Every Alphabet earnings report now feels like two companies walking onto the same stage.
The first is the Google most of us already understand. It owns the world’s most important search engine, operates YouTube, sells an ocean of digital advertising, and generates more cash than many countries could locate with both hands and a flashlight.
The second is a vastly more expensive creature being assembled behind the curtain. It consumes data centers, chips, electricity, engineers, fiber, cooling systems, and capital at a rate that would make an industrial empire feel financially self-conscious. This is Alphabet’s AI future, and management is spending as though arriving second would be a form of corporate extinction.
As I look ahead to Alphabet’s next earnings report, expected to cover the third quarter of 2026, I am not primarily asking whether the company can beat Wall Street’s revenue estimate by half a billion dollars. Alphabet generated nearly $120 billion in revenue last quarter. At that scale, a minor quarterly beat can be little more than an accounting breeze passing through a financial skyscraper.
I am asking three larger questions.
Can advertising continue growing as AI changes how people search?
Can Google Cloud preserve its increasingly impressive margins while expanding capacity?
And can Alphabet prove that its enormous AI spending is building a profitable future rather than merely constructing the world’s most expensive answer machine?
Those questions are connected. Search supplies the cash. Cloud offers the next major profit engine. AI threatens, supports, and complicates both.
That is what makes the next report so important.
The Starting Point Is Almost Absurdly Strong
Before previewing the next quarter, I have to acknowledge what Alphabet just accomplished.
In the second quarter of 2026, Alphabet’s revenue rose 24% year over year to $119.8 billion. Google Services revenue increased 15% to $94.5 billion, while Google Cloud revenue surged 82% to $24.8 billion. Consolidated operating income climbed 30% to $40.8 billion, and Alphabet’s operating margin expanded from 32% to 34%. Alphabet’s official second-quarter release shows a business that was not merely growing—it was accelerating across several of its most important operations.
Those numbers should have produced a clean victory celebration. Instead, investors did what investors always do when handed excellent results: they immediately searched the room for something to worry about.
They found capital expenditures.
Alphabet raised its expected 2026 capital spending range to between $195 billion and $205 billion, up from its previous forecast of $180 billion to $190 billion. Shares fell in after-hours trading as the market attempted to reconcile tremendous Cloud growth with a spending plan that now resembles the annual economic output of a midsize country. Reuters reported that Alphabet’s higher capital-spending outlook overshadowed results that otherwise surpassed revenue expectations.
This reaction made emotional sense even if it looked strange on the surface.
Investors love growth. We love it slightly less when growth arrives carrying a construction invoice.
The central tension is no longer whether Alphabet can generate demand for its AI products. Demand appears abundant. The tension is whether the company can convert that demand into returns that justify the infrastructure required to serve it.
Alphabet is not struggling to find customers. It is struggling to build enough capacity.
That sounds like the kind of problem every business would love to have. It is still a problem.
Advertising Remains the Machine That Pays for Everything
The market may talk endlessly about Gemini, tensor processing units, autonomous vehicles, and artificial general intelligence, but advertising still pays most of Alphabet’s bills.
Google advertising revenue reached $81.6 billion in the second quarter, up 14% from the previous year. Google Search and other advertising revenue climbed 17% to $63.3 billion. YouTube advertising revenue increased 13% to $11.1 billion, while Google Network advertising revenue declined 1% to $7.3 billion.
This is the part of Alphabet’s financial statement I examine first because Search remains the company’s economic foundation.
Cloud may be the rapidly growing second engine, but advertising is still supplying the fuel, maintaining the airport, and quietly paying for the engineers designing the next aircraft.
The reassuring news is that Search has not collapsed under the arrival of generative AI. For several years, the bear case was straightforward: people would stop typing conventional searches into Google and begin asking AI assistants for direct answers. Fewer searches would mean fewer advertisements, fewer commercial clicks, and a gradual erosion of one of the greatest business models ever built.
That threat has not disappeared. But Alphabet’s recent results suggest the transition may be more complicated—and more favorable to Google—than the most pessimistic predictions implied.
