Microsoft Earnings: AI Business Hits $37Bn Run Rate as Copilot Passes 20 Million Seats

A stable Q3 beat from Microsoft underlines that enterprise AI spending is moving from exploration to commitment, and the company's own products are among the clearest beneficiaries. Will its stock price follow suit?

4
Microsoft Earnings: AI Business Hits $37Bn Run Rate as Copilot Passes 20 Million Seats
Unified Communications & CollaborationNews

Published: April 30, 2026

Kieran Devlin

Microsoft closed its fiscal third quarter of 2026 with revenue of $82.9 billion, up 18 percent year-over-year, beating Wall Street’s forecast of $81.39 billion. The Q3 2026 earnings marked a notable bounce for Microsoft after it recently endured its worst Wall Street Quarter since 2008.

The results were driven by continued strength in cloud infrastructure and a rapidly expanding AI business that Satya Nadella, Microsoft’s chairman and CEO, described as now running at a $37 billion annual revenue run rate,Β up 123 percent year-over-year.

Copilot Adoption Accelerates, With Usage Metrics to Match

The standout commercial development in the quarter was the continued momentum of Microsoft 365 Copilot. Paid seats crossed 20 million, up from 15 million in January, with year-over-year seat adds growing 250 percent. Enterprise commitment is also deepening. Accenture now has over 740,000 seats, the largest single Copilot deployment to date, while Bayer, Johnson & Johnson, Mercedes, and Roche each committed to 90,000 or more seats this quarter.

More telling than the seat count is how those seats are being used. Copilot queries per user rose nearly 20 percent quarter-over-quarter, and monthly active usage of Microsoft’s first-party agents has grown 6x year-to-date.

Nadella said:

β€œNearly 90 percent of the Fortune 500 now have active agents built with our low-code, no-code tools, and we are seeing fast growth of our Copilot credit consumptive offer, up nearly 2x quarter-over-quarter as customers increasingly extend Copilot with custom agents tailored to their workflows.”

Nadella stated that weekly engagement with Copilot has reached the same level as Outlook. This reference point suggests, for a growing cohort of enterprise users, the tool has become a routine part of the working day rather than an occasional experiment. He added, β€œWe are focused on delivering cloud and AI infrastructure and solutions that empower every business to optimize their outcomes in the agentic computing era.”

Breaking Down Microsoft’s Financial Health With Its Latest 2026 Earnings

Net income rose 23 percent to $31.8 billion, and diluted earnings per share came in at $4.27 against an expected $4.06.

The Microsoft Cloud generated $54.5 billion in revenue, up 29 percent, while Azure and other cloud services grew 40 percent, fractionally ahead of the 39 percent analysts had penciled in. The Intelligent Cloud segment, which includes Azure, GitHub, and Nuance, reached $34.7 billion, up 30 percent year over year.

The Productivity and Business Processes segment, covering Microsoft 365, Dynamics, and LinkedIn, contributed $35.0 billion, up 17 percent. Microsoft 365 Commercial cloud revenue rose 19 percent, Dynamics 365 climbed 22 percent, and LinkedIn added 12 percent.

Capital expenditure came in at $31.9 billion for the quarter, up 49 percent, though below the $34.9 billion analysts had expected. Gross margin slipped to 67.6 percent, its lowest since 2022, reflecting higher depreciation from new data center builds and growing AI product usage. For Q4, Microsoft guided total revenue of $86.7 billion to $87.8 billion, with Azure growth expected at 39 percent to 40 percent in constant currency.

What the 2026 Earnings Numbers Reflect About the Broader Market

Microsoft’s results are a useful gauge of where enterprise AI investment is heading. The company’s commercial remaining performance obligations, contracted revenue not yet recognized, reached $627 billion, up 99 percent year over year, indicating that a significant volume of future cloud and AI spending has already been committed by customers.

One development worth watching is Microsoft’s evolving pricing model. During the analyst Q&A, Nadella confirmed that per-user businesses, covering productivity, coding, and security, will transition to a combined per-user and consumption model, with metered billing for agent usage above base entitlements. For enterprise finance and procurement teams, this is a critical practical change that will require updated cost modeling and contract governance before agent usage scales.

The Practical Takeaway for Enterprise Buyers and Microsoft Customers

For organizations using or evaluating Microsoft’s productivity stack, the quarter offers a few concrete signals. Agent mode is now the default experience in Word, Excel, and PowerPoint. A new first-party speech model, MAI-Transcribe-1, which shows a 67 percent improvement in GPU efficiency, is being developed to power transcription in Microsoft Teams; a development worth tracking for anyone invested in meeting intelligence or contact center workflows.

On the enterprise use case side, HSBC reported a 30 percent reduction in customer issue resolution time using pre-built Dynamics 365 agents. Bayer has built an in-house agent platform on Microsoft Foundry with more than 20,000 active monthly users. Neither example is a controlled pilot. Both reflect production deployments inside large, complex organizations.

The broader picture is straightforward. Enterprise AI spending is growing, Microsoft is capturing a significant share of it, and the products at the center of that growth are increasingly present in the daily workflows of the users who have them. For organizations still in evaluation mode, the peer benchmarks emerging from deployments of this scale are becoming harder to ignore.

Final Takeaway

Arguably, the most quietly significant number in Microsoft’s Q3 was not the revenue figure, but the $627 billion in contracted, not-yet-recognized obligations, up 99 percent in a year. That is the sound of enterprise technology decisions already made, by organizations that have moved past the question of whether AI belongs in the workflow and on to the harder work of making it perform. It’s worth knowing where your own organization sits on that spectrum.

CCaaSCorporate FinanceUCaaSUCaaSUCaaS & CCaaS Convergence​
Featured

Share This Post