Avory - Markets and Investing

AI vs SaaS: Seat + Consumption Can Coexist

Salesforce, Workday, Nvidia, Zoom, Block | Around the Desk

This week on Around the Desk, Sean Emory breaks down a pivotal week for AI and enterprise software.

Are seat-based models being replaced?
Or is AI expanding the value of platforms?

Using earnings and data from Salesforce, Workday, Nvidia, Zoom, and Block, Sean argues AI is enhancing durable platforms, not eliminating them. The winners are likely multi-product ecosystems with compliance depth, proprietary data, and embedded workflows. Not point solutions.

00:00 Welcome and Disclaimer
00:43 AI vs SaaS Big Week
02:13 Platforms vs Point Solutions
03:46 Salesforce Seats + Agents
07:06 Jobs Data
10:08 Buybacks + Workday
12:07 Inflation + Breadth
14:17 Nvidia + Valuations
17:26 AI Adoption + Limits
19:03 Capitulation Setup
22:08 Portfolio Updates
26:19 Closing

Disclaimer

This content is for informational and educational purposes only and does not constitute investment advice, an offer, or a solicitation to buy or sell any security. The views expressed are as of the recording date and may change. The host and affiliated entities may hold positions in the companies discussed. Investing involves risk, including potential loss of principal. Always conduct your own research and consult a licensed financial advisor before making investment decisions.

© 2026 Avory & Co. All rights reserved.

What is Avory - Markets and Investing?

Around the Desk: This is where we at Avory think out loud, challenge narratives, and look for signal through the noise. Each episode, the Avory & Co. team dives into what’s moving markets, how companies are performing, and where opportunities may be forming. We break down earnings, macro trends, and investor sentiment — all from the lens of a concentrated, high-conviction portfolio.

*** The views expressed on Avory Podcast: Around the Desk are those of the hosts and guests and do not constitute investment advice. This podcast is for informational purposes only and should not be relied upon to make investment decisions. All investments involve risk, including potential loss of capital. Avory & Co. may hold positions in the companies discussed.

Sean-1: All right.

Welcome to Avery around the Desk podcast.

I am Sean Emery, founder
and chief investment officer

here at Avery and Company.

You know, before we begin, a
quick reminder podcast is for

informational purposes only.

Nothing here discuss should be
considered investment advice.

You know, don't buy or sell anything.

You know, we're talking about.

Uh, and you should look to
do your own research, uh, and

make decisions on your own.

Investing does involve risk.

Now let's get into it.

You know, last week we spoke a little
bit about, you know, that this was

going to be an important week for.

You know, the AI narrative, the
software narrative, you know,

those two things clashing.

We got Nvidia, we got Salesforce,
we got Zoom, we got Block, we got,

you know, Workday, we got into it.

We got so many companies that I think are
so important to, uh, you know, extract

information and, you know, continue to
verify or disprove, you know, some of the

thesis that, you know, you or I, you know,
may have as it relates to, you know, AI

and its impact across the, um, you know,
the software landscape and just, you

know, in, in, in the world more broadly.

Now, you know, part of the view and, and,
and, you know, part of the, the questions

is, you know, is, is AI replacing
software, SaaS software as a service?

Or is, you know, is AI
really enhancing it?

And you know, whether consumption models,
uh, will kill off the C base models,

uh, or whether, you know, something,
uh, more nuanced will emerge and I

think the more nuanced will emerge.

Is, is simply how, you know, uh,
evolution tends to happen versus, you

know, linear thinking, which I think,
um, you know, most, uh, people in the

markets and or, you know, people, uh,
contemplating what's happening in ai.

I think they're extrapolating,
you know, ideas into the future.

You've had many of these various, um,
articles come out on X, whether it's, you

know, the ci trinia report or the report,
um, the article, you know, a couple weeks

back that kinda suggested doomsday, um.

And, you know, some are coming back with,
you know, hey, that's science fiction

and you know, we have our own view.

Look, we think, um, at the end of
the day, like, you know, there's

companies out there in the software
space or even in the non-software

space that if you're doing something.

Um, that is very singular.