Management says AI features are increasing query activity. AI Mode surpassed one billion monthly active users after its global expansion, while AI-powered Search experiences are reportedly encouraging people to ask longer and more complicated questions. Google also says these features continue sending billions of clicks to websites each week. Sundar Pichai’s second-quarter remarks presented AI as an expansion of Search behavior rather than a replacement for it.
That is exactly the story Alphabet needs investors to believe.
The company does not want AI to become a separate doorway through which users leave Google. It wants AI to enlarge the Google doorway.
What I Will Watch in Search Advertising
The 17% growth reported in Search and other advertising was excellent, but the upcoming comparison will be more demanding.
Alphabet’s chief financial officer warned that the company will begin comparing its third-quarter performance against an acceleration in Search that started during the same period last year. Currency movements may also shift from a modest tailwind in the second quarter to a slight headwind in the third. Those factors do not signal deterioration, but they make another spectacular growth rate harder to produce.
This is where investors must resist the temptation to grade a company using only one percentage.
If Search growth slows from 17% to a lower double-digit rate, the market may initially treat it as evidence that the great digital cash machine has developed a concerning rattle. I would look deeper.
I want to know whether query growth remains healthy. I want to hear whether advertisers are receiving stronger conversion rates from AI-assisted placements. I want evidence that new search formats can carry commercial messages without making the product feel like an interactive billboard trapped inside a chatbot.
That balancing act is more difficult than it sounds.
If Google monetizes AI results too aggressively, it risks damaging the experience and encouraging users to try competing products. If it monetizes too cautiously, AI queries may cost more to serve while generating less revenue than traditional searches.
Traditional search economics are beautiful because users perform much of the work. They type a query, Google retrieves a page of links, advertisers bid for placement, and the system responds quickly with relatively manageable computing costs.
Generative AI is hungrier. It must interpret the request, process context, run sophisticated models, compose an answer, and potentially use tools or agents to complete a task. Each interaction may create more value for the user, but it also requires more computing power.
Alphabet therefore has to solve two problems simultaneously: protect the revenue per interaction and reduce the cost per answer.
I will listen closely for any management discussion of monetization efficiency, advertiser adoption, commercial-query growth, and the serving costs associated with AI Mode and AI Overviews.
I do not need management to reveal every internal metric. I would, however, appreciate something more informative than “engagement is exciting,” a phrase corporations often use when the revenue has not yet arrived but everyone in the presentation would prefer to remain upbeat.
YouTube Deserves More Attention Than It Receives
YouTube is occasionally treated as a supporting character in Alphabet’s story, which is remarkable considering the platform generated more than $11 billion in quarterly advertising revenue before subscription revenue was even included.
YouTube advertising grew 13% in the second quarter. Management attributed the increase to strength in direct-response and brand advertising, particularly on television screens. The World Cup also gave the platform a significant boost, attracting a vast global audience and creating a premium environment for advertisers.
The next report will help reveal how much of that growth came from durable improvements and how much came from an extraordinary sporting event.
I do not consider event-driven strength illegitimate. If a platform attracts viewers when the world cares most about watching something, that is evidence of its relevance. Still, one-time audience surges can make future comparisons less flattering.
The more important development is YouTube’s presence in the living room.
YouTube is no longer simply a website people visit to watch a three-minute tutorial recorded in someone’s garage. It competes for television time, sports audiences, podcasts, music listening, short-form attention, premium subscriptions, and the vast human desire to watch strangers explain products they purchased yesterday.
Connected television can support larger brand budgets, while Shorts creates enormous engagement and additional advertising inventory. YouTube also gives Alphabet a defensive advantage that text-based AI competitors cannot easily duplicate. Search behavior may change, but humanity has shown no intention of abandoning video.
In fact, the more automated and synthetic the internet becomes, the more valuable trusted creators may become. People often want information from a recognizable human being, complete with facial expressions, experience, bias, enthusiasm, and the occasional microphone positioned directly in front of the person’s mouth for reasons future historians will struggle to explain.
