HomeVideos

Highlights from David A. Steinberg and Dan Ives Fireside Chat | Milken 2026

Now Playing

Highlights from David A. Steinberg and Dan Ives Fireside Chat | Milken 2026

Transcript

885 segments

0:00

Great. Great to be here with you, as

0:02

always. Great to be here.

0:03

>> Yeah.

0:04

The uh we we decided we would interview

0:07

each other today, and I'm going to start

0:09

by asking you the first question. So,

0:11

Palantir just [snorts] reported while we

0:14

were in the back room getting ready, the

0:16

CFO of Palantir called you. What were

0:18

you guys uh I know without violating any

0:21

uh you know, privacy issues, what were

0:22

you guys talking about?

0:23

>> look, I mean, to me is I they they had a

0:25

robust quarter. I mean, if you look at

0:28

their growth, it's just phenomenal. I

0:29

think it's a good example of you know,

0:32

what's the validation for this AI

0:34

revolution, right? All the CapEx

0:36

spending. Palantir, I think, front and

0:39

center. And I think

0:40

>> they grew their US business 100%.

0:42

>> Exactly. Look, and I think I think what

0:44

Palantir is doing speaks to And

0:47

obviously, it's a big focus of

0:48

everything that you're doing when it

0:49

comes to Athena.

0:51

Look, I think this is just the beginning

0:54

of the AI revolution. The soft- like

0:56

when you think about the software, the

0:57

use cases, and I think what they're

1:00

seeing is a lot of the stuff that, you

1:02

know, you're seeing with Athena, right?

1:04

I mean, it's you know, I I guess I'll

1:06

I'll go to you, okay? Like, you know, we

1:09

talk about Palantir, and I know I think

1:10

also like a lot of the stuff that

1:14

that I see with Palantir, it it reminds

1:16

me, if I go back a few years ago, right?

1:18

Where like no one gave them the respect

1:20

in terms of on Palantir, and look what

1:22

Carpenter have done. When it comes to

1:24

like Athena, it really feels like Athena

1:27

is changing data. She is. Yeah, it's

1:29

funny, because if you look at our

1:32

quarter, we grew the business by 50% top

1:35

line year over year at scale. I mean,

1:37

that's, you know, then you look at

1:38

Palantir at 100, you know, you feel a

1:40

little like you're you're sleeping

1:42

there, but at 50% growth, we were I was

1:45

super super proud of what we did. I

1:48

think there was some confusion in the

1:49

quarter, and you know, listen, we've

1:51

projected $397

1:55

million in EBITDA

1:56

for this year, and you know, we very

1:57

much expect to be meaningfully net

2:00

income positive this year, which I think

2:02

is causing confusion.

2:02

>> It's good you hit It's good you hit on

2:03

there cuz right there is

2:04

>> Well, a lot of people are talking about

2:06

that. And and the truth of the matter is

2:08

that you're projecting 397 million in

2:11

EBITDA. We expect to be, you know, net

2:13

income positive. And the restructuring

2:16

from the Marigold acquisition, which by

2:17

the way, we are just

2:19

killing

2:20

with that acquisition. I mean, if you

2:22

look at the business, we grew 29%

2:25

organically in the quarter and 50% total

2:28

in the quarter. That's because Marigold

2:30

was so far ahead of where we expected it

2:32

to be. And you know, quite frankly, our

2:34

organic business is growing because of

2:36

Athena.

2:38

In the first quarter, we saw our Agented

2:41

volumes go up by 700%.

2:45

60% of which was Athena.

2:49

So, But but also goes I remember like

2:51

when we sat down and now is the at CES,

2:54

January and and now you think you're at

2:56

Milken I can't believe you're doing it.

2:57

You The outfits just get better though.

2:59

I know. No, you you're

3:01

you could wear the same outfit. I don't

3:02

think I could pull that off. But but but

3:04

but if I go back to CES in January and

3:07

here we are at Milken in May,

3:10

I mean, to some extent like the strategy

3:12

is really playing out exactly and even I

3:15

think almost ahead of plan. Yeah,

3:17

Athena's way ahead of where we expected

3:19

it to be. 100% of our enterprise clients

3:21

have Athena now in their hands and

3:24

everybody's using it. And and quite

3:26

frankly, if you look at Zeta's return on

3:28

investment, according to Forrester, for

3:31

every dollar that's spent on our

3:32

platform, we return 600%

3:35

return on investment.