Something that is, um, you know,
somewhat of a point solution.

You know, you may be at risk, right?

So you're really looking for durable
platform companies that have multiple

modules that have, you know, context
that can't be replicated, proprietary,

you know, information in there that.

You know, isn't gonna wanna be, you
know, exposed out to, you know, public

vendors out there and kind of stay
isolated in their own ecosystem.

Um, compliance regulation,
all that matters.

And I think, uh, you know, that's, that
kind of gets lost in the, um, the weeds

here as, uh, I understand everyone's
seeing these tools extrapolating

what it could or, or cannot do.

And I think it, you know, it reminds
me a little bit of COVID, it reminds

me a little bit of, um, the NFT craze.

It reminds me a lot of, you know,
different points in time where.

You extrapolate something into the
future, uh, such as, you know, we're

never gonna go to the grocery store.

You know, people are never gonna
live in big cities and just live

remote for the rest of their lives.

Um, and many of those same
concepts I think are coming

to light as it relates to ai.

Uh, again, not to be little AI
'cause it's, you know, very powerful.

We use it internally here a lot.

Um, you know, we continue to grow
our usage, but we continue to

grow our usage across the board
in many of the things we're doing.

So again, I think it's a
more nuanced conversation.

Uh, it's more of an enhancement than a
replacement in, in many ways, and I'll

show some of those stats today across,
you know, some of the companies as well.

So let's kick it off like, um, you know,
Benioff, Benioff, uh, spoke after his

earnings, you know, CEO, founder of
Salesforce, and this is what he said.

Look, uh, he doesn't see consumption
base business models, overtaken seat

based models, consumption being,
you know, you have an agent running

and it goes and completes a task.

They're paying for that usage versus
a seat based model where again,

you're paying, you know, 9 99 bucks
a month and, you know, whatever

you do, um, it's 99 bucks a month.

And, you know, in, in one way, the CFO
likes that because you know you can

manage your, your cost structure better
when you have a fixed base pricing.

Uh, on the consumption side, you
know, it's, it's more value add, you

know, you're, you're paying for value.

Uh, so I think companies like that from
a return on investment standpoint, but

they don't like that on a consumption
standpoint, uh, because you know,

it can impact your, you know, your
line items pretty drastically.

So having guardrails in place for
consumption, I think will matter.

But you know, back to Benioff, I dunno
which Anthropic product you're using.

This is his quote.

Uh, but the one I'm using is seat based.

I dunno which open AI product you're
using, but mine is seat based.

Same thing for us.

Uh, this is about humans and agents.

Humans are seats, and
then we have agents too.

They're using APIs.

And APIs are just ways for two systems
essentially to communicate to each other.

Uh, and that's how
agents will communicate.

Uh, it's still humans and
agents working together.

This is, uh, continuing the quote
That is what is exciting about

the future of enterprise software.

Uh, that's how I see it unfolding.

It's about a world where there
are apps and agents where

LLMs extend our capability.

They make us better, stronger, and
give us the ability to do more.

Seed still exists, consumption exists.

So look, I tend to agree with that.

You know, he's clearly, uh, runs
and operates a software business,

so you could say he's more biased.

But look, there are agentic force
or agent force, which is their

offering, uh, as it relates to agents.

Is growing 800 or $800 million, uh,
revenue run rate business growing,

you know, 200% year over year.

So they're seeing real traction there.

Obviously that's a small part of
their large business, but then they

articulated that they're seeing their
seats, seat count actually increase.

Uh, so again, they're seeing
their seat count increase

while their age agent platform.

Uh, is growing pretty rapidly.

So there's more to talk about there.

The next page, um, you know,
is really Benioff articulating.

Look, Salesforce didn't hire any
engineers or service agents here in the

last year, um, but they increased the
number of sales hires by 20%, uh, because

the company has more demand than ever.

Um, so again, I think with ai, what you're
seeing is where there's a place for.