In the next earnings call, I want updates on Shorts monetization, connected-TV viewing, subscription growth, and advertiser demand beyond the World Cup boost.
YouTube does not need to replace Search. It only needs to continue becoming a larger and more profitable part of the ecosystem.
Google Cloud Has Become Impossible to Treat as a Side Business
Google Cloud’s second-quarter performance changed the conversation.
Revenue increased 82% to $24.8 billion. Operating income more than tripled to $8.8 billion, while operating margin expanded from 20.7% to 35.6%. Backlog reached $514 billion, more than $50 billion higher than in the previous quarter. Alphabet expects to recognize slightly more than half of that backlog as revenue over the next 24 months, according to its second-quarter earnings call.
Those are not the financial characteristics of a small experimental operation waiting patiently in the corporate basement.
Google Cloud has become a major business with a quarterly revenue run rate approaching $100 billion and margins that would make many mature software companies look down at their shoes.
Its acceleration came from several sources: core Google Cloud Platform services, enterprise AI applications, AI infrastructure, security, Workspace, and the initial recognition of revenue from TPU systems delivered directly to customer data centers.
That last item matters.
Alphabet is no longer offering only rented computing capacity inside Google-operated facilities. It has begun selling complete TPU systems that customers can deploy in their own data centers. This expands the addressable market and gives Alphabet another way to monetize the specialized chips it has spent years developing.
It also complicates comparisons.
Hardware sales can generate significant revenue while carrying different margins than traditional cloud services. If TPU system sales become a larger component of Cloud revenue, headline growth may remain spectacular even as the revenue mix shifts.
I will therefore look beyond Cloud’s growth rate.
I want to understand the composition of that growth.
How much came from recurring consumption of cloud services? How much came from direct system sales? How quickly are customers expanding their AI workloads? And how much capacity is Google forced to obtain from outside providers because its own infrastructure cannot be built quickly enough?
An 82% growth rate is wonderful. It is also not a number I would casually extend into eternity using a ruler.
Cloud Margins May Become the Most Important Number in the Report
Google Cloud’s 35.6% operating margin was arguably the most impressive number in Alphabet’s second-quarter results.
A few years ago, Cloud was losing billions of dollars annually. Now it produces billions in quarterly operating income. The transformation shows what can happen when infrastructure utilization improves, revenue expands across a largely fixed operating base, and management decides a growing business should eventually perform the unfashionable task of earning money.
The next quarter may be more complicated.
Alphabet said it plans to use additional third-party capacity during the third quarter as a bridge while it builds more of its own infrastructure. This allows the company to meet customer demand, but management expects the arrangement to create modest near-term pressure on Cloud margins.
That warning deserves attention.
Third-party capacity is presumably more expensive than running workloads on infrastructure Alphabet owns and optimizes itself. It is the corporate equivalent of a restaurant becoming so popular that it has to rent the kitchen next door at peak rates. The demand is encouraging. The temporary economics are less attractive.
I would not interpret a moderate decline in Cloud margin as a failure if revenue growth and backlog remain strong. Capacity shortages can indicate real demand, and using external infrastructure may be sensible if it allows Google to win customers who would otherwise turn to Amazon or Microsoft.
But I do not want the phrase “temporary margin pressure” to become a permanent resident of every earnings call.
I will be watching for evidence that Alphabet can bring new internal capacity online fast enough to improve unit economics over time. I also want management to distinguish between margin pressure caused by strategic expansion and margin pressure caused by rising competition, pricing concessions, or inefficient spending.
Those are not the same problem.
One may create future value.
The others may simply create excellent conference-call vocabulary.
Alphabet Is Spending at a Scale That Demands Proof
Alphabet spent $44.9 billion on capital expenditures in the second quarter alone. Approximately 60% of its technical-infrastructure investment went toward servers, with the remaining 40% directed toward data centers and networking equipment.
The result was negative free cash flow of $5.9 billion for the quarter, even though operating cash flow reached $39.1 billion.
