The Capex Tell

Is your growth in-line with your peers in B2B SaaS & AI? 

Benchmark yourself against actual billings data for Maxio’s 2000+ global customers.

Key takeaways from the report: 

Download the Report

This is the single most important earnings week of 2026 for anyone with meaningful equity exposure. Microsoft reports after the bell tonight. Meta and Alphabet already reported. Apple and Amazon land tomorrow. These five companies, together with Nvidia, represent a disproportionate share of the S&P 500 by market cap, which means their guidance on artificial intelligence capital expenditure is not just a story about tech. It is the story about the broader index, the story about your 401(k), and the story about whether the soft landing narrative holds through the back half of the year.

And yet almost every retail investor I know is reading this week wrong. They are watching revenue beats. They are celebrating earnings per share surprises. They are getting excited about stock buybacks. These are the surface metrics that cable TV covers because they fit in a chyron. The actual institutional read on this week is happening one layer deeper, in the capital expenditure disclosures, the operating margin trajectory, and the commentary on return timelines for AI infrastructure spending.

Let me walk you through how allocators are reading these reports, what the signals actually mean, and what you should do in your own portfolio this week regardless of which company beats or misses.

The four hyperscalers, Microsoft, Meta, Alphabet, and Amazon, are collectively on track to spend over 470 billion dollars on capital expenditures in 2026, up from roughly 350 billion in 2025. Let that sink in. Four companies are deploying nearly half a trillion dollars of capital into data centers, chips, energy infrastructure, and networking equipment in a single calendar year. This is larger than the GDP of most countries. It is the largest sustained corporate investment cycle in modern history.

The question the market is trying to answer this week is not whether the spending will happen. It is already happening. The question is whether the revenue and margin profile of AI services will scale fast enough to earn a reasonable return on that capital. If the answer is yes, the hyperscalers are at the beginning of a multi-year earnings compounding cycle that will carry the index higher. If the answer is no, we are looking at the biggest capital misallocation event since the late 1990s telecom buildout, and the consequences will not be confined to tech.

Institutional desks are not listening for the beat. They are listening for four specific things: the ratio of AI-related revenue to AI-related capex, the commentary on useful life and depreciation schedules for data center equipment, the operating margin trajectory despite the spending, and the gap between stated capex guidance and what analysts were modeling before the call. Each of these tells you something the top-line number cannot.

A simple sanity check. If a company is spending 100 billion dollars per year on capital equipment with a useful life of five to six years, the incremental revenue those assets generate needs to clear a threshold that makes the spending economically rational. Depreciation alone, ignoring any other cost, will be roughly 18 to 20 billion per year. Factor in power, cooling, real estate, and the cost of the chips that need to be replaced every three years or so, and the all-in cost of running the infrastructure is significantly higher than the depreciation line suggests.

For the spending to make sense, AI-related revenue needs to grow at a pace that covers the depreciation, the operating cost, and a reasonable return on equity. Roughly speaking, every dollar of annual capex should be associated with 30 to 50 cents of incremental annual revenue within two to three years. Anything below that range and the math starts to break down.

Right now, the hyperscalers are in a window where capex is outpacing AI revenue by a wide margin. That is not necessarily bearish. It is what you would expect during a buildout phase. The question is whether the gap is closing. In the coming earnings reports, look for disclosures about the growth rate of AI-attributable cloud revenue, and compare it to the growth rate of capex. If the revenue line is accelerating faster than capex, the investment thesis is working. If capex is accelerating faster than revenue, the thesis is getting stretched.

This is arcane, but it matters. In late 2022 and 2023, several hyperscalers extended the useful life assumptions on their server equipment from four to six years. That accounting change alone added billions to their reported operating income. In 2025, we began to see a reversal, with some companies quietly shortening useful life assumptions again because the pace of GPU obsolescence has been faster than the extended schedule implied.

Watch for language in the 10-Q filings, not just the earnings call, about changes in useful life or accelerated depreciation. A shortened useful life means higher non-cash depreciation expense, lower reported operating margin, and a tacit admission that the infrastructure spend is less efficient than the company previously implied. It does not change cash flow, but it does change the reported earnings that the market uses to value the stock.

If one of the hyperscalers announces a useful life adjustment downward this quarter, treat it as a more important disclosure than any revenue beat.

The tension in every hyperscaler call right now is simple. They want to preserve operating margins while spending unprecedented amounts on capex. Some of this works through the accounting. Capex sits on the balance sheet and depreciates over time, so the immediate hit to the income statement is smaller than the check they are writing. But operating costs, particularly power and cooling for data centers, show up in the P&L right away. Analyst consensus for several of these companies now anticipates the narrowest operating margins in three years.

If a company reports a margin that beats the lowered consensus, the stock will likely rally, but the actual signal is whether the margin is stabilizing or still compressing. A stabilizing margin with continued revenue growth is the bullish case. A margin that keeps compressing while revenue growth decelerates is the bearish case. The headline number does not tell you which regime you are in. You have to read the commentary and the segment detail.

Before each of these companies reports, analysts publish their expected capex figure for the coming fiscal year. When the company provides its own guidance, the gap between the consensus estimate and the stated number is often the most volatile moment in the after-hours session. Last year, when one of the hyperscalers guided capex 10 billion above consensus, the stock dropped 9 percent overnight despite a revenue beat and an earnings beat. The same company’s stock rallied 7 percent when it guided below consensus a quarter later, even though revenue was slightly weaker.