Um, you know, whe where you're seeing
more productivity, you're gonna lean

into, uh, where you're seeing less
productivity or you were already bloated

because you over invested, you know,
in that COVID timeframe, which I think

should not be lost in this discussion
around, you know, AI and jobs, you know,

a lot of jobs were added during COVI,
uh, again, with the idea that everything

was going to be digital and therefore
you see a lot of digital companies

retrenching a little bit in terms of jobs.

Um, but I think it's more of a
function of normalization and.

And, uh, efficiency, something
we saw meta and some others do,

you know, back in 20 22, 20 23.

Um, and you're seeing that a little
bit with some others, you know,

that have, you know, stayed bloated.

So again, trying to highlight
how there's, there's a little

bit more nuance to that as well.

There's, you know, continuing on,
you know, I think, um, you know,

one important data point here
is really around job openings.

We highlighted just software
developers last week.

But when we look at the top 10 list of
Indeed's, job postings of job openings,

what companies are looking for, I found
it pretty shocking when you looked at

the number one growing 12% essentially
year over year of software development.

So that's the fastest growing
job openings category.

It was perceived to be that, you
know, AI was going to eat software

development because that was,
that's essentially one of the best.

Use cases today.

Um, you're also seeing customer
service, another good use case for

ai, where you can have chatbots, you
know, essentially handle, you know,

many of your, um, customer service.

Uh, but you're seeing seven
point half percent growth there.

Uh, sales, uh, human resources,
uh, therapy, all these things

that, you know, a lot have,
have, um, claimed to be AI risk.

We're in year three, three and a half.

Obviously, you know, every year,
uh, AI is progressing, but.

You know, this is as of this past month.

Uh, so again, it, it's somewhat of
a, uh, narrative breaker, uh, as it

relates to jobs and, and AI roles.

Here's a, a good, interesting
thing, again, you know, a little

bit of back and forth here, but this
is, uh, Jason Lemkin from Saster.

You know, they run one of the bigger,
uh, software conferences out there and he

said, you know, they have a small team.

I think it was 12.

Uh, we cut our Salesforce seats in half.

Um, but we use so much
Salesforce data and agent force.

Net net we're spending 50 to 60%
more on Salesforce and it's worth it.

That's his quote.

Um, so again, you're seeing it, you're
seeing where there's a view that

software can be, um, you know, taken out.

First off, anthropic, uh, uh,
and OpenAI both use Slack.

They both use Salesforce.

Uh, so that is not necessarily, um.

That should tell you a lot of things.

Uh, but then you have something like
this where again, let's say you had a

sales rep and that sales rep has, uh,
you know, is set setting out leads, and

all of a sudden you can have an agent
do that, uh, and you're winning deals.

And that agent, you know, that historical
human, you are paying $200,000, um,

and you can pay $20,000 to Salesforce,
uh, instead of 9 99 bucks a month.

Um, you can pay $20,000 for
that, you know, various agents.

This is what Jason's talking
about, where consumption, uh, can

totally, uh, over consume, you know,
essentially the previous model that

someone like Salesforce was running.

So you wanna own these platforms
that have multiple modules that act

actually have real value, that have
personal information, proprietary

information that AI is enhancing.

Um, you know, Jason, he got brought
on the earnings call for Salesforce

and he says he thinks, you know, the
Salesforce model could be three x.

That's the, the, the level of
revenue because of the, the value

that can be extracted using agents.

Um, so there's a little bit of that.

Now, again, we talked about last
week's, um, software and, you

know, what can help assist there?

And we talked about buybacks and
obviously a lot of the big software

companies haven't hadn't reported yet.

So you got Salesforce and Workday, you
know, this week and Salesforce announced

a bazooka buyback of $50 billion.

Um, which again.

This is a, obviously a confidence,
um, from the team there, but also,

you know, it can act as a level of
support, uh, in that ecosystem from

a, you know, investment perspective.

Yeah, Workday.

So what did they talk about this week?

I think, you know, they're,
they're running fine.

You know, the, the business is doing well.

We don't invest in it or anything, but
it, you know, we're tracking it, uh, for

the sake of a lot of things, but mostly
for trying to understand, you know, how AI

potentially is impacting and just getting
any anecdotes that we can, you know, use.

And here's a couple quotes.

Um, by the way, uh, quote,
uh, everyone on these.