That sentence deserves to be read twice.
Alphabet generated $39.1 billion in operating cash and still produced negative free cash flow because it invested nearly $45 billion in capital assets.
This is not a struggling company burning cash to keep the office lights on. It is one of the world’s most profitable businesses deliberately channeling extraordinary amounts of money into infrastructure.
The revised 2026 capital-spending forecast of $195 billion to $205 billion represents a massive increase from what investors would have considered normal only a few years ago. Management also expects spending to rise significantly again in 2027.
At some point, shareholders are entitled to ask what all this construction will earn.
The optimistic answer is compelling.
Alphabet is building the computing foundation for a new generation of Search, advertising, cloud services, cybersecurity, enterprise software, scientific research, autonomous systems, and consumer agents. It owns specialized chips, world-class models, enormous distribution, deep technical talent, global data-center infrastructure, and products used by billions of people.
If AI becomes a foundational layer of the global economy, Alphabet may be building one of its most valuable toll roads.
The pessimistic answer is less comforting.
Every major technology company is spending heavily at the same time. Models may become cheaper and less differentiated. Competition could push prices downward. Open-source systems may weaken proprietary advantages. Customers may distribute workloads across multiple clouds. AI services could become commodities before the companies building them earn acceptable returns on their infrastructure.
In that version of the future, Alphabet will still own impressive data centers. They may simply produce less economic value than shareholders currently imagine.
The uncomfortable truth is that nobody knows which version will prevail.
Management does not know.
Analysts do not know.
I certainly do not know, although I at least have the courtesy not to place a 12-month price target beside my uncertainty and call it precision.
Depreciation Is Where Yesterday’s Spending Enters Tomorrow’s Income Statement
Capital expenditures do not hit earnings all at once. They become assets on the balance sheet and are expensed gradually through depreciation.
This gives large infrastructure programs a delayed financial effect.
Alphabet can spend enormous sums today while the full pressure on reported profit arrives over several future years. As more servers and data centers enter service, depreciation rises. Energy, maintenance, staffing, and other operating costs rise as well.
Management has already warned that higher depreciation and data-center operating costs will pressure the income statement. Alphabet also intends to continue hiring in AI and Cloud while spending more to market its AI products.
This means the next earnings report will require me to look beyond consolidated operating-margin expansion.
A 34% companywide operating margin is excellent, but segment and expense trends will tell me more about the future. I will watch research and development costs, which rose 32% in the second quarter. I will watch the losses associated with Alphabet-level activities, which reached $5.8 billion and primarily reflected shared AI research and development.
I will also watch the relationship between capital spending and incremental revenue.
If spending rises rapidly while Cloud revenue, AI subscriptions, and Search monetization also accelerate, the investment case remains credible. If spending climbs while revenue growth slows and management becomes increasingly philosophical about long-term opportunity, I will become more cautious.
There is a point at which “investing for the future” becomes a polite way of asking shareholders to stop examining the present.
Alphabet has not reached that point. But at $200 billion in annual capital spending, it has entered a neighborhood where the street signs deserve close reading.
The AI Products Must Eventually Produce More Than Engagement Statistics
Alphabet’s operational AI metrics are enormous.
The Gemini app reached 950 million monthly active users in the second quarter. Google’s first-party model APIs processed approximately 22 billion tokens per minute, up from 16 billion in the previous quarter. More than nine million developers were building with Google’s models each month. Nearly 90% of the Fortune 100 reportedly used Gemini Enterprise.
These figures demonstrate reach, adoption, and momentum.
They do not automatically demonstrate profit.
Technology companies love usage statistics because usage is impressive, easy to graph, and temporarily relieves everyone from discussing price. But infrastructure must ultimately be paid for with revenue and profit, not tokens processed per minute.
I want Alphabet to provide clearer evidence that AI is improving business economics.
In advertising, that could mean better targeting, higher conversion rates, greater advertiser returns, more queries, or new commercial formats.
In Cloud, it could mean rising consumption, larger contracts, stronger customer retention, and healthy margins.
