Google Cloud grew 82%. Why the stock still fell after hours.

I'm Shai Balor and this is Earnings

Edge. So, we're going to talk about

Google who just delivered one of the

strongest operating cores of the current

AI cycle. Yet, the stock is still down

after hours, which might sound confusing

at first because the actual business

results were exceptional. Uh, since

revenue grew 24% to nearly $120 billion.

Google Cloud is now growing at an 82%

clip. Search is still very healthy at

17%. Even YouTube advertising is growing

at 13%. And Gemini is clearly gaining a

ton of traction across not just the

consumers but large enterprises as well.

But the reason the stock is falling

isn't at all because the demand is

weakening clearly from the numbers that

just reported. Uh it's because quite the

opposite. demand is growing so quickly

that Google must spend even more now to

keep up with it. And management raises

capex guidance for 2026 and said it will

temporarily rent more expensive

infrastructure from outside providers

because essentially Google cannot build

its um own capacity fast enough. So now

the debate after this quarter is no

longer whether Google's AI strategy is

working because it clearly is but

whether the company can fund one of the

largest infrastructure expansions in

corporate history while still delivering

strong margin attractive returns for

shareholders. So let's get into more

specifics of the earnings report. Uh

most important number in the entire

report was obviously Google Cloud. This

is the one um business segments that's

actually going to move the needle for

the stock. Uh like I mentioned they grew

82% which is a crazy number at the scale

because this is now a$2 billion business

and it beat expectations by nearly two

and a half billion dollars. So it beat

expectations by at least 10%. So that

was impressive because the bar was

exceptionally high after Q1's beat. So

even the most optimistic forecast

heading to this print was still uh way

lower than what the actual business

produced. So more importantly it wasn't

just revenue growth. I mean, operating

income also tripled to $8.8 billion

while the operating margin expanded from

21% to now it's 36%. So now you have

this business that's growing 82% topline

with 36% operating margins. It's crazy.

It's huge. It's like this is a huge

moment for Google because AI

infrastructure is extremely expensive.

If you're like a Google, you're going to

have to continue to purchase not just

chips but also memory and networking

equipment. they have to continue

building these massive data centers,

secure electricity and cooling and

essentially install everything before

customers can begin using the capacity.

So for years um the biggest question

surrounding this AI buildout has been

whether hyperscalers can create

profitable businesses or simply spending

enormous amounts of money to defend

their existing positions. Uh and clearly

after this report Google is in the part

of the AI economy that's highly

profitable. I mean those margins with

that topline growth is exceptional. I

think that um Google Cloud actually

produced more operating income this

quarter than it generated in total

revenue less than three years ago. That

that tells you how much this business

has grown and not just that the margins

have just expanded at a rapid pace. Uh

and also cloud is becoming much more

important to the overall company since

it now represents more than 20% of the

total revenue and is growing several

times faster than search, YouTube and

the rest of the Google services. So, um

I think over time like this is going to

be a lot more of a predictable business

for uh Google uh regarding like the

multiple of what you should be buying

Google as a stock because historically

Google has been like an advertising

company. But this quarter really

strengthened the argument that this is a

business that has two major profit

engines digital advertising and AI

infrastructure. The next major number

from this print was the backlog. I mean

it crossed half a trillion dollars. Now

Google cloud's backlogs of $514 billion

which represents um revenue essentially

that customers have already committed to

under sign agreements but that Google

has not yet fully recognized. That's

what that backlog number means. So uh

essentially Google expects to convert

slightly more than half of that backlog

into revenue over the next two years.

Management has also said on the call

that existing customers are now

consuming more than 50% above their

original commitments while new customer

acquisitions has more than doubled from

last year. So that to me suggests that

uh Google is essentially trying to

fulfill demand that already exists and

is growing way faster than expected.

Obviously the risk um is that you can't

assume that backlog is going to

guarantee attractive returns because

Google still needs to deliver the

capacity on time. They have to manage

their energy and equipment costs and

avoid signing contracts that look

attractive on revenue but it's going to

generate a lot weaker profits and poor

margins. But I don't see that proof yet

because cloud operating margins at 36%.