3:37

So, $1 equals $6 in business. Our

3:40

clients using Athena are seeing a

3:42

substantially higher return on

3:44

investment than that. And and that

3:46

speaks like look, just like Palantir

3:48

>> way, we're supposed to be interviewing

3:49

each other. But but but to some extent

3:52

but you but but what I was saying is

3:53

like just like Palantir, what we've seen

3:56

there, it's really it's my view that

4:00

you're actually starting to see not just

4:02

validation, the actual use cases play

4:05

out. I mean, that's what that's really

4:07

what you guys are seeing in terms of

4:09

when it comes to Athena.

4:10

>> Yeah. The use cases have been

4:12

really surprising how many they are. And

4:15

And when you look at a 700% step up in a

4:17

Gentex utilization in the quarter, and

4:20

60% of that being utilized by the

4:23

platform that you put live in that

4:25

quarter,

4:26

right? And And you know, we're going to

4:28

be rolling out V2, V3, V4 of Athena, you

4:31

know,

4:32

you know, almost, you know, every month

4:34

or or every other month over the over

4:36

the next year. Let Let's talk for a

4:39

minute about SaaSageddon. Yep. Right?

4:42

So, you you're getting asked this

4:43

question. You're on CNBC, you know, if

4:46

not every day, almost every day. What do

4:48

you think? How does a company

4:51

like Palantir or us or others, how do we

4:54

react to something that's so

4:55

nonsensical?

4:57

Look, I think part of the pri- And I'll

4:59

almost

5:00

I'll almost call it it's like a we call

5:02

it the AI ghost train in terms of

5:04

Anthropic, basically that they're going

5:06

to, you know, structurally break every

5:08

piece of software and they're going to

5:09

ruin I think you have to separate out

5:12

are there going to be areas of software

5:15

disintermediated relative to Anthropical

5:18

but of course and like For sure. And And

5:20

And I think those companies could

5:22

clearly be exposed and those are the

5:25

ones either cause M&A or structure those

5:27

companies are on the loser category.

5:30

But I think for software companies, and

5:32

it's not just you, it's not just

5:34

Palantir, it's a Salesforce, it's a

5:36

ServiceNow, it's everyone. It's trying

5:39

It's trying to figure out, okay, when

5:41

does the monetization phase start? See,

5:43

my whole point is if you think about

5:46

data centers taking two or three years

5:47

to build and GPUs and the infrastructure

5:50

and the

5:51

we're only now just starting to get into

5:54

the software phase.

5:56

Yeah, cuz

5:57

all of these tech cycles start with

5:59

infrastructure. Exactly.

6:01

>> Right? And then the application layer

6:03

that sits on top of it comes next, aka

6:05

software. And that's why I think like

6:07

what we see with Palantir,

6:09

what you guys are seeing with Athena,

6:12

this is actually just the start

6:13

Snowflake seeing and others. So so my

6:16

view is that they painted software all

6:19

with the same brush and you see it

6:20

structurally if any sort of thing

6:22

happens it takes the stocks down

6:24

significantly. The reality is is that

6:27

the software use case phases are just

6:30

starting. I'm sure you see that when you

6:31

talk to customers, like when you talk to

6:34

customers today, how much different is

6:35

it today than 6 months ago? Oh, it's I

6:39

mean and I remember we started working

6:41

in AI in 2017.

6:43

So for years and years nobody even

6:46

cared. Like they would be like, "What

6:48

what What are you talking about?"

6:50

Now we're getting pulled into use cases.

6:54

We're you know, I think Athena is going

6:56

to solve far more problems for our

6:58

clients than people expect that she

7:00

will. Totally. And then look it just I

7:02

think so to your question I would just

7:05

say I think investors to me is covering

7:09

tech stocks in late '90s, it's the most

7:12

I think

7:14

incorrect narrative

7:16

that I've actually ever seen in my

7:18

career relative to software what's

7:20

happening. It will take time to prove

7:22

itself out, but the reality is AI native

7:25

stocks are never going to unseat core

7:28

install bases. If you look at the data,

7:30

if you look at ultimate lab use cases

7:32

where they're going to play on, LLMs

7:35

will partner with software companies and

7:38

over time LLMs will get commoditized and

7:40

you'll have you're going to have

7:42

hundreds or thousands around the world

7:45

relative to you know to models as they

7:47

ultimately deploy. Software will be the

7:49

hearts and lungs of AI.

