Q4 2025 Earnings Call Transcript

Good afternoon, and welcome to the Fiscal 2025 Fourth Quarter Hewlett Packard Enterprise Earnings Conference Call. All participants will be in a listen-only mode. Please note this event is being recorded. I would now like to turn the conference over to Paul Glaser, Head of Investor Relations. Please go ahead.

Good afternoon. I am Paul Glaser, head of investor relations for Hewlett Packard Enterprise. I would like to welcome you to our fiscal 2025 fourth quarter earnings conference call with Antonio Neri, HPE's President and Chief Executive Officer, and Marie Myers, HPE's Chief Financial Officer. Before handing the call to Antonio, let me remind you that this call is being webcast.

A replay of the webcast will be available shortly after the call concludes. We have posted the press release and the slide presentation accompanying the release on our HPE Investor Relations webpage. Elements of the financial information referenced on call are forward-looking and are based on our best view of the world and our businesses as we see them today. HPE assumes no obligation and does not intend to update any such forward-looking statements.

We also note that the financial information discussed on this call reflects estimates based on information available at this time and could differ materially from the amount ultimately reported in HPE's annual report on Form 10-Ks for the fiscal quarter ended October 31, 2025. For more detailed information, please see the disclaimers on the earnings materials relating to forward-looking statements that involve risks, uncertainties, and assumptions. Please refer to HPE's filings with the SEC for a discussion of these risks. For financial information we have expressed on a non-GAAP basis, we have provided reconciliations to the comparable GAAP information on our website.

Please refer to the tables and slide presentation accompanying today's earnings release on our website for details. Throughout this conference call, all revenue growth rates unless noted otherwise, are presented on a year-over-year basis and adjusted to exclude the impact of currency. Antonio and Marie will be referencing our earnings presentation in their prepared comments. With that, let me turn it over to Antonio.

Thank you, Paul. Good afternoon, everyone. Hewlett Packard Enterprise finished a transformative year with a record quarter of profitable growth and disciplined execution. Q4 revenue of $9.7 billion increased 14% year over year with non-GAAP operating profits growing faster, up 26% year over year.

Non-GAAP operating margin was a record high at 12.2%, including server around 10% and networking at 23%, matching the high end of our expectations. Non-GAAP diluted net earnings per share of $0.62 exceeded the high end of our guidance. Stronger profitability also resulted in higher than expected free cash flow of $1.9 billion for the quarter, capping a solid fiscal year 2025 performance. The underlying demand environment was strong throughout the quarter with orders growing faster than revenues.

We saw an acceleration in orders in the last weeks of the quarter, signaling solid demand for our portfolio. The strong finish coupled with the steps we have taken throughout the year to transform our company positions us well in 2026. Reflecting our continued confidence and ongoing technology leadership, we are raising our fiscal year 2026 non-GAAP diluted net earnings per share guidance and the midpoint of free cash flow guidance. Marie will provide more details about our Q4 financial results and our new fiscal year 2026 outlook in a moment. We intend to capture the opportunity in 2026 and beyond by pursuing the key priorities we outlined at the Securities Analyst Meeting in October.

These are building a new networking industry leader, profitably capturing the AI infrastructure build-out opportunity, accelerating our high-margin software and services growth through our GreenLake Cloud, capitalizing on the unstructured data market growth with our leading Alletra MP storage offerings, and driving the transition to our next-generation server platforms. Our strategy underpins our long-term financial framework. By fiscal year 2028, we are committed to generating at least $3 in non-GAAP diluted net earnings per share and more than $3.5 billion in free cash flow, with improved cash conversion cycles.

I also want to highlight that we recently announced agreements regarding our H3C stake in China. We will sell our remaining 19% stake for approximately $1.4 billion, which we expect to close during 2026. These transactions support our plan to reduce our net leverage to around two times by the end of fiscal year 2027. Looking back at fiscal year 2025, I am proud of what a transformative year it was for our company. We celebrated our ten-year anniversary with a focus on innovation and technology leadership. We completed the acquisition of Juniper Networks, strengthening our position in the networking market to create a new industry leader.

We scaled our GreenLake cloud, hybrid cloud software, Alletra MP storage, and AI businesses to new heights. And we continue to improve our cost structure through our catalyst initiatives to operate more efficiently. Overall, HPE delivered full-year revenue of $34.3 billion, a 14% increase year over year. Our revenue growth reflects solid performance across our three largest business segments and the addition of Juniper Networks.

We exceeded our full-year outlook for non-GAAP diluted net earnings per share and free cash flow, delivering $1.94 and $980 million, respectively. Networking played a pivotal role in our success. Segment revenue increased 51% in fiscal year 2025, achieving $6.9 billion with an additional four months of Juniper results. Orders for the new combined networking segment grew at a faster rate than revenues in fiscal year 2025 as the market recovered.

We saw strong double-digit order and revenue growth across our key segments of the networking market on an as-reported basis. Regarding the integration of Juniper, I am pleased with the significant progress we have made in forming a new unified leader in networking. In the five months since closing the transaction, we have brought together our teams, technologies, and go-to-market strategies, and the response from our employees, customers, partners, and the industry at large has been overwhelmingly positive. They are already seeing the benefits of our combined portfolio, the innovation we are driving, and the cohesive customer experience we now deliver.

