微软公司 (MSFT.US) 2026财年第二季度业绩电话会
文章语言：
简
繁
EN
Share
Minutes
原文
会议摘要
Microsoft reported Q2 FY26 revenue of $81.3 billion, up 17% YoY, with a 26% growth in Microsoft Cloud revenue. Key highlights include a 39% increase in Azure revenue, driven by efficiency gains and increased demand. The company invested $37.5 billion in capital expenditures, with a focus on AI infrastructure, including GPU and CPU capacity. Operating margins reached 47%, exceeding expectations, despite investments in AI and compute capacity. Q3 guidance projects 15-17% revenue growth, with continued strong performance in commercial businesses and a decrease in capital expenditures. Microsoft's AI business is expanding rapidly, with new models and services, and the company remains confident in its long-term growth prospects, emphasizing high-value solutions and tools for customers.
会议速览
Microsoft's Q2 FY2026 Earnings Call: Financial Highlights and Q&AMicrosoft's Q2 FY2026 earnings call outlines financial performance, provides non-GAAP measures, and previews Q&A session, emphasizing transparency and investor communication.
Revolutionizing AI Infrastructure: Scaling Efficiency and Sovereignty in Cloud and Agent PlatformsFocuses on enhancing AI infrastructure efficiency, expanding global footprint for data sovereignty, and developing comprehensive agent platforms with broad model choice, highlighting advancements in silicon, cloud services, and AI model customization.
Empowering Enterprises with Unified IQ Layer for Agent Development and Data ProtectionMicrosoft highlights the importance of protecting tacit knowledge within model weights for enterprise value, emphasizing the role of their unified IQ layer across Fabric, Foundry, and data in powering Microsoft 365. With Fabric's rapid growth and revenue reaching over $2 billion, the company showcases how customers leverage Foundry for agentic systems, from natural language flight search at Alaska Airlines to precision farming for Land O Lakes members. Additionally, Microsoft's Copilot Studio and Agent Builder tools facilitate agent creation for knowledge workers, with over 80% of the Fortune 500 utilizing these capabilities.
Expanding AI Agent Control Plane and High-Value Agentic Experiences for Enhanced Productivity and InnovationIntroduces Agent 365, extending governance, identity, security, and management to agents across clouds. Highlights Copilot's growing momentum, productivity gains, and enterprise adoption. Discusses advancements in Dynamics 365 agents, coding productivity, and GitHub Copilot growth, showcasing real-world examples of large-scale deployments and customer benefits.
Revolutionizing Developer Workflows and Security with AI Agents and Copilot SDKGitHub Agent HQ introduces advanced AI capabilities, enhancing developer workflows with context-aware agents and expanding security solutions. Notable achievements include significant reductions in manual triage time, extensive healthcare automation, and robust cloud migration progress. The company highlights its commitment to innovation, evidenced by strong adoption rates across core franchises and partnerships, setting the stage for future growth and customer satisfaction.
Strong Q3 Financials: Revenue Growth, AI Investments, and Shareholder ReturnsRevenue reached $81.3 billion, up 17% year-over-year. Gross margin and operating income also increased, driven by cloud services and AI investments. Shareholder returns totaled $12.7 billion, reflecting strong financial performance and strategic growth initiatives.
Microsoft's Financials Highlight Growth in Cloud Services and AI InvestmentsMicrosoft reported significant increases in commercial bookings and cloud revenue, driven by large commitments from OpenAI and Anthropic. The company also noted strong growth in productivity tools, intelligent cloud services, and AI investments, despite challenges in gaming and PC sales. Efficiency gains and strategic capacity reallocation supported revenue growth, while operating margins improved due to higher gross margins and operating leverage.
Q3 Financial Outlook: Revenue, Cost, and Investment ProjectionsOutlook forecasts revenue growth of 15-17%, with FX impacting segments differently; Cogs and OpEx growth expected at 22-23% and 10-11% respectively; investments in R&D, AI, and talent highlighted; commercial bookings to grow healthily excluding OpenAI contract impacts.
Investment Priorities, Growth Expectations, and Investor Concerns at MicrosoftThe dialogue covers Microsoft's guidance for revenue growth across various segments, including Azure, M365, and Dynamics 365, with a focus on achieving top-line growth through strategic investments. It addresses investor concerns over capital expenditure growth and Azure expansion, explaining the company's approach to balancing supply and demand, and prioritizing investments for high-value solutions. The discussion concludes with expectations for operating margin improvements and the impact of external factors like currency fluctuations and memory pricing on financial performance.