Of these AI leaders actually runs Workday,
just for what it's worth, anthropic

Google and OpenAI all run Workday.

Um, that's one.

And then later in the conversation
it was, you know, with over 11,500

global customers expansions, you
know, customers expanding with

them are largest growth engine.

In the quarter, we expanded
with customers like Anthropic.

So Anthropic this last quarter
expanded their usage with.

Uh, Workday a software platform.

So again, some of these narratives
of very linearly that, you know,

you can create your own software
that, you know, Workday, Salesforce,

zoom, uh, uh, whatever you name it.

All the software stacks are
just gonna be vibe coded.

Um, it's not really showing up with
AI Labs, expanding their usage,

AI labs, using these platforms.

Uh, it's somewhat of a narrative breaker.

And then you take the jobs angle
where software development is

the fastest growing openings, and
then customer support as well.

So again, something to think about
there, uh, from that standpoint.

Now, the macro side, you know,
just bringing up fl you know,

real-time inflation here, you
know, uh, it's running around 1.4%.

You can see how it's continued to fall.

Uh, and that is good
news, uh, for that front.

On top of that, we got super, we
got PPI today, so this is Friday.

Uh, you know, the 27th, uh, we got super
core or we got PPI today, the super core

PPI year over year, uh, you actually came
down and that aligns pretty closely to

Core PCE, which is what the Fed uses.

And you can just see here
if you're watching this, uh,

the chart between the two.

And they, they correlate really well.

Uh, and then the PPI report
leads the PCE report.

And that ticked down year over year.

Um, so we're not seeing, uh,
inflation re-acceleration.

Uh, we think that reduces pressure
on the Fed to, you know, do anything,

uh, that is hawkish, let's say.

Uh, and I think that's a good thing.

Um, on interesting.

We got, you know, somebody that.

I know Well on, you know, on the technical
analysis side, you know, we're seeing,

uh, NASDAQ buy signals expand here.

You know, just look at how you're seeing
macd buy signals expand, which is somewhat

counter to, again, some of the narrative.

So, again, underneath the surface,
I, I think you're starting to see.

Uh, some of the companies in, in the tech
sector, uh, start to expand in terms of

some of these, uh, you know, bias signals.

Um, we haven't seen this level of
bias signals in, in quite some time.

You saw it at the end of, uh,
November, and that was about it.

Uh, before that it was still,
you know, below these levels.

Um, so I think again, very important.

Now bringing up, uh, the stuff we've
been talking about, some of these, uh,

market signals really around software.

We got, uh, I believe
another signal this week.

And you're starting to see some
of these signals clustered.

The last time we saw clustering,
like this was around 2020.

Again, that doesn't guarantee that
like, you know, anything is a bottom

of sorts, but I, I think it does, uh,
acknowledge that we are in a washed up

setup, um, and that these things don't
happen in isolation and usually near,

you know, somewhat inflection points.

So you take the, you know, the prior
chart plus this chart, I think it

starts to get again interesting, like
we've been, you know, discussing.

Um, Nvidia, Nvidia had a strong report.

I think people are starting to
think about, you know, competition.

I think people are starting to think
about, you know, um, some of the su

supply chain, you know, constraints
and being able to deliver what you

wanna deliver to meet, you know, some
of, you know, buy-side expectations.

Buy-side being, you know, buy-side firms
that, you know, buy and sell securities.

Um, so look, their business is humming
along clearly, you know, their, uh, data

center business is growing 75% roughly.

Um, you know, there's no visible slowdown.

Um, the thing is, is, and I was on
Bloomberg, uh, here today actually,

and you know, one, one thing
about Nvidia is Jensen has thrown

out the number of $500 billion.

Um, there's no update to that number, so.

If you're very, very bullish on
ai, that's kind of your bogey,

so you're underwriting to that.

And therefore, if, if that happens, great.

Um, but are there any indicators
short term that signal that

we're gonna go beyond that?

Uh, and that's hard to justify, right?

So like, um, you know, when you
see a MD, you know, continuing

to expand its partnerships, uh,
continue to have products that

people are starting to resonate with.