7:50

>> Well, and and and data, right? I mean,

7:52

if if you look at it, proprietary data,

7:54

I think it's going to be the biggest

7:55

differentiator as you think about where

7:57

these models are going, right? Because

8:01

models are only as smart as the data you

8:03

feed into them. Exactly. And if you look

8:05

at and I'll talk about data again. We

8:07

have if you think of the mode around our

8:09

business, we have 555 million people who

8:12

have opted into our data cloud that we

8:14

have never fed that data into a large

8:17

language model ever, nor would we. It

8:19

simply trains our models, and then we

8:21

take our clients' first-party data,

8:23

which

8:24

includes 51% of the Fortune 100, we're

8:28

ingesting all of their data into a

8:30

consumer data platform, we're matching

8:32

the data together, and it's training the

8:34

models just for that client. One of the

8:37

really interesting things that I think

8:39

people missed about our company was last

8:41

year, our net retention rate as a

8:43

company was 120%.

8:47

And most software companies are happy at

8:48

100, we were at 120. That's because once

8:52

our clients ingest we ingest our

8:54

clients' data and we map it with our

8:56

clients' data, it's almost impossible to

8:57

fire us because you keep getting better

9:00

better return on investment for them,

9:03

therefore they keep growing with you.

9:05

But that really goes to the broader

9:06

theme here. And I think well, you saw it

9:08

in your in obviously in this quarter in

9:10

terms of those numbers, but I think more

9:12

and more we are now in the software

9:15

phase. We are and I think that's what

9:17

companies are starting to see like, how

9:19

do I deploy it? I've spent tens of

9:22

millions or more on infrastructure.

9:25

Where are the use cases

9:26

Where are the use cases Use cases aren't

9:29

not coming from Nvidia, it's not coming

9:30

from a lot of the infrastructure, it's

9:32

coming from software.

9:33

When do you think the street gets that?

9:35

Or how do we convince the street of

9:37

that? Like I I just That to me is

9:40

confusing.

9:40

>> But but I'd say David, I think like the

9:42

issue that like from a communication,

9:45

the street, and the SAS apocalypse, or

9:48

the AI ghost rate, it's not

9:50

These are conversations that I see with

9:52

so many of my some of the best software

9:54

companies in the world.

9:56

I think it's going to start Look, it

9:57

starts with the results. Just like you

10:00

guys are showing in terms of the actual

10:01

monetization.

10:03

And then I think it really starts to As

10:06

you start to see these stacks and the

10:09

infrastructure get built out, I think

10:11

investors are going to realize it's the

10:13

most disconnected narrative

10:15

in the last 25 years relative to the end

10:17

profits and others

10:19

the view that they're going to un-teach

10:20

software companies. And I think it's

10:22

really Our view is the second half of

10:24

'26

10:26

is where it really starts to play.

10:28

Because that's where the use cases start

10:31

to play. I mean, we've seen it, you

10:32

know, you showed it, you saw it on

10:34

Palantir, but it's really as we get into

10:36

the second half of the year, that's

10:38

where that starts to run. And that's our

10:40

view like software, and we think

10:42

software stocks are you know,

10:44

potentially they're up double what tech

10:46

is in the second half of the year.

10:48

Interesting. You know, it's funny, at

10:49

Zeta, this was our fifth consecutive

10:51

quarter of accelerating organic growth.

10:55

And it's it's all coming from Athena

10:57

now.

10:58

And it you know, it's one of the things

11:00

I I find interesting cuz we are a native

11:02

AI company. We We re-architected our

11:04

entire platform to put AI as native and

11:07

data as native, but foundational to the

11:10

application layer. It's been so hard to

11:12

explain the AI. When we go into a client

11:16

now and talk about Athena,

11:18

it's almost like we've been driving this

11:19

Ferrari. The AI's in the engine, but you

11:22

can't see it, and now Athena is the big

11:25

head ornament, hood ornament, sitting

11:27

right on top. We're going in with a full

11:30

Athena pitch to all of our

11:32

That's really what we're doing.

11:33

>> it almost Athena's really it's changed

11:36

the company in terms of like for sure

11:38

the go-to-market strategy

11:40

and at a time where customers across

11:43

verticals are looking at how they're

11:46

going to monetize AI. What what else are

11:48

you seeing in the markets outside of

11:50

this stuff right now? I mean I think

11:51

like right now and I just got back from

11:53

Asia for a few weeks. So like I could

11:55

tell you

11:55

>> I don't know how you do it. You like I

11:57

called you last week just to say hi and

11:59

you're like I'm going to go to Korea.