In subscriptions, it could mean more paying Gemini and Google One users.
Within the company, it could mean measurable productivity improvements that slow expense growth or allow products to be developed faster.
I do not need every AI product to make money immediately. New platforms require investment. But the scale of Alphabet’s spending has shortened the amount of time investors will accept vague promises.
The company is no longer placing a few experimental chips on the AI table.
It has pushed forward an extremely large portion of the treasury.
The Reported EPS Number May Be Misleading
Alphabet reported diluted earnings per share of $9.11 in the second quarter, compared with $2.31 a year earlier.
Anyone seeing that number without context might assume Google had discovered how to quadruple ordinary profit between visits from the accountant.
The increase was primarily caused by $98 billion in other income, largely reflecting unrealized gains on Alphabet’s equity investments. These gains are legitimate under accounting rules, but they do not reflect the recurring economics of Search, YouTube, or Cloud.
Reuters reported adjusted earnings of approximately $2.85 per share, slightly below the consensus estimate of $2.89.
This is why I separate operating performance from investment gains.
Alphabet’s equity portfolio can create meaningful value, but unrealized gains are volatile. They may reverse. They should not receive the same valuation multiple as recurring operating earnings generated by advertising and cloud contracts.
In the next report, I will focus on revenue, operating income, segment margins, operating cash flow, capital expenditures, and free cash flow before becoming emotionally involved with headline EPS.
A spectacular GAAP number can be accurate and still be unhelpful.
Accounting has many talents. Preventing investor confusion is not always among them.
Competition Is Applying Pressure From Every Direction
Alphabet is fighting different competitors across different layers of its business.
In consumer AI, it faces OpenAI, Anthropic, Meta, and a growing collection of international and open-source model developers.
In Cloud, it competes primarily with Amazon Web Services and Microsoft Azure, both of which are spending astonishing amounts to expand capacity.
In advertising, it faces Meta, Amazon, TikTok, retail-media networks, connected-TV platforms, and every company that has gathered enough customer data to suddenly describe itself as an advertising ecosystem.
In search behavior, it faces AI assistants, social platforms, shopping sites, Reddit, and younger users who sometimes prefer video recommendations to ten blue links.
Alphabet’s advantage is distribution.
Google Search, Chrome, Android, YouTube, Gmail, Maps, Workspace, and the broader Google ecosystem give the company access to billions of users. It can place new AI capabilities inside products people already use rather than begging everyone to adopt an entirely new destination.
But distribution is not invincibility.
History is full of dominant companies that possessed distribution, resources, talent, and impressive presentations shortly before consumer behavior shifted beneath them.
That is why the quality of Alphabet’s AI products matters. Google cannot rely only on making Gemini available. It must make Gemini useful enough that people choose it.
Recent delays surrounding its flagship Gemini models have raised questions about whether the company can consistently remain at the technical frontier, particularly in coding and agentic systems. Management has acknowledged areas requiring improvement while emphasizing its commitment to future models.
I will listen for concrete progress rather than competitive reassurance.
Confidence is inexpensive.
Frontier model training is not.
My Bull Case
My optimistic view begins with the resilience of Search.
AI features appear to be increasing queries rather than destroying Google’s advertising business. Search revenue is still growing at a strong double-digit rate, and Google possesses the data, advertiser relationships, distribution, and commercial intent necessary to monetize evolving search formats.
YouTube provides another durable growth platform across connected television, Shorts, subscriptions, sports, music, and creator-driven commerce.
Google Cloud has moved from chronic losses to excellent profitability while accelerating its growth. Its $514 billion backlog provides substantial revenue visibility, and direct TPU system sales open another market. Enterprise demand appears strong enough that capacity, rather than customer interest, has become the immediate constraint.
Alphabet also possesses financial resources few competitors can match. Even after its enormous spending, the company ended the second quarter with approximately $242.5 billion in cash and marketable securities. It can invest through cycles that would force smaller companies to retreat.
