Microsoft FY2026 Q4 Earnings: Azure Tops $100 Billion — What It Means for the Moat

Microsoft (MSFT) closed its fiscal 2026 on July 29 with a fourth quarter that answered the year's defining question — is all that AI spending buying anything? — with the clearest evidence a cloud business has yet produced. Revenue of $90.0 billion grew 18%, and Azure accelerated. This article covers two layers: what Microsoft actually reported, and what the numbers say about the structure of the company's moat.

What Microsoft reported

The quarter, ended June 30, 2026, came in ahead of consensus estimates on both lines: revenue of $90.0 billion, up 18% year over year, and diluted EPS of $4.81, up 32%. Net income rose 31% to $35.8 billion.

The segment split tells the real story. Intelligent Cloud grew 32% to $39.3 billion. Productivity and Business Processes grew 14% to $37.8 billion. More Personal Computing — Windows devices, Xbox, advertising — shrank 4% to $12.9 billion. Microsoft is now, decisively, a cloud company with a consumer attachment.

Three details stood out in the report:

• Azure grew 43%, accelerating from 40% the prior quarter, at a base that crossed $100 billion in annual revenue for the first time (up 41% for the full year). Management guided the next quarter to roughly 45% growth in constant currency.
• Microsoft 365 Copilot passed 30 million paid seats, with quarterly net additions more than doubling.
• The full fiscal year landed at $331.8 billion in revenue (+18%), operating income of $155.2 billion (+21%, a 46.8% operating margin), and net income of $133.7 billion (+31%).

The number under the number: free cash flow

Here is the tension the headlines hide. Microsoft spent roughly $41 billion on capital expenditure in the quarter alone — about 70% more than the prior-year quarter — and around $116 billion for the full year, much of it on servers and GPUs. The result, as of MoatScan AI's July 31, 2026 financial analysis, is full-year free cash flow of $67.0 billion against net income of $133.7 billion. Microsoft is converting about half its profit into cash; the other half is being poured into data centers. CFO Amy Hood's explanation was blunt: demand continues to exceed available supply, and capital spending will grow again in fiscal 2027.

The contrast with Meta (META), which reported the same week and also raised its capex outlook, is instructive at the business level. Microsoft can point to capex arriving alongside accelerating, contracted cloud revenue; Meta rents no capacity to outside customers, so its spending has to be repaid entirely through its own products. The same line item means different things in the two businesses — capex that shows up as accelerating revenue behaves like investment, while capex that shows up only as a bigger bill carries more risk.

The honest caveat is that GPUs are short-lived assets. Today's capex becomes tomorrow's depreciation expense, and consensus forecasts already reflect it: the FY2027 figures shown on our MSFT page have revenue growing 17.6% to $390.1 billion but EPS growing only about 9% to $19.50. Earnings growing at half the pace of revenue is what a depreciation wave looks like in advance.

What the quarter says about the moat

MoatScan AI's moat analysis of Microsoft (dated June 26, 2026 — a month before this report) scores it 77/100, Wide Moat, with Switching Costs the top pillar at 9/10: "deep enterprise lock-in through Microsoft 365, Windows, Azure, identity management." The earnings are, mostly, evidence for that thesis.

Switching costs compound with every layer a customer adopts, and Copilot is a new layer. An organization that has rolled out AI assistants tuned to its own documents, permissions, and workflows has quietly raised its own cost of ever leaving Microsoft — much as Apple's device ecosystem does for consumers, but with procurement contracts attached. And Azure passing $100 billion means workloads, data, and identity infrastructure of the kind that migrates roughly never.

The AI question cuts both ways, though — a tension covered in How AI Is Reshaping Economic Moats. The opportunity side is on full display: Azure is the toll road for AI compute demand. The threat side is quieter but real. Some slice of that 43% growth is training and inference demand from AI startups whose own economics are unproven, and if AI agents eventually become the interface to enterprise software, per-seat lock-in gets renegotiated. A Wide Moat rating is a judgment that the lock-in outlasts the transition. It is a judgment, not a certainty.

The financials, and the assumptions behind the model

As of MoatScan AI's July 31, 2026 analysis, Microsoft carries a Financial Score of 83/100 — Income Statement 8.5, Balance Sheet 8.5, Cash Flow 8, Key Ratios 8, Growth 8.5. The Cash Flow 8 is the honest grade in the set: exceptional generation, currently suppressed by the buildout.

The discounted cash flow model behind the analysis is best read as a set of assumptions rather than a single number. Revenue is modeled compounding at 9.2% — well below the 18% Microsoft just printed and consensus forecasts of 17.6% for FY2027, so the model already assumes heavy deceleration. Against that conservatism sit a 7.9% WACC and 4.0% terminal growth, which are on the generous end; moving the discount rate by a single point changes the output substantially. DCFs on mega-caps are sensitivity exercises, and the most useful question this one raises is how long capex intensity stays at today's levels — the input the model is least able to know. How each of these inputs works is explained in DCF Valuation Explained.

The scorecard

A quarter this clean still leaves the real questions open: whether AI demand at the current intensity is durable, whether the depreciation wave compresses margins faster than revenue can outrun it, and whether enterprise lock-in survives an agent-shaped interface shift. What the quarter did establish is that, so far, Microsoft is a company whose AI spending is visibly turning into revenue rather than just a bigger bill.

The live version of everything quoted here — the five-pillar moat breakdown, the six financial sections, the valuation model with its assumptions — is on MoatScan AI's MSFT analysis, and it will update when the analysis is next re-run; this article deliberately freezes the July 31, 2026 snapshot. For how Microsoft's moat compares with other companies in the archive, see the wide moat stocks list.

MoatScan AI is a free educational research tool. This article explains how a business and its competitive position work; it is not investment advice or a recommendation to buy, sell or hold any security. Figures are as of the dates stated and will change.