Strategic Allocation of CapEx for Azure and AI InnovationsThe dialogue discusses the strategic allocation of capital expenditure (CapEx) across various segments, emphasizing the importance of investing in Azure, M365 Copilot, GitHub Copilot, and other AI innovations. It highlights the need to balance short-term revenue growth with long-term investments in RD and product innovation, aiming to optimize for the highest lifetime value (LTV) portfolio and maximize overall business growth.
Investor Concerns Over AI-Centric CapEx and Revenue AlignmentInvestors seek assurance on how AI-focused capital expenditure translates into sustainable revenue and profit growth over hardware lifespan, questioning alignment with shorter contract durations.
Mitigating Capital Risk Through Long-Term GPU and CPU ContractsThe dialogue emphasizes how most GPUs and CPUs are pre-sold for their entire useful life, significantly reducing investment risk. It highlights the distinction between shorter RPOs for certain services and the comprehensive coverage of GPU contracts, ensuring stability and minimizing potential financial exposure.
Optimizing Fleet Efficiency Through Software and Moore's Law in Aging ModelsSpeakers discuss enhancing fleet efficiency by leveraging software optimization and the benefits of Moore's Law, emphasizing improved margins over time as models age, highlighting a strategy that avoids bulk gear purchases and focuses on annual upgrades and optimization.
Concerns Over OpenAI Backlog Durability and ExposureA discussion arises regarding the significant portion of backlog linked to OpenAI, prompting concerns about its durability and exposure. Perspectives from both parties highlight the need for clarity on the situation.
Confidence in Azure's Diversified RPO Ballot and Partnership with OpenAIThe speaker highlights Azure's significant RPO ballot, larger and more diversified than peers, with a 28% growth. Emphasizes the successful partnership with OpenAI, contributing to leadership in app innovation and scale provision.
Global Capacity Expansion Accelerates Amid Rising DemandDiscusses global capacity expansion efforts to meet increasing customer demand, emphasizing the importance of efficient infrastructure deployment and operation across multiple locations, rather than focusing on specific sites like Atlanta or Wisconsin.
Microsoft's Silicon Innovation and Its Impact on Inference Costs and Gross MarginsThe dialogue discusses Microsoft's advancements in silicon technology, particularly with the Maya 200 accelerator, and its potential to enhance inference performance. It highlights the company's commitment to innovation, collaboration with partners like Nvidia and AMD, and the strategic importance of maintaining a flexible and advantageous technology fleet to optimize TCO and gross margins.
Microsoft's Frontier Transformation: Leveraging AI for Enterprise GrowthMicrosoft's AI stack, including M365, GitHub, and security tools, is driving transformative growth in enterprises. The synergy between these platforms, particularly through AI agents, is enhancing business processes, customer service, and data utilization, leading to significant advancements in operational efficiency and innovation.
Azure CPU Transition & Cloud's Role in AI AdvancementA discussion on Azure's CPU operational changes and the growing recognition among clients of the necessity to be in the cloud for effective AI implementation, highlighting the cloud's pivotal role in AI progress.
AI Workloads, Cloud Migrations, and Infrastructure Balance in Modern ComputingDiscussed the integral components of AI workloads, emphasizing the necessity of compute and storage alongside AI accelerators. Highlighted ongoing cloud migrations, noting the growth of SQL Server on Azure, and stressed the importance of maintaining a balanced commercial and AI cloud infrastructure to support client workloads and new deployments.
要点回答
Q：How is Microsoft enhancing its cloud and AI infrastructure to meet customer needs?
A：Microsoft is enhancing its cloud and AI infrastructure by building it to fit the heterogeneous and distributed nature of new workloads, with a focus on geographic and segment-specific needs. They aim to increase utilization and decrease TCO using silicon systems and software. Some examples include a 50% increase in OpenAI inferencing throughput powering Microsoft Copilots, and new capabilities and efficiencies for Fairwater data centers with an AI Wan connecting both sites, resulting in nearly 1 GW of total capacity added in the quarter.
Q：What is the significance of the new Maya 200 accelerator in Microsoft's data center operations?
A：The new Maya 200 accelerator is significant in Microsoft's data center operations as it delivers 10 plus petaflops at FP4 precision with over 30% improved TCO compared to the latest generation hardware in their fleet. It will be scaled starting with inferencing and synthetic data gen for their superintelligence team, as well as for inferencing in Copilot and Foundry. This advancement underscores Microsoft's commitment to staying at the forefront of AI technology and infrastructure.