Across the industry, stakeholders have expressed enthusiasm for the combined company's ability to accelerate innovation, deliver greater value, and help organizations build secure, modern, and high-performance networks for the future. The new combined networking team is performing exceptionally well. We increased orders and revenues on a pro forma basis during the first full quarter as one team. As a combined company, we are able to better compete with our comprehensive industry-leading networking portfolio.

We saw good traction with our core customer segments, enterprise and service provider. As we integrate Juniper Networks and HPE Aruba networking, we are building a modern, secure, AI-native networking portfolio. One that encompasses campus and branch, data center switching, routing, and security. By combining our unique strengths in AIOps, Agilent AI, Silicon IP, and go-to-market scale, we are positioning HPE to capture greater market share and revenue synergies.

Our networks for AI solutions grew in fiscal year 2025, where we saw notable strength in both WAN and data center switching. We are on track to achieve networking for AI cumulative order target of $1.5 billion by the end of fiscal year 2026. Campus and Branch business had a strong performance in fiscal year 2025 with revenue up double digits with orders growing above revenues. Our AI for network solutions leverage both Mist AI large experience models and Aruba Agilent AI models, driving clear differentiation that is resonating with our customers. In our server segment, fiscal year 2025 revenue grew 10% year over year.

We signed $6.8 billion in new AI system orders in fiscal year 2025. Sovereign and enterprise bookings now account for more than 60% of the cumulative orders since 2023, demonstrating our strategy to prioritize profitable AI infrastructure build-out opportunities. Traditional servers' revenue grew double digits year over year, benefiting from a refreshed cycle as customers upgrade to the latest generation eleven and twelve servers. These support greater workload performance with quantum-proof security, higher density, and lower power consumption. I am pleased that we have returned server operating margin to approximately 10% in Q4. As we look to 2026, we will draw on our supply chain expertise to secure critical commodity supply and exercise our pricing management discipline.

We expect DRAM and NAND costs to continue to increase in 2026, the majority of which we expect to pass to the market while monitoring demand. Hybrid cloud revenue grew 5% in fiscal year 2025. We added approximately 7,000 new customers to GreenLake, ending the year with approximately 46,000 customers. Total company ARR was $3.2 billion, up 62% year over year with additional Juniper. We continue to differentiate ourselves in the market with unique cloud-native AIOps and services, which together represented over 80% of ARR.

In storage, we continue to make excellent progress in shifting our portfolio to our own IP. We closed the year with strong demand for HPE Alletra MP, with four consecutive quarters of double-digit growth in both orders and revenue. We have now shipped over 7,400 Alletra MP storage arrays, more than doubling year over year. We added more than 1,300 new customers in fiscal year 2025, further solidifying our leadership in this space.

Demand for our newer differentiated private cloud solutions, private cloud AI, and private cloud for virtualization with our HPE Morpheus VM Essential software ramped throughout the year. We set new milestones, including approximately 100 new PCAI customers. Total orders for our private cloud solutions, which also includes private cloud business edition and private cloud enterprise offerings, increased more than 20% year over year. HPE is proud to have been named a leader in the 2025 Gartner Magic Quadrant for Infrastructure Platform Consumption Services, positioned highest in execution and furthest in vision.

This acknowledgment spotlights our innovative cloud-native software and services experience, which help customers accelerate transformation and drive operational efficiency across their hybrid IT environments. This week, more than 5,000 attendees participated in our Discover Barcelona Customer and Partner event. We introduced new innovations and demonstrated the considerable progress we have made in bringing together HPE and Juniper in a short period of time. For example, in AI for networks, we announced new AIOps capabilities and common infrastructure products that deliver a consistent self-driving experience across both HPE Aruba network in Central and HPE Juniper network in Mist cloud platform.

This is an important milestone in achieving a unified experience across campus and branch and demonstrates our commitment to quickly cross-pollinating our platforms and driving common infrastructure products for investment protection. In Network for AI, we announced the first OEM switch leveraging Broadcom Tomahawk six silicon to address performance-hungry computing for AI inferencing. Our new Juniper networking data center switch connects GPUs within data centers with the world's highest performance ultra Ethernet transport-ready switch, delivering 102.4 terabits per second total capacity. And our new HPE Juniper multi-service edge router brings AI inferencing closer to the source of data generation, with performance up to 1.6 terabits per second with full duplex 400 gigabits per second connectivity.

Also in Barcelona, we extended the NVIDIA AI computing by portfolio, introducing new solutions for AI factory scale and performance. In HPE's AI factory networking solutions, we introduced the new HPE Juniper network in edge on-ramp long-haul data center interconnect with our HPE Juniper network in MX and PTX high-speed routing platforms. Integrating our Juniper routing solutions with NVIDIA Spectrum Ethernet networking and NVIDIA BlueField three DPUs enables high-speed, secure, and efficient edge on-ramp and AI data center interconnect use cases. We also announced the first AMD Helios AI RAC scale architecture with integrated HPE Juniper scale-up Ethernet networking.

The solution leverages purpose-built HPE Juniper networking data center infrastructure and software to accelerate performance and deployment of at-scale AI training inferencing for cloud service providers and sovereign clouds. In storage, we announced new solutions to accelerate AI data pipelines. The new HPE Alletra Storage MPX 10000 data intelligent nodes transform the X 10,000 into an active data layer that enriches data in real-time for AI pipelines. Finally, last month at Supercomputing 2025, we demonstrated our next-generation liquid-cooled Cray GX and supercomputing platform at data center scale.

