Sign in

    Microsoft Corp (MSFT)

    Q3 2024 Summary

    Published Jan 10, 2025, 5:10 PM UTC
    Initial Price$373.86December 31, 2023
    Final Price$420.72March 31, 2024
    Price Change$46.86
    % Change+12.53%

    What went well

    • Microsoft's significant investments in AI infrastructure are positioning it as a leader in AI, capturing demand as customers build AI solutions on Azure, despite capacity constraints.
    • Azure is gaining market share, with AI adoption helping acquire new customers and drive usage of adjacent Azure services, leading to balanced growth and stabilization in core cloud consumption.
    • AI is driving new spending across departments, not just within traditional IT budgets, as organizations adopt AI to optimize processes in areas like customer service and marketing, expanding Microsoft's addressable market.

    What went wrong

    • Microsoft's capital expenditures are expected to ramp over 50% year-over-year to over $50 billion, with media speculation of even larger spending ahead, such as a $100 billion data center, which may strain financial resources if returns don't materialize.
    • Capacity constraints in AI infrastructure are limiting Microsoft's ability to meet current AI demand, potentially impacting both current and future revenue growth.
    • Shift in IT spending towards AI projects may cannibalize other Azure projects, leading to potential stagnation in overall spending on Microsoft's platforms.

    Q&A Summary

    1. Azure AI Growth and Capacity Constraints
      Q: Why is Azure AI growth leveling off; are capacity issues affecting it?
      A: Amy explained that demand exceeds supply for Azure AI capacity, especially on the inferencing side, which may have impacted the 7-point lift to Azure growth from AI this quarter and will slightly impact next quarter as well.

    2. Azure Demand Environment and Customer Budgets
      Q: How is the demand environment for Azure amid customer budget concerns?
      A: Satya stated that they feel very good about Azure's position, as it's becoming a port of call for anyone doing an AI project, attracting new customers and seeing traction in adjacent services like data and dev tools. They are also witnessing ongoing migrations to Azure.

    3. Generative AI Investments and CapEx Spending
      Q: How does Microsoft quantify opportunities behind large AI investments?
      A: Satya emphasized allocating capital to maintain leadership in AI, both in training large models and managing inference demand. Amy added that the opportunity is significant, impacting every business process, and they will continue investing confidently, similar to their approach over the past decade.

    4. Azure Stabilization and Core Workload Growth
      Q: Can you expand on Azure's stabilization and core workload growth?
      A: Amy noted a balanced activity between new workload starts and optimizations, consistent with what they've seen throughout the cloud transition. Core Azure showed acceleration between Q2 and Q3, driven by new projects not just in AI but across other workloads, including migrations from on-prem to cloud.

    5. Microsoft 365 Copilot and Capacity Constraints
      Q: How do capacity constraints impact Microsoft 365 Copilot and Azure AI growth?
      A: Amy clarified there are no capacity constraints on Copilots, and their priority is to optimize capacity allocation to ensure per-user businesses can grow. Capacity constraints, when they occur, are more on the Azure infrastructure consumption side.

    6. AI Being Accretive to IT Spending
      Q: When will AI lead to net increases in IT spending?
      A: Satya believes that as AI tools like Copilot become standard, they shift operational expenses into tool spending, providing operating leverage. This requires cultural and process changes within organizations, and while it may take time, the rate of adoption is faster than any seen previously, evidenced by Copilot's rapid uptake.

    7. Copilots Competition and Partner Strategy
      Q: How will Microsoft's Copilots strategy adapt to industry competition?
      A: Satya explained that Microsoft's Copilot integrates deeply with business processes and applications like ServiceNow, SAP, and Salesforce. They are building a Copilot that acts as an orchestrator of other Copilots, bridging knowledge worker tools with business applications to enhance productivity.

    8. Data Quality Issues in GenAI Adoption
      Q: Is data quality a barrier to leveraging new GenAI capabilities?
      A: Satya acknowledged that successful AI deployment requires good data for reasoning and grounding. He noted that tools for addressing data quality and model tuning are maturing, enabling companies to integrate their data with AI projects effectively, especially in enterprise scenarios.

    Related Publications