Q：How is Microsoft expanding its global presence and meeting sovereignty concerns for its customers?
A：Microsoft is expanding its global presence and meeting sovereignty concerns by announcing DC investments in seven countries, supporting local data residency needs. They offer the most comprehensive set of sovereignty solutions across public, private, and national partner clouds, enabling customers to choose the right approach for each workload with the local control they require.
Q：What is the role of agents in the new platform shift according to Microsoft?
A：The role of agents in the new platform shift according to Microsoft is to serve as the new apps for the upcoming software shifts. To build, deploy, and manage agents, customers will need a model catalog, tuning services, orchestration services, context engineering, AI safety management, observability, and security services. Microsoft's focus on the agent platform is part of their strategy to prepare for future technological advancements and meet customer demands.
Q：How is Microsoft addressing the need for multiple models and region-specific models for its customers?
A：Microsoft is addressing the need for multiple models and region-specific models by offering the broadest selection of models of any hyperscaler. This quarter, they added support for GPT-52 as well as Claude 4 and 5. They are seeing increasing demand for region-specific models like Mistral and Coherent, especially as customers seek sovereign AI choices. Microsoft is continuing to invest in their first-party models to address the highest value customer scenarios, such as productivity, coding, and security, and allowing customers to customize and fine-tune models through Foundry.
Q：Why is connecting agents to enterprise data and knowledge important for their effectiveness?
A：Connecting agents to enterprise data and knowledge is important for their effectiveness because every firm needs to protect their enterprise value. This involves agents being grounded in the firm's systems of record and operational data, analytical data, as well as semi-structured and unstructured productivity and communications data. Microsoft is facilitating this through their unified IQ layer spanning Fabric, Foundry, and data, which powers Microsoft 365 and enhances context engineering with better source routing, advanced analytics, and security safeguards.
Q：Which industries have experienced growth with Foundry and what are some specific examples?
A：Foundry has experienced strong growth across every industry, with specific examples including Alaska Airlines creating natural language flight search, BMW speeding up design cycles, Land O Lakes enabling precision farming for its members, Symphony AI addressing bottlenecks in the CPG industry, and Foundry serving as a powerful on-ramp for the entire cloud.
Q：What is the significance of agents for knowledge workers and how is Microsoft addressing this need?
A：Microsoft is addressing the need for agents among knowledge workers with Copilot Studio and Agent Builder, with over 80% of the Fortune 500 using these low-code, no-code tools to build agents. Microsoft introduced Agent 365 this quarter to extend existing governance, identity, security, and management to agents across Microsoft 365 and Azure, providing controls that extend to agents built and deployed on Microsoft's cloud or any other cloud, as well as integration with partner tools.
Q：What is Microsoft's approach to creating AI experiences and what are the names of the new offerings?
A：Microsoft's approach involves building AI experiences that are intent-driven and work at task scope, leading to an age of macro delegation and micro steering across domains. They are embedding intelligence into multiple form factors like chat, new agent inbox apps, co-workers, workflows, and command-line tools. New offerings include AI experiences and Copilot family of apps, such as Copilot Checkouts with PayPal, Shopify, and Stripe integration for in-app purchases, and Microsoft 365 Copilot for organization-wide productivity.
Q：What is Work IQ and how does it enhance productivity within organizations?
A：Work IQ is a stateful agent that provides powerful reasoning capabilities over people, their roles, artifacts, communications, and history within an organization's security boundary, using data from Microsoft 365. It offers fast, accurate work results and has seen significant improvement in response quality with an average number of conversations per user doubling year over year. Microsoft 365 Copilot also enhances productivity with an increase of 10x in daily active users year over year.
Q：What notable achievements did Microsoft 365 Copilot and GitHub Copilot achieve in the past quarter?
A：Microsoft 365 Copilot and GitHub Copilot saw substantial growth with a 160% increase year over year, and now have 50 million paid seats and multiple times more enterprise chat users. They also have seen large commercial deployments with over 35,000 seats purchased by organizations like Fiserv, NASA, University of Kentucky, University of Manchester, and Westpac. In coding, GitHub Copilot subscribers increased 77% quarter over quarter, and Siemens adopted the full GitHub platform to increase developer productivity after a successful Copilot rollout to 30,000 developers.
Q：How are companies in healthcare and science and engineering benefiting from Microsoft's products?
A：Healthcare providers, such as Mount Sinai Health, are using Dragon Copilot to automate workflows with a 3x year-over-year increase in patient encounters documented. Science and engineering companies like Unilever and Synopsis are using Microsoft Discovery to orchestrate specialized agents for R&D, enabling them to reason over scientific literature, formulate hypotheses, spin up simulations, and drive new discoveries.
Q：What progress is being made in cloud migrations and Windows migrations?
A：Microsoft's new SQL Server has over 2x the adoption of the previous version. The company has also made progress in cloud migrations, with Windows reaching a milestone of over 1 billion Windows 11 users, year-over-year growth of 45%. Windows Edge and Bing have experienced double-digit membership growth, and LinkedIn has grown with 30% paid video ads growth.
Q：What were the financial results for the recent quarter?
A：Revenue for the recent quarter was $81.3 billion, up 17% and 15% in constant currency. Gross margin dollars increased 16% and 14% in constant currency, while operating income increased 21% and 19% in constant currency. Earnings per share were $4.14, an increase of 24% and 21% in constant currency when adjusted for the impact from the investment in OpenAI.
Q：How did capital expenditures and supply meet demand?
A：Capital expenditures were $37.5 billion, with two-thirds of the spend on shorter-lived items, primarily GPUs and CPUs, to meet the growing demand for Azure and first-party AI usage across services like Microsoft 365 Copilot and GitHub Copilot. The remaining spend was for long-lived assets supporting monetization for the next 15 years and beyond. The company is balancing the need to increase supply to meet growing Azure demand with the expansion of first-party AI usage and product innovation.
Q：How did Microsoft Cloud revenue grow?
A：Microsoft Cloud revenue was $51.5 billion, growing 26% and 24% in constant currency. Microsoft Cloud gross margin percentage was slightly better than expected at 67% and down year over year due to continued investments in AI, which were partially offset by ongoing efficiency gains.
Q：What were the results for the productivity and business processes segment?
A：Revenue from productivity and business processes was $34.1 billion, growing 16% and 14% in constant currency. M365 Commercial cloud revenue increased 17% and 14% in constant currency, with an increasing contribution from strong coepi dot results. ARPU growth was led by E5 and M365 Copilot, and paid M365 Commercial seats grew 6% year over year to over 450 million. M365 consumer cloud revenue increased 29% and 27% in constant currency, driven by ARPU growth and M365 consumer subscriptions grew 6%.
Q：How did Dynamics 365 and LinkedIn revenue perform?
A：Dynamics 365 revenue increased 19% and 17% in constant currency with continued growth across all workloads. Segment gross margin dollars increased 17% and 15% in constant currency, and gross margin percentage increased due to efficiency gains at M365 commercial cloud. LinkedIn revenue increased 11% and 10% in constant currency driven by marketing solutions.
Q：What were the figures for the intelligent cloud segment?
A：Intelligent cloud segment revenue was $32.9 billion, growing 29% and 28% in constant currency. Azure and other cloud services revenue grew 39% and 38% in constant currency. On-premises server business revenue increased 2% and 1% in constant currency, ahead of expectations, driven by hybrid solutions and pre-sales of SQL Server 2025. Segment gross margin dollars increased 20% and 19% in constant currency, with a decrease in gross margin percentage year over year due to investments in AI and sales mix shift to Azure.
Q：What were the results for more personal computing segment revenue?
A：More personal computing revenue was $14.3 billion, declining 3%. Windows OEM and devices revenue increased 1% and was relatively unchanged in constant currency, driven by Windows 10 and search and news advertising revenue. Gaming revenue decreased 10% and 10% in constant currency, primarily due to first-party content, while gross margin dollars increased 2% and 1% in constant currency. Gross margin percentage increased year over year due to a sales mix shift to higher-margin businesses. Operating expenses and operating income were relatively unchanged in constant currency, with higher operating expenses offset by higher gross margins.
Q：What is the anticipated revenue growth in intelligent cloud and more personal computing?
A：The anticipated revenue growth in intelligent cloud is two points, and in more personal computing, it is also two points.
Q：What is the expected operating expense for Q3?
A：The expected operating expense for Q3 is between $17.8 to $17.9 billion US dollars, with growth of 10 to 11 percent.
Q：What is the expected Microsoft Cloud gross margin percentage?
A：The expected Microsoft Cloud gross margin percentage should be roughly 65%, down year over year, driven by continued investments in AI.
Q：What is the expected revenue growth in intelligent cloud?
A：The expected revenue growth in intelligent cloud is between 27 and 29 percent.
Q：How is the expected revenue for the more personal computing segment?
A：The expected revenue for the more personal computing segment is between $12.3 to $12.8 billion, with Windows OEM and devices revenue declining in the low teens.