We have already won contracts to build five large sovereign systems utilizing this technology, including a second-generation exascale AI supercomputer for the United States Department of Energy. These announcements highlight the strength of our innovation to deliver the best network in AI and cloud solutions for our customers and partners. As I reflect on the past year, I want to highlight a few critical milestones we have achieved as a company. First, we repositioned our business, creating a new networking leader by combining the strength of HPE Aruba networking and Juniper Networks.

Second, we scaled our AI business with a focus on sovereign and enterprise customers, representing more than 60% of our bookings. Third, we advanced our cloud business with innovative offerings such as our own Alletra MP storage for both structured and unstructured data, cloud ops suite software, HPE Morpheus Enterprise, VM Essentials, and private cloud AI, which are all underpinned by our GreenLake cloud scale and experience. And lastly, we continue to improve our cost structure through our catalyst initiatives to operate more efficiently by leveraging automation and new AI technologies. We enter fiscal 2026 with a world-class portfolio and a stronger market position.

Network, cloud, and AI remain the three pillars of our strategy. Our organic investments are focused on higher-margin opportunities, and with a disciplined approach to the Juniper integration, we are positioned to accelerate value for shareholders. On behalf of the HPE management team, I want to thank our customers, partners, and team members for their dedication this year. Your support has been instrumental in making fiscal 2025 a transformative year for HPE.

We look forward to successfully executing on our strategy to achieve our fiscal 2026 and long-term plan commitments to our shareholders. With that, I will hand it over to Marie for a detailed review of Q4 financial results and our outlook for Q1 and the full fiscal year 2026.

Thank you, Antonio, and good afternoon, everyone. Fiscal year 2025 has been a transformative year for HPE. We took significant strides towards aligning with our long-term strategy, delivering on our commitments, and positioning the company for sustainable growth. We closed the acquisition of Juniper Networks, our largest ever, which has expanded our reach into the data center and significantly bolstered our scale in the networking sector.

The integration of Juniper is a top priority, and I'm pleased to share that our progress is progressing well. While it's early days, the initial synergies we are seeing are encouraging, reaffirming our belief in the potential of this acquisition to drive higher margin and higher growth opportunities for HPE. We have established a robust foundation to transform our cost structure through catalyst initiatives, which combined with the synergies from Juniper are targeting approximately $1 billion in annualized structural savings by fiscal 2028. We are pleased with the significant catalyst-related cost reductions we captured in fiscal 2025. We exceeded our target of achieving 20% of $350 million in annual run rate cost savings as our results track ahead of plan.

As part of Catalyst, we continue to optimize and streamline our portfolio to become a more agile and efficient organization. Turning to fiscal year 2025, total revenue reached $34.3 billion, up 14% year over year. Non-GAAP diluted net earnings per share was $1.94 and free cash flow was $986 million, exceeding the outlook ranges provided at our Security Analyst Meeting in October. GAAP diluted net earnings per share was a loss of $0.04, below our outlook range primarily driven by accounting adjustments related to the acquisition of Juniper and treatment of preferred stock.

We returned $886 million to our common shareholders through dividends and share repurchases, further demonstrating our commitment to delivering value to our investors. Non-GAAP operating expenses of $7.5 billion increased 11% year over year and declined 60 basis points as a percentage of revenue. Excluding Juniper, non-GAAP operating expense was down modestly year over year driven by ongoing cost management initiatives. Let me highlight some key segment-related metrics from our fiscal 2025 results.

Networking emerged as a standout performer of the year, acquisition of Juniper was instrumental in driving this success, particularly in our WAN business. Overall, networking orders were up strong double digits year over year on a pro forma basis. Meanwhile, our AI server business also had good traction, with orders totaling $6.8 billion for the fiscal year and cumulative AI orders since Q1 fiscal 2023 reaching $13.4 billion. Additionally, we saw strong double-digit year-over-year growth in Elektra MP storage orders, signaling momentum as we focus on developed intellectual property and innovation.

Now, let me walk you through our fourth-quarter performance. Revenue for Q4 was $9.7 billion, up 14% year over year and 6% sequentially, coming in slightly below the low end of our outlook range due primarily to the push out of some AI shipments. This growth was primarily driven by our acquisition of Juniper Networks and robust performance in the HPE Aruba networking, partially offset by declines in server and hybrid cloud revenue. Our Q4 profitability was another highlight of the quarter.

Non-GAAP gross margin reached a record 36%, driven by a favorable mix shift to networking, stable gross margins across our three largest business segments, and disciplined pricing strategies. Non-GAAP operating expenses rose 40% year over year, primarily driven by the acquisition of Juniper. Excluding Juniper, non-GAAP operating expenses increased by 3%, reflecting our ongoing efforts to streamline our cost structure and maintain disciplined management of discretionary spending. Non-GAAP operating margin expanded to 12%, an improvement of 110 basis points year over year and 370 basis points sequentially.

