Imagine a business whose cloud division more than doubles in a year, whose total revenue climbs 25%, and whose shares get hit for about 13% the next trading session. That is not a hypothetical. It is what happened to Alphabet this week, and if you only read the revenue line you would have no idea why.
The answer sits lower on the page, in a number that has been quietly climbing up the list of things investors care about. Free cash flow. Alphabet reported negative free cash flow of about $9.07 billion for the third quarter, against analyst expectations for a positive figure of roughly $2.70 billion, according to Alphabet’s results and CNBC’s report of the Street estimate. The market did the math and did not like the answer.
The quarter in two columns
Start with the good news, because there was plenty. Alphabet reported third quarter revenue of about $127.6 billion, up 25% from a year earlier. Google Cloud revenue came in around $30.7 billion, up 103%, according to Alphabet’s earnings release. A business growing that fast is not a business in trouble.
Now the other column. Capital expenditures for the quarter ran about $51.4 billion. That was actually a little below the roughly $55 billion analysts had penciled in, so the quarter itself was not a spending surprise. The surprise was the road ahead. Alphabet maintained its full-year 2026 capex range of $195 billion to $205 billion — a range it had raised from a $190 billion ceiling in the second quarter — and said 2027 spending would increase significantly, with further detail promised on the fourth-quarter call. Trailing twelve-month free cash flow, per Alphabet’s results, fell 73% from a year earlier to about $19.75 billion.
Read that again. A company is telling you that next year it plans to spend something like 50% more than this year, in a year when it just burned cash in a quarter. Maybe that spending earns a fortune. Maybe it does not. But the market gets to ask the question, and this week it asked loudly.
There is also a usage story underneath the revenue. Alphabet said its first-party model APIs now process nearly 32 billion tokens per minute, more than four times a year ago, and that the Gemini app has reached 1 billion users. Those are the kinds of numbers that make a bull case sound easy. They also explain why the company feels it cannot slow down, because every one of those tokens runs on hardware someone had to buy, power, and cool.
Why growth stopped being enough
For most of the past two years, the AI trade ran on a simple story. Demand was enormous, supply was short, and anyone spending to build capacity would be paid back. Revenue growth was the proof. Every earnings season that showed cloud acceleration reinforced the story.
That logic works until the bill gets large enough to matter on its own. When capex runs into the hundreds of billions, the question changes from “is demand real?” to “what does each dollar earn, and when?” Those are different questions, and the second one is much harder to answer with a growth rate. A 103% jump in cloud revenue tells you demand exists. It does not tell you the return on a $300 billion budget.
There is another wrinkle. Cloud growth that fast also means customers are signing up for capacity Alphabet has to build before it can bill for it. Spending comes first. Revenue arrives later. In a boom that gap is easy to ignore, and in a cash flow report it shows up in black and white.
If you want a simple mental model, think of a restaurant that is packed every night and decides to build three more locations at once. Sales are great. The kitchen is busy. But the checking account looks thin until the new doors open, and anyone lending money to the owner is going to ask how long that takes. Bigger restaurants are not the problem. Timing and funding are.
Microsoft told a different story with the same ingredients
Contrast this with Microsoft, which also reported this week. Microsoft posted revenue of about $92.3 billion, up 19%, with Azure growing 48%. Free cash flow was positive at roughly $18.6 billion, per Microsoft’s results and CNBC’s coverage.
Same industry, same AI demand, same enormous infrastructure bills. Different cash outcome. Reports differ on exactly how Microsoft’s capital spending for the quarter should be measured, with the sources I reviewed giving different figures, so I will not lean on a single number here. The point does not depend on it. Microsoft showed growth and positive cash generation in the same quarter, and its shares were reported roughly flat to modestly higher while Alphabet’s fell hard.
That gap is the real story of the week. The market is no longer grading AI spenders as one group. It is sorting them by who can fund the buildout from operations and who is leaning on the balance sheet to do it. 24/7 Wall St. reported that Meta, which posted negative free cash flow of about $2.12 billion in its own October 21 report, and Amazon, whose shares also fell after its own AI-heavy print, face the same test.
What this does and does not mean
It would be easy to overreact in either direction, so a few cautions are in order. One quarter of negative free cash flow is not a collapse. Companies with this much revenue growth often invest ahead of demand, and negative quarters can reverse. Alphabet’s Google Cloud business doubling is a real operating fact, not an accounting trick.
But it would be just as easy to dismiss the market’s reaction as noise, and that would be a mistake too. Investors punished the stock because the combination of rising capex, negative cash flow, and a much larger 2027 budget raises the cost of being wrong. If AI demand keeps compounding, the spending looks smart in hindsight. If it slows even modestly, a company steering toward hundreds of billions more in 2027 spending has little room to adjust.
Notice what changed. Six months ago the debate was whether anyone would pay for AI. This week it was whether the people building it can afford to keep going at this pace without stretching their finances. That is a more mature debate, and arguably a healthier one.
What to watch next
A few signals are worth tracking for anyone following this trade. First, the next set of cash flow statements. Does free cash flow recover as new capacity comes online and starts billing, or does it keep sliding as budgets grow? Second, how these companies pay for the buildout. Higher spending funded by operating cash is one thing. Funding it with new borrowing or share issuance is another, and it changes the risk profile. Third, the guidance language. Capex ranges that keep moving up are a different signal than ranges that hold steady.
Also watch whether the market keeps treating strong growth as a free pass. This week suggests the answer, at least for now, depends on what is happening underneath the growth. Investors who only track revenue headlines may keep being surprised by price moves that seem to contradict them.
The takeaway is not that AI spending is a mistake. It is that the scoreboard has changed. Revenue growth got companies into the conversation. Cash flow decides how the market prices them once they are there. Reading earnings with that in mind will not predict any single stock move, but it will make sense of days like this one.
This article is written for educational and informational purposes only and does not constitute financial or legal advice. The views and analytical frameworks presented draw on publicly available information and reported commentary from industry participants. Readers are encouraged to consult primary sources and form their own informed views on these complex topics.
