Meta Stock Q2 Earnings Preview: Why AI Capex Matters More Than Revenue

Meta hasn't missed a revenue estimate in over two years. That isn't really what's on trial when the company reports its second-quarter results after the US market closes on July 29. The real tension sitting under every research note this week is a math problem, not a mood. 

Meta told investors in April it would spend $125 billion to $145 billion on AI infrastructure this year. Three months later, it had used up barely a sixth of that budget. Somewhere in the next two quarters, the spending has to catch up, and Wednesday's print is the first real checkpoint on whether that catch-up looks controlled or alarming.

Let's break down what Wall Street expects, what options traders are betting on, everything Meta has been through since its last report, and the specific arithmetic that decides whether this quarter calms investors or spooks them further.

Meta Q2 2026 Earnings Estimates: Revenue, EPS and Capex

*Q1 2026's headline GAAP EPS was $10.44, inflated by an $8.03 billion one-time tax benefit tied to US tax law changes. $7.31 is the underlying figure analysts actually measure against.

Sources: Meta's Q1 2026 and Q2 2025 earnings releases filed with the SEC, and Street consensus as reported by Nasdaq, TipRanks and 24/7 Wall St. in late July 2026.

Two things stand out, and neither is really about revenue. Underlying EPS growth is expected to land close to flat year over year, around 1%, even as revenue grows near 27%. That gap between the top line and the bottom line is the real story of this earnings season for Meta. 

Second, management hasn't given a standalone capex number for the quarter, only a full-year range, which is exactly where the pressure sits (more on that below). One smaller line worth knowing: Reality Labs lost $4.03 billion in Q1 on $402 million of revenue, and CFO Susan Li has guided full-year losses to land "on par" with 2025's roughly $19 billion, so don't expect a surprise there either way.

What Analysts and Options Traders Expect From Meta Earnings

Coverage on Meta remains heavily bullish on paper. Depending on the aggregator, somewhere between 38 and 57 analysts carry a Buy or Strong Buy rating, with consensus price targets clustering between roughly $814 and $836. Bernstein's Mark Shmulik has kept his Buy rating into the print, and Mizuho's Lloyd Walmsley carries an Outperform rating with an $835 target, calling Meta his preferred pick in digital advertising. These are third-party analyst opinions, not something we're endorsing, and targets like these can shift within days of a print, so treat them as a snapshot rather than a forecast.

The options market, generally a cleaner read of what money is actually positioned for than research notes are, is pricing in a post-earnings move of roughly 7.4% in either direction, as per data reported by TradingKey and TS2 in the days before the print. Meta has averaged a post-earnings move of about 8.8% over its last five reports.

Key Meta Developments Since Its Last Earnings Report

A lot has happened since April 29. Here's what actually matters heading into Wednesday.

Sources: Reuters, Bloomberg, CNBC, Morningstar and Goodwin's antitrust update, reported through late July 2026.

Two of these are worth a beat of extra attention. The Anthropic talks matter because Meta is currently the only major hyperscaler without a cloud business it can point to as proof its AI spending earns money outside advertising. Anything concrete on this front during Wednesday's call, even directional, would be the first real receipt for a strategy Zuckerberg has floated since May. And the bond story matters because it's quieter but arguably more honest: investors are now demanding meaningfully higher yields to fund Meta's data centres than they did nine months ago, which is the credit market's own way of pricing AI infrastructure risk, independent of what the stock does.

The IOU Quarter: Why Meta’s AI Capex Matters More Than Revenue in Q2

Here's the number that actually decides this earnings season for Meta:

Meta actually spent less on capex in Q1 2026 than analysts expected, $19.84 billion against a roughly $27.6 billion Street estimate, even while raising full-year guidance in the same breath. That combination only resolves one way: whatever Q1 didn't spend has to show up in the next three quarters. Run Meta's own guidance through the math and you land at $35 billion to $42 billion a quarter for the rest of 2026, more than double what Meta itself spent in the same quarter last year, and sitting right next to the exact figure that spooked Alphabet's stock a week ago.

Think of it the way an Indian household thinks about a loan EMI. Commit to a certain repayment schedule for the year, underpay the first instalment, and the bank doesn't quietly forgive the balance, the remaining instalments just get heavier. Meta effectively owes the market a much bigger instalment this quarter and next than it paid in Q1. The open questions are whether it pays on schedule, pays early and spooks the stock with a fourth guidance raise, or falls behind again and pushes the problem further out.

There's a second, quieter number worth knowing, even though it won't show up in Wednesday's press release. Investor Michael Burry argued in November 2025 that hyperscalers, Meta among them, have been stretching the "useful life" of servers and network gear on their books, from roughly 2-3 years of real-world use to 5-6 years on paper, which understates depreciation and flatters reported profit today. Meta's own 2025 annual filing disclosed exactly this kind of change, cutting that year's depreciation by about $2.9 billion. Burry estimated this could overstate Meta's earnings by as much as 20.8% by 2028. Other analysts, including engineers at rival cloud providers, have pushed back hard, arguing real-world hardware utilisation doesn't support a gap that steep. 

Meta Q2 Earnings Scenarios: What Could Move the Stock

This is a map for reading the release with your eyes open, whichever way you decide to act on it. Worth remembering going in: Meta beat on both revenue and EPS last quarter and still fell 6-7% after hours purely on the capex number. A beat alone hasn't moved this stock higher in either of its last two reports.

Meta Stock Before Q2 Earnings: What Investors Should Consider

The bull case, which most of Wall Street currently holds, rests on real fundamentals: close to 27% expected revenue growth, ad pricing power that hasn't cracked, and a stock trading at roughly 18-22 times forward earnings depending on the estimate used, below its own three-year average of about 23 times. If the AI spending starts showing up as cloud revenue or durable ad-engine gains without another guidance shock, today's price could look cheap in hindsight.

The caution case rests on the IOU math above: a company that has raised its own capex guidance twice this year, is now borrowing at rising yields to fund it, and where EPS growth has stalled near zero even as revenue keeps climbing. A 7.4% options-implied swing means whichever way this breaks, it probably breaks hard, and buying right before a print like that is a different risk decision than adding to a position once the dust has settled.

Our Take

Meta's advertising engine isn't really in question this quarter. It has beaten revenue estimates for years and shows no obvious sign of stopping. What's genuinely uncertain is whether Meta can spend $105 to $125 billion in six months without either scaring investors with a bigger bill or quietly admitting the money is moving slower than planned. 

That's a sharper, more answerable question than "is AI a bubble," and Wednesday's print settles it either way. Whatever your starting view on Meta, watch the capex number before you watch the revenue number this time. It's doing all the talking.