half calendar 2026, second half?

I mean, I assume at some point it has to hit. Yeah. I think we've always said the compute and network buildout would be both bigger and earlier than the storage buildout. I think it is because the training of large language models and the algorithms that are being used for AI does not require a lot of storage, right? Nobody is creating a second copy of all the internet's data to train OpenAI. That buildout, of course, will be bigger and earlier than storage. I think, as we said, storage is 80% of the storage use is actually from inferencing. As inferencing becomes a bigger part of the overall AI landscape, you will see more storage consumption.

I think we are seeing it in our business also in the data lake, data modernization business, where people are building large-scale multi-terabyte repositories to bring all their data together. I think if you were to look at what we said, we said, "Hey, the number of proof of concepts would tick up in fiscal year 2026." We are seeing that. I think with regard to broader enterprise AI adoption, we've always believed it's use case by use case. It's not a horizontal technology that gets deployed. It really depends on the ROI from the use cases that we see. There, we are seeing, like we said, healthcare and life sciences ahead of the market. Some of the other ones kind of heavy in proof of concept, but not yet in production. Okay. Thank you. Thanks, Peter.

Your next question comes from the line of Ramzi Mohen with Bank of America. Please go ahead. Yes. Thank you so much. George, you said at the end of your prepared remarks that you would be capturing share in the market. I was wondering how much of that is predicated on some of the flexibility you have in pricing given the supply that you have secured versus product portfolio mix capability. How are you thinking about that? I have a follow-up. I mean, broadly speaking, Ramzi, in a transaction, the majority of the value is in software and the data platform that the customer is signing onto, right? They train their teams on the use of the platform. I think our growing differentiation, not only with cloud, but cyber and bringing more advanced data management capabilities, is the leading indicator.

I think when you are trying to displace a competitor, having a kind of flexible set of tools, both things like Keystone or subscription-type services and the ability to price aggressively is helpful. Okay. Okay. Thanks, George. Just to follow up, you just raised your public cloud gross margin range, and you're already kind of at more than halfway through that range with slight increase in revenue in the quarter. I'm just kind of wondering, what's a natural ceiling for this business that you see? Maybe you can just maybe Wissam can just comment on cash flow as well, which is a little bit weak in the quarter. Just thinking through sort of the trajectory over there, that'd be helpful. Thank you. Sure, Ramzi. Let me start with the first part of your question.

On the cloud gross margin, look, this is the second quarter we're operating within our upgraded target range. We feel comfortable operating within this range. We've seen the mix in that business shifting a little bit more towards software. As we talked about before, we've had some runoff and depreciation. Having said that, there could be potentially an upward bias, but it's too early to talk about that today. I would say we're still comfortable operating within that sort of 80-85% range that we published last quarter. With respect to the cash flow, there's a bit of seasonality in our cash flow. Q2 is typically lower. In some cases, it's sometimes the lowest point in the year.

Also in Q2, we had the last installment of the tax payment, which had to do with the transition tax that was related to the law that was enacted back in 2017, I believe. Those are, I would say, the couple of factors influencing Q2 cash flow. Okay. Great. Thank you so much. Thanks, Ramzi. Your next question comes from the line of Perun Singh with Oppenheimer. Please go ahead. Yeah. Hi, Ann. Thank you for taking my questions. I really want to dive a little bit more into your AI wins. You talked a little bit about AI differentiation, the AFX, and what is helping you win versus competitors. I wanted to understand, what are you hearing from customers? Technically, do you feel you are there yet? If you could share what mix of your AI wins are with existing customers as opposed to taking share from new customers?

I think with our AI wins, just like with our all-flash business, we are bringing on a lot of new customers. That, I'm talking about the on-prem business. The cloud business has been a very strong contributor to new clients. As we have shared, roughly half of all our cloud customers are net new to NetApp, right? That pattern has stayed the same for a very, very long period of time. If you talk about the enterprise storage business, the number of new customers that are signing on to us has been strong. A lot of that is because of unified data management, hybrid cloud kind of deployment options, and an increasing number of customers coming because of our built-in NIS2 compliant cyber resilience solutions. Those are the key sources of wins. I think with regard for enterprise AI, we feel really, really strong. Yeah.