12:02

I think part of it is like the only way

12:04

you could understand what the demand

12:06

looks like you have to see it globally.

12:09

So I think it's very easy while the

12:10

bears they're negative on AI in the 35th

12:13

floor in New York City office building.

12:15

So you don't see AI when you're in Metro

12:18

North. And I think the reality is that

12:20

the demand that we see it's 12:1 demand

12:24

and supply. Wow. And obviously you're

12:25

going to get to be at Milken he'll be

12:27

talking about the massive AI

12:29

>> in about 20 minutes. But but the reality

12:30

is like the chip demand is unlike

12:33

anything we've seen. It's a super cycle

12:35

in terms of memory and overall chips.

12:38

But guess what? In terms of that capex

12:40

build out it's all getting built out for

12:43

what the use cases are going to be. And

12:45

I think more and more you're seeing I'm

12:47

seeing from CIOs in so many companies

12:50

they're now going into that next phase.

12:53

And I think there's an arm race, right?

12:55

Like which are the stack players that

12:57

are going to benefit? Like who are some

13:00

of the niche software companies where

13:01

some of them might get acquired? Some of

13:03

them might fall apart. And I think it's

13:05

trying to it's it's those that stay on

13:07

the treadmill will get passed by.

13:10

Well, that's I mean if you look at guys

13:12

like Palantir, right? Didn't they go

13:14

through something like this a few years

13:15

ago? So I look Palantir is a good

13:17

example like and obviously you know

13:19

being such a supporter of them over the

13:21

years

13:22

you know so many times

13:23

>> way, watching you talk to like the guys

13:25

who run that company while we're in the

13:27

other room. The These are your boys,

13:28

right? Like you get along really well

13:30

with them. Well, part of it is like I'm

13:31

a huge believer in what Karp and

13:34

Palantir have built. And but but it's

13:37

very easy to look at them today

13:40

just like it's very easy to look at

13:41

Nvidia today. You don't see a few years

13:43

ago. You don't see where like they were

13:46

building it all on government. They were

13:48

building the infrastructure. No one

13:50

believed in No one gave them a multiple,

13:52

and look where it is today. No, so I'm

13:54

saying like that's why like like I said

13:56

like it's a lot of the stuff that you're

13:59

going through that I think other

14:01

software companies that are that are

14:03

thriving are going to It's very similar

14:06

to what Palantir a few years ago where

14:08

it took time for them to execute, show

14:11

the vision, and then investors

14:13

ultimately rewarded them. And that's why

14:15

I just think right now we are going

14:17

through this massive prove me phase

14:20

for the rest of software. Because I

14:21

think right now it's like, okay, there's

14:23

Palantir, and they're maybe put up here.

14:25

And everyone's trying to figure out who

14:28

else

14:29

is going to join them. Is it Snowflake?

14:32

Is it MongoDB? You know, is it Zeta So

14:35

everyone's trying to figure out who

14:36

those software Where does Oracle fit?

14:39

Where does ServiceNow? Where does

14:40

Salesforce? And I think that's why right

14:42

now we're in that phase. Interesting.

14:44

So, we're sort of in the We've got

14:46

infrastructure and software in the same

14:49

sort of concentric circle growing

14:51

together right now.

14:52

>> Well, cuz like right now like every

14:54

investor, they're focused on like

14:56

hardware, chips. I I don't care if it's

14:59

easy or year up here. You own memory

15:01

stocks. You own chip. Software is still

15:04

do not enter zone. I think that's the

15:05

wrong trade, but it's really because

15:07

like you're not getting paid to own

15:10

these stocks right now because the more

15:12

And I think that creates the massive

15:13

opportunities.

15:14

>> Right. If you look at the reactions like

15:16

ServiceNow and others,

15:17

but if you look at how it plays out,

15:20

okay, as companies go from like the

15:23

prove it phase or maybe the skeptic

15:25

phase from an investor to here, that's

15:28

the huge multiple re-rating. And then

15:30

And that's what's going to separate the

15:32

companies where we'll be here a year

15:34

from now. And there will be stocks that

15:37

you look back on and be like, what? How

15:39

did no one think because it was the

15:42

prove it phase. And I think that's we're

15:43

going through that now, not in chips.