You know, whether it's with meta,
you know, making a deal with them,

uh, Nutanix making a deal with them,
which is pretty cool to see given

that we have owned, uh, Nutanix.

And so, you know, I think there's,
there's some question marks there

as the durability and or just, uh,
air pocket and, you know, investors,

I think long-term investors,
you know, don't really care.

Um, but you, you, again, you have a
lot of trading activity that happens

at the markets these days that are
trying to time, you know, every cycle.

So I think that's a little bit of
the, the, the rationale of why, you

know, someone like Nvidia has been.

Um, not as popular here in the last
year, uh, from an investment perspective.

And, you know, some then, you know,
try to triangulate that to, you know,

bubble this, bubble that, and, you know,
here's a, here's a cool chart, um, by,

you know, you know, in, in general, uh,
Phil Rosen, um, that I, you know, saw

and, you know, looking at the valuation,
uh, discrepancies between, you know, the

2000 Cisco trade at 130 times earnings.

Oracle at 120.

Microsoft had 60 intel at 47.

Um.

And today it's Google at 26, Nvidia at 24,
apple at 30, uh, and Microsoft at 21, 22.

Um, so far different environments
and setups, uh, than that.

Uh, the correlation, uh, between
NASDAQ and software, uh, has is, is

basically the lowest it's ever been.

Um, historically, these things
move somewhat in tandem,

and you saw that break.

Obviously it's the AI narrative,
everything I just explained above.

Super important.

But ultimately I think, you know,
this is gonna resolve in two ways.

I think the correlations will
eventually come together.

You know, on one side you could
say, you know, NASDAQ catches down.

Uh, or you can say, you know,
that, um, you know, software, you

know, uh, moves higher or they
meet somewhere in the middle.

Like again, there's
multiple options there.

Um, ultimately, you know, everything we
talked about on the software side is our

view on the software side, specifically
public software, uh, names that are very

durable, very more platform centric.

You know, these are good companies.

Um.

The AI traction starting to show up
again, it's just trying to highlight some

of the different antidotes that we see.

You know, Adobe's AI spend, you know,
this is like email receipt data and stuff.

It's starting to show up in, um,
in, in some of the data out there.

So, Firefly data, ai, assistant
data, uh, is still picking up pace.

You see, you know, anthropic
here on this, on, on here,

continuing to take, uh, share.

We've talked explicitly about how
Zoom has a pretty large investment.

Uh, today, uh, in Anthropic.

So that's important from that standpoint.

Uh, but Anthropics continuing to win.

They're more than half of the
billings, uh, for LLMs today.

Um, and you're seeing that in other
forms of data as well, the ramp up at

an anthropic accuracy of these LLMs.

So again, I'm just trying to, you know,
I came across this stat where, you know,

there's a couple different, um, tests
that these AI models take and one of 'em.

Uh, is explicitly trying to, you
know, multimodal, multidimensional,

multi, uh, you know, uh, topic,
uh, of how accurate these models

can be when they take these tests.

Now, uh, to be quite frank, like if you
gave any human a test that was across all

different fields, uh, the average person
would probably have a very low score.

Uh, and so these thing, you know, these AI
models are probably smarter in that sense.

But are they smart enough to do
everything, which is essentially

the a GI concept, right?

And their scores are 50%.

So, um, they're still a long
ways away, uh, from, you know,

being, uh, accurate on everything.

Uh, so it's something to think
about as, you know, we continue

to move past in this stuff.

Um, you know, the next page
shows just, I mean, this, the,

the chart summarizes everything.

You see a massive spike.

And put volumes for software.

And this was a six day rolling,
this was from last week, right?

So you could just see the capitulation,
the level of anxiety, the amount

of angst to, uh, you know, get
out, uh, and or just by puts.

And again, um, these are things you
see closer to bottoms, not tops,

uh, revisiting the, the path to
calm that we, we posted, uh, you

know, maybe two or three weeks ago.

And we li we labeled out like seven
things that we thought were me, like

important potential paths to, to
calm the markets specifically on,

you know, the AI disruption fears.