These improvements were supported by Catalyst cost savings and Juniper synergies. For the quarter, our non-GAAP diluted net EPS was $0.62, exceeding the high end of our guidance, while GAAP diluted net EPS was $0.11. The difference reflects the exclusion of certain items, including amortization of intangible assets, Juniper-related acquisition costs, stock-based compensation expense, and cost reduction plan expense, partially offset by adjustments for taxes and other adjustments. Our annualized revenue run rate or ARR grew by 62% year over year, reaching $3.2 billion. This growth reflects the strength of our GreenLake platform, the accelerating adoption of our software solutions, and the incremental contributions from Juniper.

GreenLake continues to grow its footprint, adding around 2,000 new customers in the quarter, bringing our total to approximately 46,000 customers by year-end. I'm particularly pleased with our Q4 free cash flow of $1.9 billion, well above our expectations, bolstered by strong Juniper collections and better-than-expected profitability. Now let me provide some color by segment. Starting with networking, which is the cornerstone of our transformation strategy.

HPE is uniquely positioned to lead in the networking market, offering an industry-leading secure AI-native networking portfolio that spans campus and branch, data center switching, routing, and security solutions. In Q4, networking generated revenue of $2.8 billion, representing a 150% year-over-year increase and a 62% sequential growth. Q4 revenue benefited from the first full quarter contribution of Juniper results, alongside continued growth in our HPE Aruba networking business. We saw double-digit growth pro forma year over year across WAN, Campus Branch, and Security.

We are particularly encouraged by the profitability of this newly consolidated networking business, which delivered an operating margin of 23%. While this represents a 140 basis point decline year over year, it marks a 220 basis point improvement quarter over quarter driven by robust gross margin performance and higher revenue. Although we will not report Juniper's results separately going forward, we are pleased to note that integration synergies have already been materializing, enabling Juniper to deliver an eight-year high in operating profit margin during Q4. We remain focused on continuing to unlock the value of this integration.

We are combining our two networking sales teams into a unified organization and implementing a new sales coverage model to drive efficiency and alignment. Starting in January, we will also introduce a unified sales compensation plan, promoting consistency across the integrated networking team. These actions position us well to build on the momentum we have established and capitalize on the significant market opportunities ahead. Moving to our server business.

In Q4, server revenue totaled $4.5 billion, representing a 5% decline year over year and a 10% sequential decrease. This performance primarily reflects the timing of AI service shipments during the quarter and lower-than-expected US federal spending. Despite these headwinds, we were encouraged by robust server order growth across both traditional server and AI offerings, with demand significantly outpacing revenue in this period. Momentum in traditional server was driven by the continued shift toward next-generation platforms, which contributed to higher average selling prices.

Our gen 11 and gen 12 platforms now comprise 98% of our traditional server revenue mix. As we look ahead, we will maintain a disciplined focus on balancing profitability and unit growth for our traditional server business, emphasizing volume and services attached to sustainable long-term cash flow. Turning to AI systems, orders were strong in the fourth quarter, reaching $1.9 billion, largely fueled by demand for sovereign customers. Additionally, our AI server pipeline remains multiples of our backlog, underscoring the substantial interest we are seeing from sovereign and enterprise customers.

It is worth noting that we expect AI demand to remain uneven, as some of our larger sovereign customers are placing orders with extended lead times, which may defer shipments to future periods. We successfully delivered an operating margin of approximately 10%, consistent with our outlook. We improved our margin performance by 340 basis points sequentially, a result of our disciplined approach to managing AI volumes, executing on traditional server pricing, and reducing operating expenses. Moving to our Hybrid Cloud segment.

We reported revenue of $1.4 billion for the quarter, reflecting a 13% decline year over year and a 5% decline sequentially. While below our outlook for flattish revenue quarter over quarter, this performance reflected our strategy to sharpen our focus on higher-margin HPE-developed solutions while intentionally reducing our exposure to low-margin non-IP-related businesses. As part of this strategic pivot, we are encouraged by the continued momentum in our innovative offerings. Orders for Elektra MP grew strong double digits year over year, underscoring the growing traction of this solution in the market.

We are also seeing good growth in private cloud AI orders, which more than doubled sequentially. And we closed the year with approximately 100 new logos in this space. Hybrid Cloud operating margin for the quarter came in at 5%, representing a 280 basis point decline year over year and a 90 basis point decline sequentially. This reduction primarily reflects the scaling of operating expenses as we continue to invest in innovative and transformative solutions.

Turning to Financial Services. Our Financial Services business delivered $889 million in revenue, roughly flat sequentially and down 2% year over year. Financing volumes totaled $1.5 billion, reflecting consistent demand within the segment. Operating margin expanded meaningfully to 12%, up 230 basis points year over year and 160 basis points quarter over quarter, driven by favorable lease portfolio mix and lower bad debt levels.

Our Q4 loss ratio held steady at approximately 0.5%, while return on equity reached 21%, our highest level in over five years. These results underscore the strength and resilience of our financial services portfolio. For the quarter, we delivered strong operating cash flow of $2.5 billion and free cash flow of $1.9 billion, reinforcing our disciplined approach to financial management. Generating robust free cash flow and successfully integrating Juniper remain top priorities as we execute our fiscal 2026 strategy.

Our Q4 cash conversion cycle improved last quarter by five days to thirty days. This was driven by a decrease in days receivable largely due to strong collections for Juniper, including early payments, and a decrease in days of inventory due to lower purchases, offset by a decrease in days payable due to higher vendor payments. Inventory ended the year at $6.4 billion, reflecting a 19% decrease year over year and an 11% sequential decline. We continue to demonstrate our commitment to a balanced capital allocation strategy.