15:47

In software. Right. Cuz chips are

15:48

already seen that. We've already seen

15:50

that. And right now, but go back I could

15:53

go back in terms of like these periods

15:55

of AI revolution. Go back a year ago.

15:58

Or go back like 2 years ago. There was a

16:01

video It can only be a video. AMD can't

16:03

be successful.

16:04

Memory stocks

16:05

>> Intel can't win. Intel can't win. Micron

16:08

can't benefit. SK Hynix, no, it's the

16:10

memory stocks. They're What about

16:12

SanDisk? They're not No one's going to

16:14

pay a multiple like that. So, the point

16:15

is like now it's consensus and that's

16:18

that's where the crowd is. Software,

16:21

it's still No, it's the I I believe as

16:25

an investor that so many of them are

16:27

structurally troubled. But I disagree

16:30

with that because ultimately the hearts

16:32

and lungs of these use cases will be

16:34

software. We're going through from May

16:38

to let's say like December is the prove

16:40

it phase for software companies. Well,

16:42

it's interesting cuz if you look at most

16:44

of the infrastructure and and I mean,

16:45

listen, they've built some insane

16:47

companies. Like these are great great

16:49

great companies. But they're effectively

16:51

selling the infrastructure once. Yes.

16:53

When we sell a client, it's a

16:55

subscription for a very long time. And

16:58

if you look at those use cases,

17:00

you know, you you would think software

17:02

companies that win it I want to go back

17:04

to your original premise, right? Cuz I

17:05

do agree

17:07

they're going to be what I call workflow

17:09

management companies that have built

17:12

platforms that

17:15

if they don't pivot, they're going to

17:16

have some real problems. Sure. Because I

17:17

think agents are going to

17:19

disintermediate them. As well as

17:21

seat-based models.

17:22

>> Yeah, of course. Well, the seat-based

17:23

model people were beating the living

17:25

crap out of us for charging a

17:27

subscription fee and a utilization fee

17:30

year 5 years ago when we went public.

17:33

And somebody was asking me the other

17:34

day, well, it seems like you've done

17:36

really well on the transition

17:38

to subscription. I'm like, we didn't

17:40

transition. Like we're doing we've been

17:43

we're we're 60%

17:45

>> your premise before you

17:46

>> That was the premise.

17:47

>> It was a hurdle before everything. That

17:48

was the whole thing. And they beat the

17:50

living out of us at the IPO for that.

17:53

Like nobody wants a utilization model.

17:55

You got to have a seat-based model. You

17:56

got to have a CPU model. And we just

17:59

didn't think that model scaled.

18:02

But it looks like now everybody's trying

18:04

to move to sort of your subscription fee

18:07

plus your utilization fee. Because of

18:09

the worries, too. And again, this also

18:11

goes back to a lot of AI alarmist, job

18:14

growth job's going to everyone

18:16

everyone's going to lose their job. And

18:18

and I think a lot of that is more of a

18:20

dystopian type narrative. I I think a

18:22

lot of that is going to be I'm not

18:23

saying there's going to not going to be

18:24

areas around that, but the view and I

18:27

think you're starting to see changes now

18:29

even in the job market and some of the

18:31

ones, you know, that that we've seen.

18:33

I'm just saying I think that's also

18:35

playing into this narrative about seats

18:37

going down, job growth, pricing power's

18:40

going

18:40

>> Well, if you don't need the employees,

18:41

why do you need seats?

18:42

>> No, but that's

18:43

>> Right. But but but then to you like the

18:46

question I would say for a lot of

18:47

software companies like, are you losing

18:49

pricing power?

18:51

Are you? And like what would you say to

18:53

that? Like you know, there was like cuz

18:54

you know, there

18:54

>> Well, we we we grew 50% last quarter.

18:57

And if you, you know, looked at the

18:59

business, it we're adding operating

19:01

margin this year and getting more

19:03

profitable this year. The the the

19:05

interesting thing when you think about

19:07

sort of how this all plays out, right?

19:10

This morning I did a I was I was invited

19:14

to a panel at the Milken Global

19:15

Conference with some of the largest CEOs

19:18

in the world.

19:19

And I then ended up having Coke or diet

19:23

Coke with the Wait, cuz you only drink

19:25

diet I only drink diet Coke. I only

19:27

drink I love I have I have I love diet

19:29

Coke. My wife doesn't love me drinking

19:30

it. I probably shouldn't be announcing

19:32

this on a live stream, but I had a diet

19:35

Coke with a senator a US senator today.

