The important change is not that Microsoft became a worse business. It’s that the stock stopped pretending gravity was optional. At roughly $390 a share and a 23.2x P/E, $MSFT sits well below its own 52-week high near $555 and well below the kind of premium investors were paying when “AI” was enough to suspend arithmetic market pricing and current finance data. For a company that produced about $318.3 billion in trailing revenue, $125.2 billion in trailing net income, and roughly 39% net margin, that reset matters because valuation is finally doing some of the work for you, instead of all of it.
Start with the business. In its fiscal third quarter ended March 31, 2026, Microsoft reported $82.9 billion of revenue, up from the prior year period, with $31.8 billion in net income and $46.7 billion in operating cash flow in the earnings release. Management also said Microsoft Cloud revenue reached $42.4 billion, up 20% year over year, and that Azure and other cloud services grew 33% in the same release. That is the whole case in one sentence: this is a mature company still putting up growth numbers that would make younger software names smug on social media.
Now the valuation. Using the current market cap of roughly $2.91 trillion and trailing revenue of about $318.3 billion, $MSFT trades near 9.1x sales current quote and financials. On earnings, the market is paying about 23x for a company with a fortress balance sheet, more than $78 billion in cash and short-term investments at the latest quarter end, and just $57.0 billion of total debt Microsoft quarterly balance sheet data. That is not “cheap” in cigar-butt terms. But for elite quality, it is far from absurd.
Compared with the Magnificent Seven, Microsoft looks unusually sane. $AAPL trades around 37.4x earnings with a market cap above $4.5 trillion current finance data, yet its business is more exposed to hardware cycles and replacement timing than Microsoft’s subscription-heavy model. $GOOGL is cheaper at about 27.5x earnings and remains a serious alternative, but its profit engine is still materially tied to advertising even as Google Cloud improves current finance data and Alphabet results. $AMZN trades around 29x earnings current finance data, but that multiple flatters a quarter helped by a $16.8 billion pre-tax gain from Anthropic and masks the fact that trailing free cash flow collapsed to $1.2 billion because AI capex is eating like a teenage linebacker Amazon Q1 2026 results. Microsoft’s accounting is simply cleaner.
Against enterprise software, Microsoft also holds up well. Oracle’s latest quarter was undeniably strong: fiscal Q4 2026 revenue rose to $15.9 billion, cloud revenue grew 27%, OCI revenue jumped 62%, and remaining performance obligations reached more than $138 billion Oracle’s FY26 Q4 release. That explains why $ORCL commands about 25.2x earnings despite a much smaller and narrower platform current finance data. If Oracle deserves a mid-20s multiple for accelerating cloud infrastructure, Microsoft does not look expensive at 23x for a broader moat, better diversification, and stronger ecosystem lock-in.
The real objection is not valuation. It is capex discipline and AI monetization. Every hyperscaler is spending like an oil wildcatter who just found religion. Amazon said it expects about $200 billion of capital expenditures in 2026 in its February 5, 2026 release. Alphabet warned that depreciation growth would accelerate in 2026 as a result of elevated investment on its Q4 2025 earnings call page. Oracle is leaning hard into AI data center demand in its latest release. Microsoft is in the same arms race. If enterprise customers prove slower to turn copilots and AI workloads into durable budget lines, returns on all that spending could lag the narrative.
Still, Microsoft has a better setup than most because it already owns the customer relationship. It does not have to invent a new habit; it can attach AI to Office 365, GitHub, Dynamics, Azure, and security products that enterprises already buy company overview and earnings materials. That matters. Selling more to an installed base is cheaper than persuading the world from scratch. Buffett would call that a nice way to avoid unnecessary heroism.
So: is now a good time to take a position in $MSFT? Yes — buy. Not because it is statistically cheap versus the broad market. It isn’t. Buy because among mega-cap compounders, Microsoft offers one of the better combinations of quality, durability, balance-sheet strength, and now a valuation that no longer demands immaculate execution. If you want the cheapest multiple in big tech, buy something else. If you want the best business at a merely fair price, Microsoft is near the top of the list.
What to watch next is straightforward: when Microsoft reports again, will Azure growth and AI attach rates remain strong enough to justify the data-center spend, or will capex start outrunning monetization in a way the market can no longer ignore?