So that shows that the infrastructure is

already running um right now is being

monetized extremely well. So I'm

anticipating the backlog will convert as

well to maybe not the same margins but

similar enough. Uh also the Gemini

adoption is moving well beyond the

experimental stage since Gemini app now

has reached around 950 million monthly

users. Uh Gemini models are now

processing I think it was around 22

billion API tokens every minute which is

insane because just last quarter it was

16 billion. So now that's a huge

sequential jump in tokens per minute

which is everything in this like

inference boom that we're about to

experience right we are experiencing

right now and also uh man said that

nearly 90% of the fortune 100 companies

are now using Gemini enterprise and more

than 9 million developers are building

with Google's models each month so these

are all really impressive figures and I

think it shows that there's real this is

Gemini is being used at real usage at

scale since now companies are beginning

to put the AI into real production which

requires a lot of continuous access to

computing power, security, storage,

databases, networking, software tools,

like takes a lot for uh a companies to

specifically not just capitalize on AI

but use AI on a day-to-day basis. And

that's exactly why Google's full stack

strategy is becomes so valuable because

Google isn't just selling Gemini as a

standalone chatbot like the open AI is

like you have to go uh use chat GBT like

Gemini is truly this connectivity layer

inside their ecosystem like it's

connected across Google cloud workspace

search advertising Android chrome cyber

security and data analytics so the same

investment in Gemini can therefore

create revenue across several different

products which is why Google really

doesn't need Gemini to win all these

model benchmarks that we're seeing all

over Finit. Like it truly just has to

have a good enough product because their

cloud platform as of right now is

supporting the Nvidia GPUs, Google TPUs,

Anthropics Cloud, open source models and

a wide range of development tools. So

even if a customer chooses another

company's model because as a better

benchmark score has a better specific uh

query that it handles better than what

Gemini does it doesn't matter because

Google will still make money by

providing the infrastructure where the

model is trained deployed secured and

operated. Um, this also talk about the

me I should say it's the legacy

business. I mean the search business

like it's still uh a big chunk of their

business but this was the segment that a

lot of people two years ago said that AI

was going to disrupt and this is going

to turn Google into a dinosaur. Uh but

if you were listening on to the podcast

future equities podcast you knew that we

believe the complete opposite and the

signs are still there that the search

business is much healthier than a lot of

the original AI bear cases suggested

because it's growing 17% to more than 63

billion despite the rapid growth that

we're seeing right now across cloud chat

GBT perplexity and other AI first

products. Uh, so I think that a lot of

people believe that these platforms

would quickly weaken Google's core

business by replacing the traditional

search queries, but that really hasn't

happened so far. And also not just the

defensibility that Google's had on the

search business. They're actually using

AI to force a tailwind on that search

business because now management also

talked about AI the AI mode has

surpassed 1 billion monthly users while

uh a lot of like the AI mode and AI

overviews now generating additional

search activity rather than just simply

replacing

existing searches. So I think I think

the company also mentioned something

about bringing down the cost of serving

AI generated answers even as it adds

more advanced capabilities which is huge

because the search debate really has two

parts to it. The first question is

obviously whether people will continue

using Google in an AI world and the

current revenue and engagement that

we've seen so far suggest that they

absolutely will. The second question

though was whether Google can make money

from AI generated answers without

destroying the economics of search. It

was a really tricky balance to benefit

from AI but without eroding your legacy

business. I do think that this quarter

offered encouraging signs that serving

costs are falling while advertising

tools and monetizations are improving

which to me is a really great thing to

see. Um I also think that it's really

important to note that Google's already

placing like clearly marked

advertisements inside AI mode and

they're also using Gemini to understand

longer more detailed searches that were

previously difficult to monetize. Uh

obviously advertising advertisers are

going to be using tools such as the AI

max and performance max. Uh and they're

probably going to they are already

actually seeing better conversions

results at similar returns on

advertising spending which is great

because that increasingly will resemble

Google's transition from being a desktop

to mobile rather than the collapse of

search. I also think that the interface

is changing. So uh I think that you're

seeing Google's ability to adapt the

advertising model around this new AI

world at a rapid pace and that's great

to see. Also I mean regarding that uh

YouTube that it was a solid quarter as

well. I mean it grew 13% to 11 billion

beat expectations. Uh I I do think it

was really nice to see that the

subscriptions continue to grow faster

than the advertising through YouTube

premium music and other paid products.

So uh YouTube is obviously the least

sexy of these business units that we're

talking about for Google. But it does

give Google that massive global platform

where Gemini can improve

recommendations, search, advertising,

content creation and commerce. So the

strength now like across search,

YouTube, and cloud is like really

important. Uh that they're all like

humming at the same time because Google

isn't really funding the AI transition

from a deteriorating core business.