During the quarter, we returned $171 million through dividends to common shareholders and an additional $100 million via share repurchases. At the same time, we reinforced our financial strength by improving our pro forma net leverage ratio from 3.1 to 2.7 times, primarily due to an enhanced cash position resulting in a net pay down of $2 billion of term loan debt. In terms of portfolio optimization, as announced previously, we are selling the entirety of our remaining interest in H3C in transactions valued at $1.4 billion, subject to regulatory review and approval. We expect to conclude both sales in 2026 and intend to use the proceeds to further deleverage our balance sheet, aligning with our strategic objective of maintaining a strong flexible financial position.

Before I get into the details of our guidance, let me first address the industry-wide commodity cost inflationary environment and provide some context around the actions we are taking. We are monitoring the DRAM and NAND markets daily and taking mitigating actions to preserve our margins. This includes partnering with our suppliers, taking pricing actions, and working with our customers to shape demand. Overall, we expect to pass through the majority of component cost increases while monitoring demand elasticity with our customers.

These dynamics are factored into our outlook, with our server business most exposed, followed by storage, and then networking. We will continue to focus on what we can control while navigating the environment as it evolves. Turning to our FY 2026 outlook. We are reaffirming our revenue growth outlook range of 17% to 22% on a reported basis or 5% to 10% on a pro forma basis as was provided at our security analyst meeting.

We expect our revenue mix to be approximately 46% in the first half and 54% in the second half, which is a bit more back-ended than our typical seasonality given the composition of our AI server backlog and pipeline. We are raising our full-year networking revenue growth outlook to 65% to 70% on a reported basis, implying approximately $11 billion as we see strong traction in the marketplace for our combined portfolio. The approximately $11 billion in revenue now translates to a mid-single-digit growth on a pro forma basis. Our FY 2025 pro forma baseline shifted approximately $100 million related to the move out of non-core assets to corporate and other aligned with our restated financials for the new segmentation effective November 1, 2025. We are optimistic about our networking outlook with the commencement of our sales day one on January 1 when we combine our sales forces.

We will update you on our progress as we move through the integration. We expect operating profit margin in the low 20% range driven by top-line growth and our cost optimization initiatives, resulting in networking constituting greater than 50% of our total operating profit for the year. We are reaffirming our cloud and AI revenue growth outlook of mid-single to low double-digit rate growth and operating margin of 7% to 9%. Given the increasing mix of sovereign customers and our AI backlog, we expect the majority of the backlog to be realized in the second half and beyond.

We remain focused on prioritizing profitable server growth, implementing pricing actions to counter rising commodity costs, while balancing the shift to higher-margin owned IP. We are raising our fiscal 2026 non-GAAP diluted net EPS outlook range to $2.25 to $2.45. We expect to recognize approximately 53% of EPS in the second half. Our revised outlook for seasonality versus what we provided at SAM reflects the rapidly shifting component environment. We now also expect a higher GAAP diluted net EPS range of $0.62 to $0.82. These estimates reflect a fully diluted share count of 1.44 billion, non-GAAP tax rate of 14%, and OI and E of approximately $650 million.

In addition, given faster-than-expected benefits from the integration of Juniper, we are raising the midpoint of our FY 2026 free cash outlook and now expect a range of $1.7 billion to $2 billion, which includes approximately $700 million in costs related to the Juniper and Catalyst program. Our slightly increased cash expense outlook reflects accelerated implementation of catalyst-related initiatives at FY 2026. For Q1, we expect total revenue will be between $9 billion and $9.4 billion, with sequential revenue decline roughly in line with historic seasonality. For networking, we expect revenue to grow 145% to 155% year over year on an as-reported basis, or the high end of our updated pro forma revenue growth target range of mid-single-digit.

This growth is driven by strength in our backlog and Juniper seasonality. We expect continued strength in the business and synergy realization to drive an operating margin rate in line with our full-year guidance. In Cloud and AI, we continue to see the impact of lumpiness in AI server revenue and expect a sequential decline in the AI server revenue with the majority of AI deals shipping in the latter half of the year. Given the expected mix shift towards traditional server and benefits from recent pricing actions, we expect operating margins for cloud and AI to be slightly above the high end of our FY 2026 target range.

On a consolidated basis, we expect Q1 total operating expense to decrease sequentially. Combined with our commodity cost pricing mitigations, we expect our non-GAAP total operating margin rate to be up slightly sequentially. Consequently, we expect non-GAAP diluted net EPS between $0.57 and $0.61 and GAAP diluted net EPS between $0.09 and $0.13. In closing, FY 2025 was a year of transition for HPE, as we repositioned the company for its next phase of growth, completing the integration of Juniper and taking decisive action on our cost structure. There is more work ahead, but we believe we have the right strategy, the right portfolio, and a clear path to making HPE a leaner, more efficient company aligned with the networking, cloud, and AI needs of our customers.

I'm confident in the opportunity in front of us, and we remain firmly committed to consistent execution and the financial framework that we outlined for profitable growth and strong cash generation. With that, I'll turn the call back to the operator to begin the Q&A. Thank you.

Thank you. We will now begin the question and answer session. The first question will come from Amit Daryanani with Evercore. Please go ahead.