Number one was CapEx growth moderates.

And like we haven't
explicitly heard any of that.

There's some anecdotes of it, but
ultimately I think, you know, if someone

like a Microsoft stock continues to get
hit because they're spending too much,

and that's the market's perception, the
mar the, the company's going to react

and they don't react by, you know.

By, you know, detracting their
spend, but more by slowing it

down so you get a deceleration.

Uh, and that would be fuel for, you know,
a market, you know, transformation there.

Uh, valuations reset.

Obviously we've seen that.

Uh, position washed out.

You know, I just showed the put volume.

I showed some of the other indicators
that highlight how positions

have been fairly washed out.

We know funds are essentially, you know,
neutral to negative, um, integration.

So this week with the software
names, we saw a lot of integration

where again, someone like Zoom
saw, uh, 10 of its top 10 largest

deals, uh, incorporate a paid ai.

C uh, you saw Salesforce's,
uh, agent force.

You know, accelerate, uh, to 200%
growth on $800 million revenue base.

So still small, but growing fast.

And so there's a lot of like good things
that we're seeing, uh, that show that

they're showing up in the earnings.

Uh, last one, you know, last
couple is like buybacks.

So again, we, we heard
buyback announcements from

pretty much all the big.

Uh, players out there.

Uh, and then last, there's two more,
but I think more importantly is

the security fears cybersecurity.

So Google wrote a, a piece on shadow
AI agents, um, just basically showing

how, you know, companies have employees
and these employees are now, uh, you

know, incorporating some of these
AI tools and you kind of have these

AI agents that are shadowed to the
organization extracting some of the

information from the organization
that I think that proves risk.

To the organizations and eventually,
uh, this will have to either slow

down or put guardrails around.

Uh, we also saw that the head of meta ai,
head of, uh, AI safety, um, she was using,

you know, one of these AI tools and, um.

Her emails were, you
know, essentially deleted.

Uh, so again, if you have someone in that
position and they're having troubles, uh,

from a AI perspective, I think that is
likely to continue to creep out there and,

and show up in in other organizations.

And therefore, again, be, um, uh, a
risk philanthropic made a post on,

you know, 24,000 fraudulent accounts
were generated, uh, using Claude.

Uh, so again, more insights there.

Last couple pages around.

Earning specific to, you know,
companies in our portfolio.

You know, someone like Clear, secure,
you know them for the airports,

but they're much broader than that.

You know, they're are becoming
much broader than that.

They're moving into, um, more identity.

So we think, uh, in the world of ai,
you know, one of the biggest issues is

gonna be around identity and having an
identity provider, uh, that you can trust.

Uh, is gonna be very, very important.

So they, you know, they, they signed
a deal with, uh, Mount Sinai and

other hospital systems, and Uber and
LinkedIn to verify, you know, for

verification purposes and bookings
went from, um, you know, teens to 25%.

So revenue was, you know,
16% in bookings are 25.

So it speaks to the idea of
a continued re-acceleration.

You know, they increase
their dividend by 20%.

You know, they have, you know, nearly
a billion dollars in cash, no debt.

So, very interesting and,
and good to see there.

Uh, we saw a Zoom report and,
you know, 10 of their 10 largest

deals, again, incorporated ai.

Seven of their 10 deals in their ZCX uh,
division were taken from a competitor.

So that should knock a little bit of.

You know, the view that, uh, competition's
gonna take anything from Zoom.

They've grown every year since COVID, uh,
and now they're displacing, you know, some

of the incumbents and some of the other
parts of, um, their business segments.

So, zoom phones, zoom Contact
Center, uh, AI companion, uh, is

starting to, you know, monetize.

And then they won deals from, you
know, a leading retailer, a major

insurance company, and, and major
League baseball, which obviously I like.

Um.

A hundred thousand plus customers
at Zoom are growing 9%, that's

now 33%, uh, of total revenue.

So again, they're creeping up there and
you saw, uh, bigger deals with their,

what's called, uh, remaining performance
obligations, growing double digits.

So again, bigger deals are, are,
are being won, and that's why you're

seeing those numbers take higher.

Lastly, block, block
reported really good report.

There's two aspects to the block.

Story number one is
growth is reaccelerating.

Here, if you're looking at
the chart, you can see it.

You know, growth, uh, gross profit
dollars, which is their version of

revenue, really, uh, growing 24% up
from 18%, up from 13%, up from 9%.

So four quarters in a row of, uh,
accelerating growth cash app going from,

you know, nine to 13 to 24 to 33% growth.

So again, very, very good.

Um, underlying, uh, you know,
metrics there, there was other

metrics that they shared.

Uh, but again, we've been very
constructive on, on, on them

for, you know, uh, a while here.

Uh, you know, especially post our,
our visit to, um, either their

investor day in San Francisco, the.

Other story is that, you know, efficiency.

So they, they built, uh, their, their, you
know, they acquired Afterpay, they, uh,

you know, four or five years ago now, and
you know, that that came with employees

and then, you know, cash App and Square
had, you know, two different operations.

And I think they're trying
to become more efficient.

So they had a, you know, a, a, a
pretty big, uh, uh, layoff, uh, that.

Will put them in a position
from a margin standpoint, uh,

continue to accelerate that.

So you have accelerating top line
growth, which signals that the

company is performing pretty well.

You know, them trying to flatten their
business so they can operate even faster.

Uh, they have, you know, they open
source a product called Goose,

which is kind of like their AI
agent that they use internally.

And, uh, clearly that's
showing some benefits there.

But again, I think the story there
is really around normalizing,

flattening the, the, the organization
from some of the over hiring that

they did, you know, in prior years.

And again, that's showing up in revenue,
numbers accelerating, and we think a lot

of that's gonna drop to the bottom line.

Uh, and they'll reinvest some of that,
uh, into the business going forward.

But 59 million now.

Cash app, monthly active users,
they added a million in the quarter.

Um, in terms of, uh, banking s
which are basically the crown

jewel for, uh, any digital wallet.

Uh, so now they have, you know,
I think over 9 million, you

know, primary banking actives.

So a lot of.

Uh, key, uh, indicators that suggest
that that business is healthy.

And obviously it's unfortunate
for the people that are let go.

Um, but from a business perspective,
you know, clearly, uh, they're trying

to flatten it very similar like I, I
think what Meta did years ago and, and

be closer to the product, closer to the
teams so they can operate at even higher

pace and, you know, take 24% growth and
accelerate that here in the coming years.

Um, and so all of that is good news.

The ne last thing is, you know,
we have obviously different

companies reporting next week.

There's, um, you know, there's, you
know, uh, Okta and, and, and Best Buy.

So we can see how many Mac Minis are
being purchased and, um, you know, on

the employment side or on the, on the,
um, economy side, you know, someone

like a Target and Costco also report.

Uh, we also get, you know, Viva,
which is in the software space.

Uh, box, which is, you
know, in the software space.

And, you know, Aaron Levy has been a,
you know, pretty good advocate of, of ai.

Um, so anyways, in, in closing, you
know, are we bottoming in software?

You know, we, we never know
what's gonna happen in a week, but

structurally I think, you know,
valuations, you know, are contin,

you know, continue to be attractive.

The companies are doing well, buybacks
are starting to become active.

You know, hiring is fairly stable.

We, we continue to see, if you looked
at the a DP, you know, weekly reports,

um, jobs are starting to accelerate.

Um, AI demand is showing up that's great
in these, in these software companies.

It's showing up in other places too.

Uh, and ultimately it looks like,
you know, signals all across the

border are starting to cluster here.

Um, and that we're not gonna have like
this structural decay of these, you

know, some of these software companies,
again, we're big believers in ai.

A lot of companies will get disrupted.

Uh, but as long as you focus on I
think the right things, um, both,

uh, at the company level, but
then the, the data level, I think.

Ultimately what you see is, uh, these
companies have the op, you know,

the opportunity to, to grow much
faster going forward than they were.

Um, that's it.

Uh, we'll be watching all the
data, so we'll see you next week.

It'll probably be a little
bit shorter next week.

Um, so keep that in mind.

Have a good one.