We don't have we are ahead of all the competition. Most of the competitors have kind of point products. They don't have an integrated stack. They don't have hybrid. They don't have all of those pieces that we have. We feel good. Clearly, in our install base, we have an enormous advantage because we hold all the data. As we talked about at our Insight conference, we can bring AI to your data rather than copy all your data over to do AI. Got it. Before my follow-up, anything just sorry. Just a quick follow-up. Assuming that you keep winning and your revenue trajectory is going to improve on the product side with incremental workloads coming into AI inferencing, add to that, you have Keystone going faster, easier comps with Spot, increasing public cloud mix.

Wouldn't it be reasonable to assume that off of 5% normalized growth this year, you should see an acceleration into next year? Would that be the right way to think about it? I just say we will guide fiscal year 2027 when we guide fiscal year 2027. All of the comments you made are accurate in terms of a growing part of our business is coming from faster growing components of our business like cloud, Keystone, all-flash, and AI. I think that as that mix gets to be a bigger part of our business, you should expect acceleration. We will guide fiscal year 2027 when we get there. Thank you so much. Appreciate it. Yep. Thank you. Your next question comes from the line of Ari Terjanian with Cleveland Research Company. Please go ahead. Hi, all. Thanks for the question.

First, just on the quarter, good to hear the uptick in large deal activity. I know, George, at the start of the year, you mentioned some large deals in the pipeline. What do you think drove some of the larger deal activity this quarter? Was it execution, or maybe there is some pull forward ahead of potential shortages? And then my second question, it seems like OpEx was down year-to-year. It seems like it has been trending that way. How sustainable is that? Is that more timing related of hiring? How should we think about that in the second half and next year? Thank you. Yeah. I think, first of all, with regard to wins, listen, this is execution, right? We said we had good pipeline, and we are executing to that. I do not think that there are pull-ins due to supply chain, right?

I mean, listen, we have never said there's a supply chain issue in our kind of our pipeline. I just will just say, "Hey, these are wins." You always get some that get closed earlier than others and some that get closed later. I think that our strength in the business helped offset a significant kind of deterioration in US public sector due to the shutdown. We feel good about that. With regard to OpEx, I'll let Wissam comment on it. Yeah. With regard to OpEx, look, this is a matter of timing. We expect it to be potentially slightly higher from here for the second half. We do want to continue to invest in the business. We want to invest in the growth of our business. I'll leave it at that. Thanks, Ari.

Your next question comes from the line of Amit Darjanani with Evercore. Please go ahead. Hi. This is Hannah for Amit. I was just wondering, for all-flash, the growth accelerated rather well this quarter. Can you just touch on what drove that? Was it pricing, portfolio refresh, or was it AI? I think that from a workload perspective, AI was noticeable. With regard to the mix, listen, I think we have refreshed the full set of systems last year, Hannah. We are seeing the benefit of that as customers have qualified the systems and started to use it. We saw strong results, especially in the high-performance flash area where our offerings did really well in the quarter. Thanks, Hannah. Your final question comes from the line of a CEO merchant with Citigroup. Please go ahead. Great. Thank you for taking my question.

I was wondering, you answered partly that about the high-performance mix shift. I think there was some mix shift that worked the other way in the prior quarter. Just as we look ahead, the guide for the overall gross margin sticking higher here, if you can just kind of lay out how you're thinking about the mix here. What's embedded in terms of public cloud revenues? Because obviously, that's the higher margin. What's embedded in terms of public cloud revenue growth expectations for the third quarter? Thank you. Hey, Asia. Thanks for the question. Look, we do not guide to that level of granularity. I think I've said enough with respect to product gross margin to sort of give an indication as to how we expect the next couple of quarters to shape up.

With respect to public cloud, I would expect, as I said, from a margin perspective, we're comfortable in where we are operating, exiting Q2. From a revenue perspective, we talked about the growth being very much in line with what we've seen over the past, let's say, couple of quarters. Hopefully, this helps. All right. Thanks, Asia. I'm going to turn it over to George for some final remarks. Thank you, Chris. Strong execution and operational discipline enabled us to deliver an outstanding second quarter. Our focus on AI, all-flash, and public cloud continues to yield top-line growth in an uncertain macro environment. In the quarter, we introduced substantial innovation to extend our differentiation and further solidify our leadership position.

Looking ahead, we remain confident that our visionary approach to a data-driven future will enable us to outgrow the market and capture additional share, driving value for customers, partners, and shareholders. Thank you. Have a great Thanksgiving. Ladies and gentlemen, this concludes today's call. Thank you all for joining. Goodbye. You may now disconnect.