Q：How should one think about the capacity expansion and its effect on Azure growth?
A：The speaker suggests that there is a direct correlation between CapEx spend and seeing an Azure revenue number, implying that the CapEx spend, especially on CPU and GPU, will translate into Azure revenue growth.
Q：What are the primary focuses of the company's investment strategy?
A：The company's primary focuses of investment include solving for increased usage in sales and the accelerating pace of 365 Copilot and GitHub Copilot, investing in long-term research and product innovation, and allocating GPUs and capacity to AI talent. This strategy has led to improvements in products and serves to grow Azure capacity to meet demand.
Q：What is the significance of investing in various layers of the stack?
A：Investing in various layers of the stack is significant as it benefits customers and shows in revenue growth across the business. It also reflects in X growth as the company invests in its people.
Q：How should investors view the company's capital growth and portfolio GM profile?
A：Investors should view the company's capital growth and portfolio GM profile with a focus on not just maximizing one business but also on allocating capacity to build the best LTV (lifetime value) portfolio. This includes investing in products like Azure, M365 Copilot, GitHub Copilot, and Dragon Copilot, which all have a GM profile and lifetime value.
Q：What are the average durations of the company's revenue streams?
A：The average duration of the company's revenue streams is 2.5 years, which has increased from two years last quarter. The revenue streams are a combination of different contract durations, with the majority being Azure contracts that have longer durations.
Q：How does the company manage the risk associated with hardware CapEx investments?
A：The company manages the risk associated with hardware CapEx investments by ensuring that most GPUs are already contracted for most of their useful life, which is typically beyond the six-year capitalization period. Additionally, the company uses software to optimize and run the latest models on an aging fleet, which provides a revenue stream that matches the useful life of the hardware.
Q：What is the company's view on the impact of OpenAI and the contract with them?
A：The company views the partnership with OpenAI as a great collaboration, with Microsoft serving as the provider of scale. The contract is seen as a key part of being at the forefront of AI innovation. Although not much can be disclosed, the company expresses high confidence in the resiliency of the overall revenue picture, which includes a diverse range of solutions across industries and geographies, contributing to a significant and growing RPO (Revenue Pursuit Opportunity).
Q：Can the company provide any qualitative insights into the magnitude of future capacity additions?
A：The company does not provide specific quantitative insights into the magnitude of future capacity additions in the transcript provided. However, the December quarter saw an extraordinary addition of 1 GW, which suggests that capacity additions may be accelerating. Investors are advised to look at the magnitude of capacity additions in upcoming quarters, regardless of how they are allocated.
Q：What is the company's strategy for global capacity expansion?
A：The company is focusing on adding capacity globally, including in the United States and other locations mentioned, to meet customer demand and increased usage. The emphasis is on multiyear plans for capacity expansion rather than focusing on specific sites.
Q：How does the company plan to optimize its data centers and AI systems?
A：The company plans to optimize data centers by ensuring power, land, and facilities are available and by rapidly adding GPUs and CPUs as they are completed. The goal is to achieve the highest possible utility and efficiency in the operations of these data centers.
Q：What are the company's thoughts on the significance of its silicon innovation in AI?
A：The company is very excited about its silicon innovation, particularly with the Maya 200 accelerator for inference, which has demonstrated remarkable performance. This innovation reflects the company's long-term approach to developing its own silicon and working closely with its in-house team on model and system optimizations. The company aims to maintain a diverse and innovative fleet of technologies to stay advantaged in terms of total cost of ownership (TCO), rather than being locked into one solution.
Q：How is the company's AI stack contributing to enterprise transformation?
A：The company's AI stack, which includes Microsoft 365, security solutions, and GitHub, is having a compounding effect on customer benefits. The most important database for many companies is the one underlying Microsoft 365, due to its comprehensive capture of company data. New tools, such as Work IQ and the integration of AI systems across Microsoft's services, are transforming companies by aiding in coordination and leveraging enterprise data for customer service, marketing, finance, and other business functions. Additionally, agents and services like those in Fabric and Foundry are helping businesses to build their own agents for further transformation.
Q：What operational changes were made in relation to CPUs on Azure, and what do they imply for cloud migration and AI delivery?
A：The company has operational changes related to the CPU side of Azure, which imply that AI workloads require not just AI accelerator compute but also computers with storage to support the agents running on GPUs. This indicates that cloud migration is essential for proper AI delivery, as customers need a full range of infrastructure elements, including compute and storage, in the cloud region where they deploy their workloads.
Microsoft Corp.
Follow
