Microsoft sells the software the working world runs on: Windows, Office, and the Azure cloud, now with AI assistants layered across all of it. It is the most complete software franchise on earth. At $509, you pay a full price for that completeness.
Scale is 0–50%/yr. The shape of the verdict: a fine compounder whose base case clears neither hurdle. Only the bull case clears the 12% growth bar.
Azure grew 43% last quarter and crossed $100 billion in annual revenue for the first time. The contracted backlog sits at $678 billion, roughly two years of future revenue already signed. When the growth engine has that much fuel booked, the debate shifts from "will demand show up" to "can they build fast enough."
Microsoft's trick is distribution: AI assistants are being sold into the hundreds of millions of seats it already owns through Office and Windows. That is a monetization path no AI startup can copy. The open question is how much customers will keep paying once the novelty fades.
Office subscriptions, Windows, LinkedIn, gaming, and cloud: when one leg wobbles, the others carry. Full-year revenue crossed $332 billion, up 18%, with net margins above 40%. This diversification is why Microsoft rarely has a truly bad year.
Microsoft is spending enormous sums on AI datacenters. So far the returns justify it: revenue up 18%, margins intact. The risk is that the spending continues after the demand curve flattens. Every quarter is a check on whether the buildout is still earning its keep.
This is a wonderful compounder at a full price. At 28 times earnings, the market already believes the Azure and Copilot story. The base case (~9.4% a year) is a fine return and a weak substitute for either hurdle. The note stays warm for a price that lets quality do the compounding.
At 28x earnings and $3.8 trillion, expectations are sky-high. A single quarter of Azure deceleration would re-rate the stock hard. Perfection is the assumption; anything less is the risk.
Datacenter spending is at historic levels. If AI demand growth slows, Microsoft owns a lot of expensive capacity. The backlog covers about two years; beyond that, the bet has to keep paying.
AWS is reaccelerating and Google Cloud is growing fast. Azure's lead is durable but not unassailable, and price competition in cloud never fully goes away.
Microsoft's bundling of AI into Office and Windows attracts antitrust attention on multiple continents. Remedies could slow the Copilot monetization flywheel.
Growing 18% on $332 billion means finding $60 billion of new revenue a year. Every year the hurdle gets higher. Compounders this large eventually become the market, not beat it.