They're actually funding it from a place

of offense like they're doing it because

they have the luxury to do so without

without having to need to do so.

Especially when the Google services

revenue is already growing is growing

15%. That uh the income as well is

growing 20% and that segment alone has

maintained like operating margin above

40%. Like that means that Google's

existing business continues producing

enormous profits even while the company

invests aggressively in the next

generation of AI infrastructure. Let's

talk about why the stock is down.

Listen, Google raised it capex number

and market doesn't like that. Um they

never like hypers scale raising capex

numbers. 2026 capex outlook went to $200

billion from $185 billion. It's a 15

bill bill billion jump. Uh so the

company is now spending more on capex in

a single quarter than it did all of 2022

which is crazy to see. This is the new

AI economy that like this is a buildout

that is providing exceptional returns at

a rapid scale that I think it's worth

it. I mean management also indicated

that spending will increase

significantly again in 2027 meaning that

the this investment cycle isn't close to

speaking. So the market is now therefore

asking whether even exceptional cloud

growth can justify a capital spending

program of this size. I think it does

but that's probably one of the reasons

of why stocks down. The second reason in

my opinion is Google plans to use more

third-party computing capacity during

this third quarter because the internal

supply cannot keep up with the demand.

Uh I think management talked about this

as like this bridge while Google builds

more of its own uh capacity. I do think

that trade-off is obviously rented

infrastructure costs more than the

infrastructure Google owns which will

create some kind of near-term pressure

on the cloud margins. But again like

we're talking 36% operating margins.

This cloud business is going 82%. Like

what are we what are we poking holes

here? I mean, Google, yes, Google is

effectively choosing to accept lower

margins for several quarters rather than

tell major customers to wait, but these

are exceptional numbers that they're

producing already on the Google uh cloud

business. Like, sacrifice a couple of

points on the bottom line. That's fine

in my opinion because strategically,

this makes sense to do so because losing

a large enterprise or like these

frontier AI labs during a period of

limited capacity is going to allow that

customer to build its long-term

infrastructure around a competing cloud

provider and you don't want to do that

for Google. So, I think the market's

just punching the stock a little because

it wants proof that this temporary

bridge doesn't become permanent, which

leads to the third reason I think the

stock's down. Free cash flow. Uh, as

crazy as it sounds, they had negative

free cash flow this quarter. Uh, I mean,

Google generated $39 billion of

operating cash flow during the quarter,

but they spent $45 billion of capex. So,

that obviously leads to a negative $6

billion. That's a huge change for a

company historically known for producing

enormous amounts of cash while

repurchasing stock. Uh also Google is

funding several expensive priorities at

the same time uh since the company has

begun training Gemini 4 which uh the CEO

has described as a much larger base

model needs to compete at the next

frontier of AI. Obviously these are this

is an expensive race that uh Google is

part of. AI is a prisoners dilemma with

the companies competing in have the

deepest pockets but listen upside is

there. All this money is going towards

capex is the same infrastructure that

can support several revenue streams

since a stronger Gemini model can

improve search advertising workspace

cloud like the list goes on. The

downside is obviously that the frontier

AI development is becoming a lot more

expensive than a lot of people

anticipate a year or two ago and there's

no guarantee that the company can

maintain a permanent tech week. Like

we're at the fluid stage of AI. Nobody

can predict what these companies will be

like in three years because the rate of

change has been exponential. But I do

think this quarter was much stronger

than the stocks after hours the client

suggests. And I think that uh Google

showed that the demand is real, cloud

growth is accelerating, Gemini is

reaching meaningful scale and search

remains healthy and AI infrastructure

can generate excellent operating

margins. So I do think that Google's AI

strategy is working operationally well,

but the market is just continuing to

debate the financing cost of scaling it,

but that's nothing new. I do think that

um the market will look past whatever

heartburn it has from that $15 billion

increase in capex, a temporary free cash

flow negative quarter because Google

controls one of the most complete AI

platforms in the market. Uh but this

next phase that they're entering is

going to they're going to have to show

that the cloud margins can recover after

the temporary use of outside Vacity that

the backlog can actually convert into

profitable revenue uh and that the free

cash flow is going to rebound once the

most aggressive stage of the

infrastructure buildout passes. But for

now, I think I view the stock's weakness

as a reaction to the size and timing of

the spending cycle rather than any signs

of deterioration in their underlying