Microsoft reported on 29 July 2026 that fiscal 2026 revenue rose 18% to USD 331.8 billion, while Azure revenue passed USD 100 billion and Microsoft 365 Copilot reached more than 30 million paid seats. The Wall Street Journal’s framing of AI as only part of the problem fits the results: the surrounding cloud platform, capacity and operating model now carry much of the work.
The figures matter for architects, admins and IT managers because they show AI demand arriving through existing Microsoft estates. Copilot adoption, Azure consumption and cloud commitments are becoming budget and platform questions, not only model selection questions.
Cloud growth in the figures
Microsoft Cloud revenue was USD 59.3 billion in the fourth quarter of fiscal 2026, up 27% year on year. Intelligent Cloud revenue was USD 39.3 billion, up 32%, and Azure and other cloud services revenue increased 43%.
The earlier quarters showed the same direction. In the second quarter, Microsoft Cloud revenue crossed USD 51.5 billion and Azure and other cloud services revenue increased 39%; in the third quarter, Microsoft Cloud revenue reached USD 54.5 billion and Azure and other cloud services revenue increased 40%.
Microsoft also said in the third quarter that its AI business had passed an annual revenue run rate of USD 37 billion, up 123% year on year. The company does not split Azure growth by workload in the earnings release, so the figures do not show how much of the increase came from particular AI services or other cloud consumption.
Copilot scale and Microsoft 365 impact
Microsoft 365 Copilot reached more than 30 million paid seats by the end of the fiscal year. Microsoft 365 Commercial cloud revenue increased 14% on a reported basis in the fourth quarter, or 16% when adjusted for a prior-year comparison that benefited from 2 points of in-period revenue recognition.
For Microsoft 365 administrators, the paid-seat figure is an adoption signal rather than a deployment design. The earnings release does not state revenue per Copilot seat, customer distribution, or how many seats are in active daily use.
Commercial remaining performance obligation increased 84% to USD 678 billion in the fourth quarter. That figure points to contracted demand, but it does not remove the need for tenant-level planning, cost control and change management in individual organisations.
Infrastructure spending becomes visible
Microsoft’s capacity build-out appears in its cash flow and balance sheet. Additions to property and equipment were USD 115.9 billion for fiscal 2026, compared with USD 64.6 billion in fiscal 2025.
In the fourth quarter alone, additions to property and equipment were USD 35.8 billion, compared with USD 17.1 billion in the same quarter a year earlier. Net property and equipment stood at USD 313.1 billion on 30 June 2026, up from USD 205.0 billion on 30 June 2025.
The cost side also moved. Service and other cost of revenue was USD 94.3 billion for fiscal 2026, compared with USD 74.3 billion in fiscal 2025, while Intelligent Cloud cost of revenue rose to USD 57.9 billion from USD 40.2 billion.
Microsoft’s own risk language reflects the same issue. The fourth-quarter release named inability to develop and expand adequate infrastructure, as well as outages, disruptions or capacity constraints, among the factors that could affect results.
What customers should take from it
The earnings releases show a business with large AI and cloud demand, but they do not answer every planning question for customers. They do not provide regional capacity detail, workload-level margins, or a breakdown of Copilot usage beyond paid seats.
That leaves the practical work with enterprise teams. Architecture decisions still need to account for region choice, data placement, latency, resilience, networking, identity, monitoring and spend controls.
The OpenAI investment line also affects earnings comparisons. Microsoft reported full-year GAAP net income of USD 133.7 billion, up 31%, while non-GAAP net income excluding the impact from OpenAI investments was USD 128.8 billion, up 22%.
For technology teams, the clearer operational signal is the combination of Azure growth, Copilot paid seats and infrastructure spending. AI adoption is now tied to the same disciplines that already govern cloud platforms: capacity, cost, resilience and governance.
Turning Copilot seats, Azure growth and AI demand into running services brings the pressure back to Azure architecture, hybrid cloud, cost and platform engineering, which ECS’s Infrastructure topic covers. The programme gives attendees sessions and practitioner experience on the design and operational choices behind these deployments. Bring the capacity, cost and governance questions that are difficult to settle from quarterly figures alone.
Photo: Derrick Coetzee from Berkeley, CA, USA, CC0, via Wikimedia Commons, cropped
This article was written by the ECS editorial team with the assistance of Amal Agent, run.events' AI assistant, which supported the research, drafting and proofreading.