19:37

And what I can tell you is

19:39

whether it's a global CEO of which there

19:42

were I'd say 15 to 20 Fortune 500 CEOs

19:46

in in the room.

19:47

Uh or the senator,

19:49

the job repercussions of AI are the

19:53

third rail discussion. Mhm? Nobody wants

19:56

to have it.

19:57

But unless we start having it, we're not

20:00

going to figure it out, right? And And

20:03

it's something like

20:04

when you look at sort of AI,

20:07

I look at Athena.

20:09

Last quarter, not only did we increase

20:12

our Gentik uh utilization by 700%,

20:16

but clients who adopted Athena used 78%

20:21

less labor to manage the platform. Now,

20:24

what I say lovingly is and they can use

20:27

those employees to be repurposed into

20:29

other functions in their business.

20:30

Productivity, yeah. Right? And And I

20:32

think that's that's going to be a big

20:33

part of it, but I think there's another

20:35

part of it that we're going to have to

20:36

think about where where do some of these

20:39

jobs go? But then I would say and I

20:42

would tell you like my whole career

20:44

coming from Asia to here and so many

20:47

times I'd be in Taiwan and you see you

20:50

know, you see a fab 18 hours a day. You

20:53

know, the amount of production and I

20:55

land in New York Airport there's a fist

20:56

fight at Dunkin' Donuts. And then you're

20:58

like there's a reason

20:59

>> Why are you in Dunkin' Donuts? At And

21:01

New York Airport.

21:01

>> Okay. And then you're like there's a

21:03

reason we're 17th in math.

21:05

At that point for so much of my for so

21:08

much of my career, now for the first

21:11

time

21:12

US is ahead of China when it comes to

21:14

tech. So, I can make the other argument,

21:16

the amount of data centers going to get

21:18

built, the ripple effect, engineer,

21:20

private companies, innovation. So, so so

21:23

that's why like I do think the view of

21:26

AI you have to address it, yeah, but I

21:29

do think going back to like even weaving

21:31

into like software and private credit

21:34

and seeds and pricing model, I do think

21:37

we're going to look back and be like the

21:39

bark was a lot worse than the bite 100%

21:42

100% 100%. And by the way, we're at, you

21:45

know, 4 4 and 1/2% unemployment for the

21:48

third or fourth year in a row.

21:50

Three years ago, and this is I think

21:53

I'll show how incredibly old I am. I

21:55

think this is my 25th or 26th Milken

21:58

Global Conference.

21:59

Three years ago, the entire conference

22:02

all

22:02

>> when you started to go to Milken when

22:05

you were 14 years old? I loved you.

22:07

That's a me That's That's why I love

22:09

you. So, 14 years ago was your first

22:12

Milken? That's a me Well, in in all

22:14

seriousness, three years ago, all

22:17

anybody could talk about was aging

22:19

global population, and we're never going

22:21

to have enough people for all the jobs

22:23

we're creating. Now, all we're talking

22:26

about is AI is going to disintermediate.

22:28

As you said, pendulums swing hard, but

22:31

they're almost always wrong. They

22:33

overcompensate,

22:35

and I think we are going to find great

22:38

great jobs for people. But but when you

22:40

look at the difference between the

22:42

winners and the losers, I think

22:44

productivity accelerants

22:47

are going to be major. Tomorrow, I'm

22:49

moderating the lunch in the main

22:52

ballroom, which is Yeah, it's huge. It

22:55

was It was nice of Mike to ask me to do

22:57

that. They obviously they probably asked

22:59

10 people and you know, nine said no,

23:01

and then they ended up on me. But

23:05

it's all about

23:07

is this going to be the step function

23:10

in the productivity curve, right? Cuz we

23:12

we've been stalled for a few years. We

23:14

had this massive productivity that

23:16

started with the computer.

23:18

Then we had this massive productivity

23:20

that started with the internet. And for

23:22

the last 5 to 10 years, we've been very

23:24

stalled in productivity. The question is

23:27

does is AI the next step function? Or is

23:30

it just going to keep us sort of growing

23:32

at 2 or 3%? And and I don't think

23:34

anybody's going to know the answer to

23:35

that for a few years.

