Is Microsoft Overvalued? A Framework, Not a Verdict

The question this article will and will not answer

Asking whether Microsoft is overvalued is really asking a sharper question: what does today's price already expect the business to do, and can it deliver that? We will not stamp the stock cheap or dear. We will show you the business as it stands in July 2026, read the price as a forecast, set that forecast against Microsoft's own record, and then lay out both honest readings so you can settle it yourself. You can pull the live numbers alongside this from the Microsoft report on Tenet.

The method matters because "overvalued" is not a fact about a company. It is a relationship between a price and a set of expectations. Change the growth you assume and the same price flips from expensive to fair. So the work is to make those assumptions explicit.

What Microsoft is today

Microsoft is a software and cloud business of enormous scale, and its revenue mix tells you where the growth case lives. In fiscal 2025, which ended in June 2025, the company reported revenue of about $282 billion and net income of roughly $102 billion, according to its 10-K. Server products and cloud services, the line that contains Azure, was the single largest source at about $98 billion.

Figures are from Microsoft's fiscal 2025 10-K and Tenet data as of July 2026, listing the largest lines rather than every category. Two things stand out. First, cloud and enterprise software, not Windows or gaming, are now the center of gravity: Azure and the commercial cloud drive the story, which is why demand for artificial-intelligence services is watched so closely. Second, this is a business built on recurring, contracted revenue, the kind that renews rather than resets each year. When you buy the stock, you are mostly buying that stream of enterprise cloud and subscription income and a bet on how long it keeps compounding. The deeper business view is in how we analyze Microsoft.

The quality checklist

Before touching the price, the business has to clear a quality bar, and Microsoft clears it with room to spare. We look for durable margins, heavy cash generation, sensible capital use, and a defensible position. Run the fiscal 2025 figures from Tenet data:

• Margins. Gross margin was about 69 percent and operating margin about 46 percent, both high and steady, the signature of software economics.
• Cash generation. Microsoft produced roughly $136 billion of operating cash flow and about $72 billion of free cash flow, the cash left after capital spending.
• Capital allocation. It returned large sums to shareholders, about $18 billion in buybacks and $24 billion in dividends, while still investing heavily in data centers.
• Competitive position. Enterprise software carries steep switching costs: once a company runs on Azure, Microsoft 365 and its identity tools, moving off them is slow, costly and risky.

One number deserves a note rather than applause. Microsoft's free cash flow, though enormous, has grown more slowly than net income lately because capital spending on data centers has climbed sharply, to roughly $65 billion in fiscal 2025. That spending is a bet on artificial-intelligence demand. If the demand shows up, the capital builds the next leg of growth; if it disappoints, that cash was spent against a weaker return. Read the balance sheet next to it: Microsoft held about $95 billion in cash and short-term investments against roughly $112 billion of total debt, much of it lease obligations, so its finances are sound though no longer a pure net-cash fortress. The point of a checklist is to surface exactly this trade-off rather than wave it through.

What today's price already assumes

This is the half most buyers skip, and it is where the overvaluation question actually lives. In July 2026 Microsoft traded near $387 a share, a market value around $2.9 trillion, against fiscal 2025 diluted earnings of $13.64. That is a trailing price-to-earnings multiple of about 28.

Judge that against Microsoft's own recent history, not against other companies. Over the prior five fiscal years the stock generally traded in the high 20s to high 30s times earnings, touching roughly 40 times at its 2025 high before pulling back. At about 28 today, buyers are paying near the lower end of that band for each dollar of earnings. On its own history, then, the multiple is not stretched. The harder question is what that multiple assumes.

Here is the useful way to frame it, an approach sometimes called expectations investing: rather than asking "what is Microsoft worth," ask "what does this price already expect Microsoft to do." A multiple of 28 carries an earnings yield near 3.5 percent, meaning a year of current earnings buys you about three and a half cents on the dollar. For that to become a strong return, earnings have to grow, and keep growing. The price is a forecast, and buying the stock endorses it. That mindset is developed in understanding market expectations, and the underlying idea of what a business is truly worth is the subject of intrinsic value. Building that worth up from the cash the segments generate is the exercise walked through in step-by-step company valuation.

Two honest readings follow from the same number, and both are defensible.

• The "not overvalued" reading. Microsoft grew earnings per share from $5.76 in fiscal 2020 to $13.64 in fiscal 2025, a compound rate near 19 percent, behind a widening moat in cloud and enterprise software. If Azure and artificial-intelligence services keep that engine running, 28 times earnings is a fair or even modest price for a business of this durability, and the multiple leaves room to expand.
• The "overvalued" reading. A $2.9 trillion company cannot compound at 19 percent forever; the law of large numbers pulls growth down. If earnings growth fades toward the low teens while capital spending stays heavy, then paying 28 times leaves thin compensation, and any stumble in cloud demand could compress both earnings and the multiple at once.

A quick sense check makes the stakes concrete. Suppose Microsoft grows earnings at 12 percent a year for a decade and the multiple drifts from 28 down toward the low 20s as growth matures. A buyer would still do reasonably well, but much of the return from earnings growth would be offset by the shrinking multiple. Now suppose cloud and artificial-intelligence demand keep earnings compounding in the mid-to-high teens and the market keeps paying a premium. The same starting price looks comfortably rewarding. The two assumptions, growth and exit multiple, swing the outcome across a wide range, and small changes in either move the answer more than the current multiple does.

Notice that the disagreement is not about the multiple. It is about the growth rate you plug in behind it. That is where your judgment does the work.

What would have to be true for you

Because we are not handing you a verdict, the work now shifts to you, and it narrows to a few questions only you can answer. This is the "what would have to be true" test, the same reasoning laid out in building an investment thesis: for Microsoft at today's price to reward you, several things must hold, and you have to judge how likely each is.

1. Does Azure keep compounding? Much of the growth case rests on cloud and artificial-intelligence demand growing at a healthy clip for years. If that decelerates faster than expected, 28 times earnings gets harder to defend.
2. Do the data-center bets pay off? Microsoft is spending roughly $65 billion a year building capacity. Whether that spending earns a strong return, or merely defends share in a capital-hungry race, changes the free-cash-flow picture materially.
3. What return do you actually need? At an earnings yield near 3.5 percent, your return depends on growth clearing a real bar. If you demand a wide margin of safety, this price may not offer one; if you are buying quality to hold for a decade, you may weigh it differently.
4. What is your time horizon? A demanding multiple is far easier to hold for ten years than for one, because time lets earnings grow into the price. Your holding period changes the math, a point explored in when to buy a stock.

There is no answer key. Two careful investors can run this list and disagree, because they bring different growth views, return targets and horizons. That is normal, and it is why the label cannot be outsourced.

So, is Microsoft overvalued?

That is yours to decide, and now you have the pieces to decide it well: a business that clears the quality bar comfortably, a multiple that sits inside its own historical range, and a growth assumption that quietly decides the whole answer. Our job was to give you the framework and the current figures, not the conclusion. To go deeper, read how we analyze Microsoft for the full business view, then open the live Microsoft report on Tenet and test the growth you would need against the numbers as they stand today.