Good afternoon. Thanks for taking my question. I guess if I look at the fiscal 2026 EPS free cash flow guide, you folks are kind of raising both those numbers while revenue guide as well as the unchanged. Assume this is just networking mix that's helping you a bit over here.

But, you know, can you spend just some time talking more about how are you thinking about the memory headwinds in the fiscal guide versus what you were expecting back at SAM, how do you think this memory cycle ends up being different versus what we saw back in '17, '18? That'd be really helpful. Thank you.

Well, good afternoon, Amit. This is Antonio. I will start, and then Marie will provide further details. Look, we are very confident in the new guide that we provided, which obviously raises the EPS and the free cash flow.

It has to do with the combination of the mix of the business, obviously, networking revenue increased now to the mid-single digits. And the continuous actions that we take across the company. Right? Catalyst is slightly ahead of what we wanted to be.

But net-net, you know, the driver of this is all the execution that we've seen now in Juniper. Now we are a networking-centric company, clearly drives all of that. And then on the component side, look, it looks a little bit like the early part of the COVID time frame with it's all about the allocation of supply as we go forward that drives cost increases. We already, by the way, Amit, have implemented price increases in the month of November.

So that's already in place. And we have very strong capabilities in our supply chain to secure the allocation of components we need. And we have discipline in passing through the cost through our pricing, which again we already did in November. Everything we know as of today is in our guide.

So what we know today is already included in the guide for both revenue and EPS.

And, Amit, good afternoon. Maybe I'll just add just to clarify on the guide itself so you are correct. We did raise the guide the midpoint by 5¢, and that's actually driven by the revenue that moved out of Q4 into Q1. So we got some profit benefit there. And, also, as you correctly said, some of the stronger networking backlog that we're seeing.

So overall, look at the prudent guide. If we can do better, we will, and, you know, we're pleased with the cash flow as well. I think that's a great example of just some of the strengths that we saw in Juniper collections in Q4 because we had very good cash flow in Q4, and that actually trickling through into '26 as well. So overall, they're the two drivers of the increases in the guides, Amit.

Very good. Thank you, Amit. Operator, next question please.

The next question will come from Samik Chatterjee with JPMorgan. Please go ahead.

Hi. Thanks for taking my question. I guess, Antonio, just following up on response to Amit's question. You did talk about acceleration in orders towards the end of the quarter, and you're referring to price increases in November.

Maybe if you can just sort of help us in terms of how you're interpreting the increase in orders that you saw towards the end of the quarter, and do you see sort of them largely being in response to price actions you're going to take in November? Just trying to understand what the drivers are, what you're seeing at your end. Thank you.

Yeah. Thank you, Samik. We saw a very linear quarter in Q4, and in the last few weeks, we saw an acceleration of that. And it's true across the entire portfolio.

Look, we closed the first quarter as a fully integrated networking business. The commodity cost has limited impact in the networking business because obviously DRAM and NAND are not really applicable to the networking business. And that business did really, really well. We saw orders growing faster than revenue there.

Then we saw continued momentum in our Alletra storage with double-digit year-over-year growth in both orders and revenues. And then the traditional server business orders did well. It's early to say that it was a pull-in of sorts on demand, but I will tell you that we felt prudent at this point in time considering the early signals we got from our suppliers on DRAM. To take the actions on pricing and we expect that the NAND part will follow in 2026. And therefore we will do the same.

The customers obviously have budgets that end on December 31. Most of the customers are on calendar year budgets. And so you should normally expect an acceleration of orders in the last part of the year. But we have conversations with customers about what we expect on commodities so they can make the right decisions and place the right orders. We really focus on conversion short term versus guaranteeing pricing long term.

Because obviously all our agreements have price protection. And therefore, we will be very upfront with customers when to place the orders and what to expect in terms of supply. Very good. Thank you, Samik.

All right, operator. Next question, please.

Next question will come from Tim Long with Barclays. Please go ahead.

Wanted to ask on ARR and the GreenLake side, if I could. Could you just talk a little bit about traction you're seeing with the as-a-service models more broadly? And then, obviously, the ARR jumped up a little bit with Juniper Mist being added. If you could just talk a little bit about what adding Juniper does to this part of the model?

Does this help maybe accelerate, obviously, the networking part, but are there any broader benefits by having the two companies combined on the GreenLake side? Thank you.

Yeah. Thanks, Tim. Look. All the ARR we are from Juniper is in the software subscription services.

Because remember, our ARR is a combination of call it, the SaaS, which is a software subscription on GreenLake. And that's a combination of networking with Aruba Central, obviously, all the hybrid cloud, software, with Morpheus, VM Essentials, Zerto, OpsRamp, and the like, and also the storage business, right, with Alletra MP. And then we have the GreenLake Flex, which is the pure consumption model, which obviously is inclusive of hardware and financing. And the financing is only for the operating leases part of the equation.

But the addition of Juniper comes with all the software subscription that they are tied to things like Mist, and Apstra and so forth. In fact, one of the key announcements we made yesterday when I was in Barcelona is the integration already of Apstra with OpsRamp as a part of GreenLake. That's an acceleration of ARR. But look, we now have a baseline that represents 80% of ARR is software and services.

