The Latest Microsoft Earnings Reassured Investors. Meta’s Didn’t.

Microsoft (NASDAQ: MSFT) and Meta (NASDAQ: META) reported earnings within a day of each other this week, and both companies are spending record amounts on the same bet: artificial intelligence infrastructure. Microsoft earnings beat every major number analysts were watching, and the stock jumped 8 percent in after-hours trading. Meta beat on revenue and still got punished. Same spending story, same 24-hour window, two completely different verdicts from the market.

What Microsoft Earnings Got Right

Microsoft’s fiscal fourth-quarter results released Wednesday cleared the bar analysts had set and then some. Revenue came in at $90.0 billion against expectations of roughly $87.6 billion. Diluted earnings per share reached $4.81, or $4.74 on an adjusted basis, comfortably ahead of the $4.24 adjusted consensus. Net income rose to $35.8 billion.

Azure, Microsoft’s cloud business, was the headline. Cloud revenue reached $59.3 billion for the quarter, up 27 percent from a year earlier, and Azure itself grew 43 percent, its fastest pace since early 2022, according to Microsoft’s own earnings release. For the full fiscal year, Azure revenue crossed $100 billion for the first time. Microsoft 365 Copilot passed 30 million paid seats.

There’s a detail worth flagging, honestly. Part of the EPS beat came from a $3.2 billion gain tied to Microsoft’s stake in AI lab Anthropic, which added 33 cents per share. Microsoft noted separately that earnings would have been 7 cents lower had the same accounting applied to its OpenAI stake instead. That doesn’t make the underlying quarter weak. Azure’s growth and the Copilot seat count are operating numbers, not accounting ones. But it’s a reminder that a headline EPS beat isn’t always a clean read on the business.

Not everything grew. Windows and devices revenue fell 7 percent. Xbox content and services dropped 10 percent. Those are smaller pieces of the business, and the market largely looked past them.

Retail sentiment mostly matched the professional read. One widely upvoted post on Reddit summed up the buy-the-dip mood with a promise to “buy the dip and keep buying” regardless of near-term swings. That’s not analysis, it’s conviction, but it tracks with how the stock actually traded this week.

Where Meta’s Quarter Fell Apart

Meta’s numbers tell a different story, and the gap is instructive. Revenue came in at $60.8 billion, up 28 percent year over year and ahead of estimates. But diluted earnings per share landed at $6.18, well short of the roughly $7.18 to $7.23 analysts expected. Net income fell to $15.8 billion, down 14 percent.

The real damage was underneath the headline. Operating margin dropped from 43 percent to 31 percent, partly on $2.4 billion in legal charges and $1.18 billion in severance tied to May layoffs. Free cash flow collapsed to $784 million against $31.08 billion in quarterly capital spending. Reality Labs, Meta’s virtual and augmented reality division, posted a $4.62 billion operating loss on just $431 million in revenue, a narrower loss than Wall Street expected, but still a loss on a business generating almost nothing in return.

Meta narrowed its full-year capital spending guidance to $130 billion to $145 billion, raising the floor from its earlier $125 billion estimate and holding the ceiling, even after missing on earnings. That’s the detail that unsettled investors most. The company isn’t slowing down the spending that just squeezed its margins. It’s committing to more of it at the low end.

One financial newsletter covering the print warned that Microsoft’s spending pattern risks becoming “a poor investment” if the cash flow math doesn’t hold, a caution that applies just as directly to Meta’s quarter. On Bogleheads, one forum member summed up the broader unease that’s been building around this cohort of stocks all year, pointing out that the largest technology names now “make up 35% of the S&P 500,” a concentration level that magnifies exactly this kind of single-week divergence for anyone holding the index.

The One Line That Decided Both Stock Moves

Strip away everything else and the difference comes down to one question: is the spending showing up anywhere yet? For Microsoft, the answer this quarter was yes, in the form of accelerating Azure growth and a backlog of signed contracts. For Meta, the answer was no, at least not in a way that offset the cost of building it.

UBS analyst Karl Keirstead flagged the mechanics behind Microsoft’s own soft spots, noting that Microsoft shifted scarce computing capacity toward its own products, and that “both Azure and the M365 segments fell a bit short,” a nuance that didn’t stop the stock from rallying but is worth watching next quarter. Bold Wealth Partners chief investment officer Jason Lemire told Fortune that investor sentiment on AI capital spending has inverted: “It used to be the more the better, but now it is the less the better.”

That shift is the real story of this earnings season, and it’s the same one this site covered when Alphabet and Tesla posted a similar capex-versus-cash-flow split two weeks ago. The pattern is repeating company by company. Spending alone no longer buys goodwill. Spending that visibly converts to growth does.

Two Ways to Read This Week’s Numbers

There are two reasonable ways to respond to a week like this, and they lead to different actions depending on what you’re actually deciding.

• If you’re contributing to a retirement account or brokerage account on a regular, automated schedule, this week’s earnings prints aren’t a reason to change anything. The data on trying to time contributions around single earnings reactions, in either direction, consistently favors staying on schedule over guessing right. One good or bad quarter doesn’t change that math.
• If you’re specifically weighing whether to add new, discretionary money to Microsoft because of this report, that’s a different decision, and it deserves a different process. An 8 percent pop confirms that this quarter cleared expectations. It doesn’t tell you whether the stock’s current price already assumes Azure will keep growing at 43 percent indefinitely, which is much harder to bet on repeating.

If this week’s split verdict has you second-guessing which company “deserved” its stock move, Nassim Taleb’s Fooled by Randomness is worth a look. It’s a clear-eyed case for why one quarter’s result, good or bad, rarely tells you as much as it feels like it does.

TheCapitalist.com may earn a commission from qualifying purchases through this link. This is an educational partner placement and does not affect our editorial coverage.

Both of those positions are defensible. What’s not defensible is treating a single quarter’s stock reaction as proof of anything beyond what it is: the market’s read on one earnings report against the specific numbers it was expecting going in.

Microsoft earnings bought the company a quarter of patience. Whether that patience extends to next quarter depends entirely on whether Azure’s growth rate holds, and whether Meta can show the same kind of return on its own spending before investors run out of it.

For educational purposes only. Not financial advice.

Frequently Asked Questions

Reader Poll: