Meta Platforms (META): Q2 2026 results, my verdict

2026-07-29 · By Lubin Danilo, founder of Lubin Investment

META: see the full analysis on Lubin Investment

Meta reported on July 29, 2026 revenue of $60.8 billion (+28%), above consensus, but earnings per share of $6.18, down year over year due to an explosion in AI infrastructure spending. My quality screen validates 23 out of 25 criteria. Yet my model judges the stock heavily overvalued. Here is why.

What just happened: advertising holds up, profit falls

Meta Platforms reported after market close on July 29, 2026 its second-quarter 2026 results: revenue of $60.8 billion, up 28% year over year and slightly above analyst expectations ($60.26 billion estimated). The engine remains advertising on Facebook, Instagram, and Meta's other apps, where demand shows no sign of slowing. But earnings per share came in at $6.18, down from $7.14 in the same quarter last year, and well below the $7.23 the market expected. Operating margin (the share of revenue left after operating costs) fell to 31%, from 43% a year earlier. For the current quarter, Meta guided to revenue between $61 and $64 billion.

The cause of this gap between accelerating revenue and falling profit is known and openly owned by management: capital expenditures (capex, the money spent on data centers and AI chips) reached $31.08 billion in this quarter alone, and Meta now targets between $130 and $145 billion of capex for the full year 2026. CEO Mark Zuckerberg summed up his confidence in the trajectory of Meta's own AI models: 'I'm quite happy with the trajectory that we're on. We've released a few models that I think are quite impressive on our early scaling ladder, and we're in the process of scaling much larger and more advanced models.' In other words, Meta is knowingly sacrificing near-term profit to fund a technology race it considers decisive (the same capex trade-off I analyzed at Microsoft the same day in my guide on capex, good or bad signal).

Why my quality screen validates 23 out of 25 criteria

For very large companies like Meta, where rich qualitative data is available (governance, competitive position, market structure), my screen expands from its usual 10-criteria model to 25, to capture more nuance. Meta passes 23, a very high score. The fundamentals remain excellent: a 32.8% net margin (nearly a third of revenue ends up as net profit), 17.8%/year sales growth on average over five years, a share count declining 1.65%/year thanks to buybacks, and net debt that would take just 1.36 years of free cash flow to repay.

Qualitative factors matter just as much in this score. Meta's moat (the competitive advantage that protects the business), as identified by my screen, rests on two pillars: network effects (the more users on Facebook and Instagram, the more valuable the platform becomes for each of them) and economies of scale. Meta has no major client dependency (its biggest advertiser accounts for less than 15% of revenue), keeps gaining market share against rivals like TikTok, and its growth remains organic rather than bought through acquisitions. On governance, Mark Zuckerberg has been founder and CEO for 21.8 years (well above the 5-year threshold I prefer for judging a track record of decisions) and owns roughly 13.5% of the company's shares, a very strong personal alignment of interests. In June 2025, Meta invested $14.3 billion in the startup Scale AI, valued at $29 billion, and hired its CEO Alexandr Wang to lead its AI efforts, a capital-allocation bet my screen views favorably.

Two points, however, cause the score to fail or add nuance. Cash conversion (the share of accounting profit that actually becomes available cash) reaches only 37%, far below the greater-than-100% target: exactly the same phenomenon seen at Microsoft this quarter, where money earned goes straight back into building AI infrastructure before it has time to accumulate as cash. And my screen flags that Meta's share buybacks are not always executed opportunistically (meaning bought more heavily when the stock is cheap), which can reduce their efficiency for shareholders over time.

Reality Labs: an expensive bet, but no longer just a black hole

Reality Labs, the division behind the Quest virtual reality headsets and connected AI glasses, generated $431 million in revenue this quarter, up 16% year over year thanks to strong growth in AI glasses, offsetting lower Quest headset sales. But the division also posted an operating loss of $4.62 billion for the quarter alone. This deficit is nothing new: Meta has owned it for several years as the price of building a position in what it considers the next computing platform after the smartphone. The difference from two years ago is that AI glasses are starting to show real commercial traction, changing the nature of the bet: less a distant science-fiction scenario, more a product gradually finding its market.

The price: a premium of more than 60% according to my model

Meta currently trades at 57.9 times its trailing-twelve-month free cash flow, a rich multiple that has mechanically climbed this year precisely because capex is compressing cash generated today, while the future revenue this investment promises has not yet shown up. My reasonable buy-price model, which projects the actual five-year cash-per-share generation trajectory, comes out at $233.27, versus a current price of $585.61: a premium of roughly 60%, one of the largest in my entire coverage universe this quarter.

The question to settle is similar to Microsoft's, but in the advertising business rather than cloud: is the market today paying for an AI promise (generative models, connected glasses, better-targeted ads) that still needs to be proven in real cash, or does this promise justify a legitimate premium given Meta's dominant position in its core advertising business? The 23-out-of-25 quality score says the underlying business is solid. The price says you already need to firmly believe in the success of Meta's AI bet to avoid overpaying today.

What could derail the thesis

The most direct risk remains capex return on investment, as with all tech giants this earnings season. A Meta-specific risk adds to that: near-total dependence on advertising revenue, highly sensitive to a broad economic slowdown that would cut advertiser marketing budgets. Finally, Reality Labs continues to represent a multi-billion-dollar-per-quarter bet whose large-scale commercial success still needs to be confirmed over several years.

Meta reported on July 29, 2026 revenue of $60.8B (+28%, above consensus), but earnings per share of $6.18 down year over year, as operating margin fell from 43% to 31% under a $31.08B AI capex bill this quarter ($130-145B guided for 2026).
My quality screen validates 23 out of 25 criteria: 32.8% net margin, 17.8%/year sales growth, network-plus-scale moat, founder CEO with 13.5% ownership. Weak point: cash conversion of only 37%, like Microsoft, absorbed by AI investment.
Reality Labs (Quest, AI glasses) generates $431M in revenue (+16%) but a $4.62B operating loss this quarter, a bet Meta owns on the next computing platform.
The 57.9x P/FCF is very high. My model targets a $233.27 buy price against a $585.61 current price, a premium of roughly 60%. This is not investment advice, do your own research.

FAQ

Why is Meta's earnings per share falling when revenue grows 28%?

Because AI infrastructure capital spending ($31.08 billion this quarter) pushed operating margin down from 43% to 31%. Advertising revenue is growing well, but a growing share of that money goes straight into capex rather than net profit.

What is Reality Labs, and why does Meta tolerate its losses?

Reality Labs covers the Quest virtual reality headsets and AI-connected glasses. The division loses $4.62 billion per quarter, but Meta sees it as a bet on what could become the next computing platform after the smartphone, funded by advertising's very solid profits.

Why is Meta's quality score out of 25 instead of 10?

For very large companies where rich qualitative data is available (governance, competitive position, market structure), my screen expands from its usual 10-criteria model to 25 to capture more nuance. Meta passes 23 of them.

Should you buy Meta stock after these results?

The quality of the business is real (23 out of 25 criteria), but my reasonable buy-price model targets $233.27 against a $585.61 price, a premium of roughly 60%. Paying this price means firmly betting on the success of Meta's AI bet. This is not personalized investment advice, do your own research.

About the author

Written by Lubin Danilo, founder of Lubin Investment. A self-taught individual investor, I find fundamental analysis fascinating, and it has delivered excellent results. For three years now, my performance has beaten the S&P 500. But analyzing every stock took too much time: sites with incomplete data, calculation methods and criteria never aligned with mine. And spotting the best stocks was just as time-consuming, even with my own well-defined checklist. So I put my software development background to work to build this software, base my investment strategy on its results, and share it with people who share the same passion as me. It judges a company's quality and its price separately, using criteria drawn from the financial literature (Warren Buffett, Michael Mauboussin, Aswath Damodaran).