And that's very, very high. As we continue to grow the networking business, which now we are at the core, our larger networking-centric company with a tremendous amount of innovation, if you look at innovation this week, has been significantly higher than any other event we have had in the last few years. Rami came strong and articulated strategy about cross-pollinating the Mist and Aruba Central. That means both will get the benefits of each other, which means more software added through the AI ops in addition to the fact that now we support dual hardware on both platforms.

We also announced the new Juniper QFX fabric, which is the first direct liquid cooling. He took advantage of our direct liquid cooling. And that has also a component of software that will go into the subscription. So that's what we're seeing.

Obviously, now we have a $3.2 billion ARR and from there we expect to continue to grow at the bigger base, obviously. And that's where we are excited about what comes next with GreenLake. And all the AI ops intelligence we build around it.

Thank you, Tim. Next question please.

Next question will come from Eric Woodring with Morgan Stanley. Please go ahead.

Hey, guys. Thank you so much for taking my question. Antonio, I'm going to go back to kind of where Amit and Samik were touching on the commodity stuff. You know, truly want to understand your thoughts on the pass-through of demand elasticity because you know, we can look at server DRAM contract pricing up 50% in April alone.

DRAM is obviously a considerable part of the server bill of materials. You know, I guess, simple math would say you'd have to raise pricing 15% just to account for the contract pricing in 4Q alone. So I guess maybe just is that the implication that you're kind of referencing here? And second, what are you assuming for demand elasticity just because if we take your kind of cloud and AI guide it does assume a notable acceleration in growth through the year.

So I'd love to just understand, are you talking about pricing increases, that can be considerable? And two, how do you expect demand to respond? Thank you much.

Thank you, Eric. I think you are spot on. I think, you're not that far, to be honest with you. You may be a little bit short, in fact.

I will say, look. We have made pricing changes to reflect exactly what you say, which is the DRAM cost and the percentage of the mix of that in the BOM in our servers. And we expect that the NAND of that will follow as we go forward. As for demand elasticity, look.

There are benefits to upgrade because we have shown customers that you can take a generation 10, which is maybe four years old, down to one, seven generation 10 servers down to one. That helps reduce your energy cost by 65%, get better performance on a per-core basis, more density. At the same time, you can pay off that investment in less than two years. The depreciation of that return is very, very fast.

As we said in my remarks, we are going to monitor the cost and the demand. I think you will see a rebalancing over time between units and revenue. Remember, more than two-thirds of our AUP is structural. I believe there are unit growth that we'll continue to see but maybe a little bit more muted than maybe expected six months ago.

But the balance, the revenue will grow as implied in our revenue guidance because of these AUP mix shifts. I think you're spot on in your thesis.

And I'll just add, Eric. You know, look. We're using multiple tools. You know, you hit on pricing, but, you know, also, look.

We've been down this track before. We know that, you know, other tools like demand shaping are critical to use through this process. So we will be looking at how we can shape demand, you know, with the parts that are available while we try to balance some of our key relationships. But as Antonio said, you know, our guidance really reflects the best estimate of the impact of commodities and the actions as of now.

And by the way, another point, Eric, is that we have frame agreements, think about large enterprises that buy on a catalog of preconfigured products. They are priced protection guarantees there. Obviously, there is an ability for us to raise prices as a part of the changes we see in the industry.

Thank you, Eric. Next question please.

Next question will come from Wamsi Mohan with Bank of America. Please go ahead.

Yes, thank you. I wanted to clarify some of the comments around seasonality. In your slides, you talk about the $9 billion to $9.4 billion in 1Q. 1Q is a decline consistent with historical seasonality. But I think where you said that there were some pushouts of servers from 4Q to 1Q.

So should that not be driving much better seasonality and sort of a higher outlook in fiscal 1Q? And similarly, you noted that the AI server lumpiness and sort of timing will drive the second half of the fiscal year much higher relative to the first half. It's like different seasonality. So I'm just trying to piece those pieces together what's baked in from pushouts into 1Q and would the seasonality have been even worse had those pushouts not occurred.

So hopefully you can just maybe put that all in some context. Thank you.

Sure. No problem, Wamsi. So why don't I just clarify the seasonality with respect to Q1, then I'll turn to Antonio to talk about the AI timing. So with respect to Q1 seasonality, the way to think about it is just bear in mind that from a Q1 perspective, spot on.

We did have some AI deals that moved out of Q4 into Q1. It is in line with our normal sort of historic seasonality for Q1 revenue. So sort of use that as an anchor to think about the year. I think in the prepared remarks, I did mention that we have a split between revenue 46% in the first half and 54% in the back half. So use that as another way to sort of think through the seasonality.

So hopefully, gives you the context on Q1. And then I'm gonna turn it to Antonio to talk specifically about the back half and how we see the AI shipments sort of placing themselves throughout the year. Antonio, over to you.

Yeah. Wamsi, look. On the AI conversion, right, because now more than 60% of our orders are in sovereign and enterprise, and obviously enterprise is a very large number of deals that get through, but they are smaller in size compared to potentially a sovereign AI cloud, which obviously is much larger. But when I think about that, look, they are longer to convert for a number of reasons.

Number one, obviously, is the whole procurement process is much longer. Get the funding locked and so forth. The data center readiness, the availability of power and cooling, and the like. And some of these deals, by the way, are not current technologies.