If Alphabet can sustain advertising growth, expand Cloud revenue, normalize Cloud margins after temporary capacity costs, and monetize AI across its ecosystem, today’s spending could establish a powerful long-term advantage.
In that scenario, the company is not recklessly overspending.
It is buying infrastructure before demand becomes even larger.
My Bear Case
My cautious view begins with capital intensity.
Alphabet’s revised spending plans are staggering, free cash flow has come under pressure, and depreciation will continue rising. Management already expects another significant increase in capital expenditures in 2027.
The company is investing ahead of returns that remain difficult to measure.
Search monetization may become less efficient if AI-generated answers are more expensive to serve or reduce the number of commercial clicks. Regulators may continue challenging Google’s distribution arrangements and advertising practices. Cloud margins may retreat as third-party capacity, hardware revenue, energy costs, and competitive pricing alter the mix.
Cloud’s recent growth may also create expectations no large business can sustain. An 82% growth rate becomes dangerous when investors begin treating it as normal rather than exceptional.
Then there is the possibility that AI models become less differentiated. If customers can move workloads easily among providers or adopt capable open-source systems, the industry may produce enormous usage without equally enormous profits.
The bear case is not that Alphabet lacks good products or talented people.
It is that the industry may spend too much money building capabilities customers eventually expect to receive cheaply.
Technology has a long history of creating miracles and then charging almost nothing for them.
What Would Make the Next Report a Success for Me?
I would consider the upcoming earnings report encouraging if Search and YouTube maintain healthy growth after accounting for more difficult comparisons and currency effects.
I want Google Cloud to deliver strong underlying growth even if the reported rate naturally slows from 82%. I would accept modest near-term margin pressure if management demonstrates that it comes from temporary third-party capacity and rapid customer acquisition rather than deteriorating competitive economics.
I want backlog to remain strong and customer consumption to continue expanding.
I want greater clarity on the path from AI usage to AI revenue.
I want capital expenditures to remain connected to observable demand, not merely the fear that a competitor might build something first.
I also want operating expenses and depreciation to grow at rates that leave room for operating leverage over time.
What would concern me?
A sharp deceleration in Search with little explanation.
Cloud growth driven mainly by lower-margin hardware while core consumption weakens.
A significant Cloud margin decline without a credible recovery path.
Another major increase in capital-spending guidance before the current investments begin producing measurable returns.
Or management repeatedly answering questions about monetization with statistics about engagement.
Engagement is valuable.
The electric company still prefers money.
My Final View Before Alphabet Reports Again
I approach Alphabet’s next earnings report with cautious optimism.
The company’s core advertising business remains stronger than many expected at this stage of the AI transition. Google Cloud has become a genuine profit engine, and enterprise demand appears extraordinary. Alphabet’s distribution, infrastructure, technical talent, proprietary chips, and financial resources give it a formidable position.
But the price of maintaining that position is rising dramatically.
Alphabet is attempting to protect Search, challenge larger cloud providers, train frontier models, expand Gemini, sell AI to enterprises, build custom chips, construct data centers, power those facilities, develop autonomous vehicles, and persuade investors that all of this belongs inside one coherent economic machine.
That is an ambitious assignment, even for Google.
The next earnings report will not settle the AI debate. No single quarter can. It will, however, offer another piece of evidence about whether Alphabet’s spending is creating durable operating leverage or simply increasing the admission price to remain competitive.
I am less interested in whether the company “beats” expectations by a few cents than in whether the underlying relationship between growth and spending continues to make sense.
Advertising must remain resilient.
Cloud must remain profitable.
AI must begin demonstrating economic returns commensurate with its appetite.
Alphabet has already shown that it can attract users, advertisers, developers, and enterprise customers. Now it must show that it can build enough infrastructure to serve them without turning free cash flow into an endangered species.
That is the real earnings preview.
The ad business is financing the present. Cloud is trying to justify the future. AI is sending everyone the bill.
When Alphabet reports again, I will be watching to see whether those three parts still fit together—or whether the world’s greatest information company needs to provide shareholders with a clearer answer.
Comments
Post a Comment