The point is that the market has become acutely sensitive to capex signaling because investors are trying to price the probability that these buildouts will earn a return. Guidance that comes in above consensus is now frequently interpreted as a red flag, not a green one. This is a reversal from 2023 and 2024, when higher capex was treated as bullish evidence of AI demand.

If you own an S&P 500 index fund, roughly 30 percent of your equity exposure is tied to the fate of these five companies. Your portfolio is already exposed to the capex question whether you think about it or not. The right response is not to sell the index. The right response is to understand what you own and to hold a thoughtful view on whether the current concentration is appropriate for your time horizon.

There are three practical adjustments to consider this week. First, check the actual weight of the largest seven or eight tech names in your portfolio. If you hold a broad market index, a total market ETF, and a Nasdaq 100 fund, you may be running a 40 or 45 percent effective allocation to the same handful of companies without realizing it. This is exactly the kind of stealth concentration that an aggregated view on a platform like Empower’s dashboard will surface immediately.

Second, consider whether your equity exposure is balanced between growth and quality factors. In a regime where Big Tech capex guidance is driving index-level volatility, a pure growth-factor tilt will amplify the swings in either direction. A quality factor, which selects for high and stable return on equity, healthy margins, and low leverage, tends to dampen the volatility because it includes companies outside the AI capex story. A blend of both, weighted toward quality in the current regime, is the institutional default.

Third, do not try to trade the individual earnings reports. After-hours options volatility on these names is priced for exactly the event you are trying to trade. Unless you have a specific thesis and a specific edge, the expected value of an earnings play is negative after volatility crush. If you want exposure to the upside of a specific company, buy the shares and hold them. If you want protection against downside in a concentrated position, look at longer-dated puts that survive the event rather than zero-day options that depend on a specific overnight move.

The reason retail investors lose money in weeks like this is not because they pick the wrong names. It is because the emotional volume of earnings coverage pulls them off their plan. A CEO says something bullish on a call and the stock runs. A CFO gives soft guidance and the stock dumps. Investors react in real time, and those reactions compound into a portfolio that looks nothing like the one they set out to build.

The institutional alternative is not cleverness. It is automation. Set your target allocation, set your rebalancing triggers, and let the system fire. Platforms like M1 Finance let you build a pie-based allocation with automatic rebalancing, which removes the step where human emotion typically destroys value. When Microsoft guides capex above consensus and the stock gaps down 6 percent in extended hours, the automated response is to rebalance into weakness if the allocation has drifted. The human response is to panic and sell. Guess which one wins over a decade.

The automation framework I walk through in the Wealth Architecture Blueprint is not about trading. It is about removing the moments when you would otherwise override your own plan. That is what institutions do by design, through risk committees and policy constraints. You can do it through tooling.

Spend 20 minutes looking at the capex guidance from Meta, Alphabet, and Microsoft. Do not read the coverage. Read the actual press releases or the 10-Q commentary. Note whether guidance for the coming fiscal year is above, below, or in line with what analysts were modeling. That one data point will tell you more about the next six months of the index than any pundit segment.

Check your effective tech concentration. If more than 35 percent of your equity book is tied to the magnificent seven, decide if that is intentional.

Write down one question you want answered by the time the week is over. Not a prediction. A question. Something like, is the AI revenue line at the hyperscalers growing faster than capex? If the answer is yes after this week, that is your signal to stay overweight tech. If no, that is your signal to rotate toward quality.

Do nothing else. Do not trade the earnings. Do not buy zero-day options. Do not cash out because someone on TV said the bubble is popping. The week is a data-gathering exercise, not a trading event.

The Wealth Architecture Blueprint walks through the full framework for setting target allocations, monitoring factor exposures, and automating rebalancing so that weeks like this one produce information rather than panic. To get the Blueprint, reply BLUEPRINT and I will send it over.

For the Velocity framework, which is specifically focused on how capital compounds when you keep emotional decisions out of the loop, reply VELOCITY.

Share MSL with three people and I will send you the Portfolio Positioning Playbook on the house. Ten referrals gets lifetime premium access. Your link is in the footer.

That’s 5 sales in 4 months. 28 all time. And the performance?

14.6%, 17.6%, and 17.8%, net annualized returns on sold works held longer than one year (See all 28 at Masterworks.com).

It’s not from stocks, private equity, or real estate… it’s from contemporary and post war art. Crazy, right?

With Masterworks, you don’t need to be a BILLIONAIRE to invest in multi-million dollar art anymore.

Historically, the segment overall has had attractive appreciation and low correlation to stocks.*

Masterworks targets works featuring legends like Banksy, Basquiat, and Picasso, identifying what they believe to have significant long-term appreciation potential, not just at the artist level but at the level of individual artworks.

As one of the largest players in the art market, with $1.3 billion invested over 500 artworks, they pass critical advantages through to their 70,000+ members to add art to their portfolios strategically.

Looking to diversify your investments in 2026?

Click here to skip the waitlist

^{*According to Masterworks data. Investing involves risk. Past performance is not indicative of future returns. See important Reg A disclosures at masterworks.com/cd.}