- Threshold crossed. Microsoft reported on July 29 that Azure revenue exceeded $100 billion on an annual basis for the first time in the company’s history, as Q4 FY2026 Azure growth accelerated to 43% in constant currency — well ahead of the 40% the company had guided.
- Fact. Microsoft 365 Copilot reached more than 30 million paid seats during the quarter, more than doubling sequentially, with customers holding 50,000 or more seats rising sevenfold year on year.
- Stake. The results arrive as investors debate whether enterprise AI adoption is generating enough revenue to justify the sector’s capital expenditure cycle — Microsoft’s numbers, alongside concurrent beats from Amazon and Alphabet, suggest adoption is accelerating rather than hitting a ceiling.
Azure Accelerates Past Guidance
Microsoft’s Q4 FY2026 results, released after market close on July 29, showed the Intelligent Cloud segment — which houses Azure — generating $39.3 billion in revenue for the quarter, up 32% year on year. Azure’s own growth rate accelerated to 43% in constant currency, surpassing management’s prior guidance of 40% and reversing the trend of sequential deceleration that had concerned investors earlier in the fiscal year.
Total company revenue for the quarter reached $90 billion, up 18%, against analyst expectations of around $87.7 billion. The full fiscal year 2026 exceeded $331 billion, with commercial remaining performance obligation — a forward-revenue indicator — reaching $678 billion, including commitments tied to Microsoft’s investment in OpenAI.
CEO Satya Nadella characterised Copilot’s traction in terms that go beyond adoption metrics: “This is the first time where you really have an enterprise-wide tool, which has both a per-seat and usage-based pricing,” he said on the earnings call, framing the product as a new category of software rather than an add-on to existing Microsoft 365 plans.
Copilot’s Enterprise Breakthrough
The 30-million paid-seat figure for Microsoft 365 Copilot is the headline that enterprise software analysts had been waiting for. The company added a net 10 million seats during Q4 alone, and usage intensity — measured by weekly active engagement — now sits at levels comparable to established products like Outlook and Teams. That last metric matters: high-engagement software is stickier and harder for competitors to displace.
GitHub Copilot, the code-generation product aimed at developers, posted revenue growth exceeding 60% in Q4 following a shift to usage-based pricing. The product now counts enterprise customers across industries that previously had no reason to interact with developer tooling, as non-engineering roles adopt AI-assisted workflows.
The Copilot momentum stands in contrast to earlier anxieties about enterprise AI adoption speed. As recently as a few months ago, Intel’s 59% AI revenue jump was read partly as infrastructure pull-forward rather than evidence of end-user demand. Microsoft’s Copilot numbers now provide the demand-side confirmation that the infrastructure thesis needed.
Capital Expenditure and Capacity
Microsoft spent $41 billion in capital expenditure during Q4, adding 31 new data centres across five continents — part of a full-year total of 88 new facilities opened in FY2026. For Q1 FY2027, the company guided capital expenditure above $50 billion, with roughly two-thirds directed at short-lived assets like CPUs and GPUs to expand capacity quickly. The long-lived portion — land, buildings, and physical infrastructure — represents a slower-moving bet on where demand will be in five to ten years.
Demand currently exceeds available capacity, which is the key constraint on Azure’s near-term growth. Management guided approximately 45% constant-currency Azure growth for Q1 FY2027, suggesting the acceleration seen in Q4 is expected to continue rather than mean-revert, as new capacity comes online.
The combination of accelerating cloud growth, proven Copilot adoption, and forward guidance that implies no near-term ceiling positions Microsoft’s results as the strongest case yet that the current AI infrastructure cycle is translating into durable enterprise revenue — not just anticipated future value.