Maybe they are for their back half set of technologies, particularly with NVIDIA, Beta, and Ruby. So you have a combination of factors there. In regard to the push out of the deals, look, there were some deals for the government related. They take time.

They take time to really get the machine up and running again. Remember, we're just now a handful of weeks here since they came back online. That may take an extra full set of weeks. Then there was one particular deal that they weren't ready with the data center.

And so we deploy a number of pods and those other pods will take a number of incremental weeks to get it done. So we expect that the back end of the year will have the biggest part of the AI revenue conversion. But at the same time, we continue to stay focused on those two segments because the focus there is because of our ability to play and win with the right margin profiles and the right working capital.

Thank you, Wamsi. Next question please.

Next question will come from Aaron Rakers with Wells Fargo. Please go ahead.

Yes, thanks for taking the question. I wanted to ask about the networking business. It looks like on a pro forma basis using Juniper's results, grew kind of in the low to maybe mid-teens range year on year this last quarter. It looks like you're guiding kind of the pro forma number to kind of grow in that mid-single-digit range.

Is quarter and in that range for the full year. I'm curious as why necessarily you see a deceleration in that? Is that conservatism? And can you talk a little bit about Juniper's positioning in some of these AI fabric build-outs, what your discussion has been with customers thus far?

I think Juniper has had a position in some larger build-outs in the past. Thank you.

Yeah. No worries. Why don't I just talk a little bit about the revenue and how we're thinking about it? Yes.

You're right. You know, look, really pleased with the results I'd say, for Q4, both in terms of both revenue and operating margin for the business. I think you're seeing the transformative power of those two networking companies coming together and just, you know, what we can drive here. What I would say is you think about, you know, the rest of the year, we have kept revenue, you know, we did raise revenue actually in terms of the range that we gave you for networking.

And a couple of things just to bear in mind, we do have a critical milestone in the integration that's taking place in this current quarter, which is the integration of both of the sales forces. So, you know, it's early days, and at this point, you know, we've got a prudent outlook given where we're at relative to the integration itself. So a couple of things just to bear in mind in terms of drivers. You know, there is some seasonality I mentioned earlier.

We see some commodity pressures that you heard us talk about on the call, and then, obviously, we've got the product mix in terms of Bitmo cloud and AI. But I'd say overall, you know, we're comfortable with the guide that we've given you so far for networking for the year.

So, Tim, look. This is in my mind, it's very simple. We have an incredible portfolio. In the campus and branch, we continue to make tremendous traction.

Both platforms are winning in the market. There is more about integrating the sales force, getting them stabilized from an account coverage perspective. I think the channel will be the opportunity for us to drive up outside because they are all super excited about that part of the business. And look, we had major wins on both sides.

And so we expect that to continue in 2026. On the data center searches side, we expect that business to grow at or above market as we go forward. Networking for AI, we said we expect to achieve $1.5 billion by 2026. Obviously, some of that will cut into 2027 because you have the backlog conversion that has to take place. The one area that I am actually more excited about in many ways is the routing business. The MX platform is the standard for on-cloud on-ramp cloud.

It's an amazing product that really is winning in the market. And even the PTX for the DCI, meaning the data center interconnect for long haul, is the reference. That business has a very large backlog that will convert later in the year. Marie referenced this, I think it's important that all of you remind yourself that Juniper is a back-end loaded conversion to revenue.

That has been always the case. That's why we have given this seasonality in line to also what HPE is going to do. But based on what we know today, and the line of sight of the integration, and still have to go through the sales integration on January 2, we still raised the revenue guidance to now call it the mid-single digits, call it the five-plus percent which is almost double of what we give you the security analyst meeting at the midpoint. Which was between 2-5%.

So that tells you we are growing confident. And as we go forward, we will see how the team executes, but look, we believe there is a tremendous opportunity and the goal there is all the goal for us is all about execution.

Thank you, Aaron. Right, operator, last question please.

And your final question today will come from Asiya Merchant with Citigroup. Please go ahead.

Great. Thank you for squeezing me in here. Marie, if I can just ask about the sale of the assets that you have and how that is reflected in your OI and E. I believe the OI and E figure didn't really change from what you provided at the analyst event. So how should we think about the sale and how that should be factored into the OI? Thank you.

Yeah. Look, Asiya, at this point, I assume you're talking about H3C. Look. Everything's factored into the OI and E numbers that I gave in my remarks. We'd actually planned for this all along in '26, so it's all captured there in the prepared remarks that I gave you in the call today, Asiya. Thank you.

Yeah. No. Thank you. Thank you for the patience today.

Look, Marie and I provided a lot of details. We provided a lot of data. I'm sure you will have follow-up questions that the IR team will handle. But we felt that this was the right time to give you as much data as possible as we enter 2026. And, look, I will say that 2026 is a year where HPE has the opportunity to not only deliver what we committed but really drive the transformation of this company to new heights based on the fact that we are now a networking-centric company.

And on that foundation with the latest innovation, we will deliver great experiences, growth opportunities, in both cloud and AI. But we will maintain that strong discipline, focus on growth, and operating margins ultimately drive profitable growth and free cash flow. And I believe our strategy is working and the Juniper integration is working. That's what I will leave you today.

And I take this opportunity to thank you for your coverage and feedback. And if I don't speak to you, I wish you and your families happy holidays.

Conference has now concluded. Thank you for attending today's presentation. You may now disconnect.