23:37

>> But the use cases will be the key. And

23:40

now we enter the software phase, where

23:43

obviously that's a huge part of you

23:44

know, where Athena and Zeta Plus. Well,

23:46

the one good thing about our stock right

23:48

now is it's buy one get one free. And

23:50

you know, we bought back I think 25 or

23:52

30 million dollars worth of our stock

23:54

last quarter. We're going to continue to

23:56

aggressively buy back our stock. And and

23:58

you know, as long as it continues to be

24:00

uh, you know, buy one get one free,

24:02

we'll keep doing that.

24:03

>> also part of it, too, is that you can

24:04

just execute on the plan, show the

24:07

numbers, build it out, and everything

24:10

else, you know, takes care of itself.

24:12

Damn, what else do you want to talk

24:13

about tonight? I mean, I would just say

24:14

like who

24:16

when we talk about like SaaS apocalypse

24:19

and just everything that we're seeing

24:21

from cu- Can you just like maybe like

24:23

without obvious

24:23

>> I asked you the question. Now you're

24:25

asking me a question. Bu- bu- bu- bu-

24:27

without without without naming customers

24:31

can you just explain

24:33

how the conversations specifically have

24:36

changed? Like if you just go back to

24:39

>> So, back to what we were talking about.

24:41

Athena has changed the narrative with

24:43

our com- customers in ways that I didn't

24:46

even think possible. So, I think a lot

24:48

of our customers just looked at us as a

24:50

marketing platform. And we did a good

24:52

job, and we had a good return on

24:53

investment. And when Athena was first

24:56

sort of introduced at Zeta Live and then

24:59

really we got behind it at CES and then

25:01

she was born you know, in the first

25:04

quarter we did that big birth

25:05

announcement

25:06

>> birth announcement

25:07

Uh, you know, when she was born clients

25:10

started looking at us more like an

25:12

innovation engine

25:14

and less like oh, they can just help us

25:16

with return on investment for marketing.

25:18

Yeah. And it's opened up entirely new

25:21

doors and new use cases. One of the

25:23

things that I thought was the coolest

25:25

statistic we reported in the first

25:28

quarter was that clients who use

25:31

multiple use cases on our platform,

25:33

which is the single biggest driver of

25:35

revenue unlocked, went up 50%

25:39

year over year.

25:40

Clients who use us for, you know,

25:42

customer acquisition, retention,

25:43

monetization went from doing one

25:45

to two or more went up 50% and that is a

25:48

flywheel that's going to continue for

25:50

many, many years. I mean, I would just

25:52

say to that point from my perspective, I

25:54

think the main thing that software

25:56

companies right now are struggling with

25:58

is like a lot of acronyms, a lot of

26:00

great marketing stuff.

26:02

But We call it we call it roadmap AI.

26:05

It's like we're seeing a lot of roadmap

26:07

AI nowadays. But but but customers don't

26:10

want

26:11

roadmaps. No, so that's the that would

26:13

be the one thing that I would say is

26:15

that I'm seeing a lot of companies that

26:17

I meet with whether it's private or

26:19

public

26:20

it's like they have a lot of everybody

26:22

but they have a lot of roadmaps.

26:24

But right now customers want

26:27

action. Look, I think it goes back to

26:29

like part of why I think Palantir has

26:31

been so successful is like

26:33

if you work with them as a customer and

26:36

you do their own co-

26:37

they

26:38

they could show results. And I think but

26:40

but like to some extent like you guys

26:42

are almost like there there is a lot of

26:44

similar DNA in terms of like this is not

26:48

top like in the you're seeing

26:50

>> We've because of you we're spending more

26:51

time with them and getting to know them

26:53

better and and it's been really

26:55

interesting. But what I would say is the

26:57

thing that the Palantir guys do just so

27:00

incredibly well

27:01

is they solve problems and they execute.

27:05

And and that's sort of key to success in

27:08

almost any business. Solve your clients'

27:10

problems

27:12

and execute. And Palantir does that. I

27:15

mean, they do that as well as anybody.

27:16

And and it's 8 of we we feel like we do

27:19

it in in all the world as well.

27:21

But it is it is crazy to see

27:25

sort of we'll go It's funny. If we're on

27:27

a pitch,

27:28

we might go into a room where seven

27:30

other companies have said they do the

27:31

exact same thing we do. And then you

27:33

actually

27:33

>> Like they're they're just making it up.

27:35

And then we come in and they're like,

27:37

"Oh, we

27:37

I said I'll I can say this to a client.

27:40

Great. Test us.

27:42

If I don't cut your cost to create

27:44

customers by at least 50%, I will give

27:47

you your money back."

27:49

Even on a test.

27:51

Cuz there's so many things like that's

27:52

where like where you're going to thrive

27:54

and obviously have a massive install,

27:55

but but a lot of it comes down to that,

27:57

right? It's like proof of concept. It's

27:59

proof of concept. And if you think about

28:01

it, that's one of the reasons our net

28:02

retention rate's so high. Mhm. Because

28:04

if you land with a client and then you

28:06

massively expand with the client, your

28:08

net retention rate's going to continue

28:09

to go up. We've given long-term guidance

28:12

of 110 to 115% net retention rate. We

28:15

were at 120 last year and we were well

28:17

ahead of that, the top of the range in

28:19

the first quarter of of 2026, which is

28:22

traditionally our lowest net retention

28:25

rate quarter of any year for a whole

28:27

host of reasons, and we were still well

28:29

above the 115.

28:31

That's because we come in

28:33

and say, "We'll prove ourselves."

28:35

And when you do that, clients move more

28:37

budget, which then naturally drives your

28:39

your net retention rate. But

28:42

it it I do think that it's crazy how

28:45

much road map AI is out there versus the

28:49

companies that are really delivering and

28:50

really executing.

28:51

>> Well, that's the

28:52

I think that's the separation. And

28:54

you're seeing like from cybersecurity to

28:56

software to that is like a huge

28:59

>> are good cybersecurity companies at this

29:01

point?

29:02

>> to to me it's like when you look like

29:04

CrowdStrike and Palo Alto,

29:07

they're almost playing a different game

29:09

than everyone else in cybersecurity

29:10

because of I could argue like Palo Alto

29:13

in terms of like cyber arc acquisition,

29:16

what George has done at CrowdStrike in

29:17

terms of building out the AI suite and

29:20

what they've been able to be successful

29:22

cuz right now when you think about

29:23

cybersecurity, that's going to be a huge

29:26

piece of every

29:28

>> Oh, for sure.

29:28

>> you know? For sure. So, I So, I just

29:31

think like what what we're seeing in

29:33

cyber

29:35

is very similar to what we're seeing

29:36

across other areas of software. And

29:39

looking like physical AI is just

29:42

starting, right? Like when you think

29:43

about like autonomous and robotics and

29:46

what we see with physical like like that

29:48

phase right now is just starting to play

29:49

>> I think world AI is going to be very

29:51

big. And I I think you're seeing a lot

29:53

of that come down the pike.

29:54

>> And then when you think world AI, it's

29:56

like US, you still have less than 5% of

29:58

companies gone down the AI path. Asia

30:00

ex-China less than 2%. Middle East has a

30:03

huge build out in regard obviously, you

30:06

know, conflict now, but if you look at

30:07

the data centers that are getting built

30:09

in Middle East

30:09

>> Yeah.

30:10

>> Europe continues to build

30:11

>> and and once the the conflict is done,

30:13

that'll come right back.

30:14

>> data centers outside the US are

30:16

happening in the Middle East, which just

30:17

shows Well, they're the closest to the

30:19

energy, right?

30:20

>> Well, and also and who's who's really

30:21

front and center of all of those deals

30:23

US tech. Right. So, I just think and

30:25

then also what we're seeing on the

30:26

government side is just it's I I think

30:30

it this next phase right now where

30:32

you're going to see just massive

30:33

government spending on AI. It's a huge

30:36

part of the administration that we've

30:37

seen in the beltway. So, you would sum

30:39

it up by saying you think the back half

30:41

of the year, how does that look? Oh, I

30:43

think that I mean we've said it we think

30:44

tech stocks are up, you know, another 12

30:46

to 15% second half of the year, but I

30:49

think stars of the show

30:51

will be the software companies that

30:54

could produce show the use cases. We're

30:56

in a skeptical phase now, but I think

30:58

that this is now we're now going into

31:01

the software phase of AI. Well, and I

31:03

would tell you for Zeta in particular,

31:04

with in addition to everything else we

31:06

announced

31:07

coming out of the first quarter was the

31:09

highest pipeline we've ever had. Our

31:10

pipeline was up 40%. So, I I I see it in

31:15

the adoption and I see it in the use

31:17

case.

31:17

>> And the timing. So, look guys, first of

31:20

all, thanks everyone for coming and as

31:22

always

Interactive Summary

Loading summary...