Eric Schmidt at the Morgan Stanley Technology Conference


Uploaded by Google on 08.03.2007

Transcript:

MARY MEEKER: We thought it made sense to have someone
speak very high level about technology trends on the first
day of our conference.
And the list of potential speakers is remarkably short.
And we're thrilled to welcome Eric Schmidt.
Eric, you've worked at some of the most innovative technology
companies, more--
also Mo--
enterprises ever: Xerox, Park, Bell Labs, Sun Microsystems.
You've been an engineer in Silicon Valley for more than
two decades.
You're on the board of Apple.
You're the chairman and CEO of Google.
You're uniquely suited to share your views on how
technology may evolve over the next five to 10 years.
We have 140 plus companies here from the technology area,
hundreds of investors dining on a wonderful lunch in a box
for the next 45 minutes.
And what are the five to 10 things in 15 minutes or less
you think we should be thinking about?
And don't worry, we'll also move to questions about
Google, and audience Q&A after Eric is done.
Thanks.
Thanks for coming.
ERIC SCHMIDT: Thank you.
Thank you very much, Mary.

It's an honor to be here and to join you for lunch.
In a technology conference, in an industry that is changing
so very, very quickly, and in the years that I have had the
privilege of being part of the technology industry, nobody
gets it right.
And I've certainly not gotten it right.
But I've learned some lessons that I want to share with you
and I want to make some observations about scale.
And let me begin by making an observation about travel.
Everybody here travels all the time.
And what I was trying to think is, what happens
when I travel now?
What do I see?
What do I hear?
And the answer is that I hear the sound of mobile phones
everywhere, all the time, all around the world.
And using that as a metaphor, the numbers are quite
interesting.
The 2 billion phones.
It took 20 years for it to get to 1 billion mobile phones.
It's going to take take four years to get to the second,
take three years to get to the third billion.
So if you think about it historically, we took this as
a sign of our own wealth and our own
privilege and so forth.
And now everybody wants a phone.
And everybody's going to use a phone.
And that is fundamentally the technology story.
We're in a business that's declining costs.
We're in a business that's driven by a
whole bunch of rules.
The first rule, which everybody here knows very
well, is called Moore's Law.
And the easiest way to think about Moore's Law
is that it's a factor--
everybody says doubling every 18 months.
It's a factor of 10 every five years.
And it's a factor of 110.
And when you think of it that way, you begin to understand
the underlying compounding is not done.
It's just beginning.
And there's something about humans, with respect to
compound interest, where we cannot do the math right.
We cannot foresee 10 years from now what a world of 100
times better really, really looks like.
If you look--
the numbers, and these are some set of numbers.
There are many, many different sources of research.
Probably the best is actually done by Mary.
277 broadband users in 2006, worldwide
growing to 413 million.
Anybody here remember the last time you were in a hotel
without a broadband connection where the dial-up modem didn't
work and your Verizon or CDMA 3GSM card didn't work, and you
were really upset?
What did you do 15 years ago when you were in every hotel
that was like that?
Maybe you had a life.
Maybe you had dinner with your friends or something.
And now you can't anymore.

MARY MEEKER: You can have box lunches at The Palace.
ERIC SCHMIDT: If you assume that at least in the majority
of the Western world everyone's going to have a one
megabit connection, which is roughly what 3G will offer,
then you see our future is to be always connected.
At the same time, the big screen revolution is also
driving behavior.
There's a lot of studies that indicate that these much
larger screens that people use are also much more productive.
So on the one hand, we talk about the mobile revolution
and small handheld devices.
But there's a similar revolution going on with very
large devices.
So you have a combination of broad--
essentially what we'll consider low speed
connectivity at a megabit.
And then these huge fiber based screens with enormous
amounts of information coming to them.

There is a return to massive data centers.
And I want to talk about that because it changes the way you
think about technology.
And of course, the combination of high quality screens and
the power of CPUs and so forth mean that things like personal
computers and Macintoshes are going to do very well over the
next while.
If you stopped making new media, it would be possible to
take all current media that's available today and put it on
a single hard drive in 10 years.
That's the good news.
The bad news is that the rate of production of new digital
media of all sorts is expanding as quickly if not
more quickly than any of the other scenarios that I'm
describing.
So of course the race continues, even greater
capacity and so forth.
It's the same speech.
It's the same message.
But we fail to do the compounding.
And it's clear to me that the quadruple play, the
combination of television, mobile, and fixed phone, and
internet on fiber, and coax, and others is very much going
to be the way in which we're all going to consume this.
And it will have competition between the various operators
of how they can do that most technologically.
When I look at this, I see real problems that matters to
consumers, information.
How are you going to find everything in this?
Well, of course that's search.
Health, number one priority for everybody, huge issues in
the US and everywhere else in the world.
By the way, you're going to find that with search.
And entertainment, which of course you're going to find
also with search.

If you think about the technology that as this plays
out, and you go back to my earlier point, there is
another underlying transition which is confusing for many
but is really the thing that's underlying and driving this.
It's generally known in a marketing term called Web 2.0.
But I want to talk about it technologically.
If you accept everything that I described, everything that I
talked about is going to happen, and you accept the
rate of Moore's Law which is clearly going to happen
because the physicists are busy working as we speak, then
we have an opportunity to change the underlying
computing architecture.
And I'm going to call it cloud computing.
Various terms have been used to describe this.
But it's important to understand, technologically,
its history.
In the 1980s, literally in this hotel, many companies in
earlier versions of conferences like this talked
about client server computing.
And the characteristic of client server computing was
that it was a proprietary operating system, typically on
a PC, a proprietary server based operating system, and
proprietary protocols in between it.
The internet has changed all of that.
And a new set of companies, in which Google, many of the
other companies represented here, certainly eBay, Yahoo,
Amazon, and so forth have exploited it.
The service is computed on the server and you need a
relatively thin client, a relatively simple browser.
The underlying technology that was [UNINTELLIGIBLE] of this
is called LAMPP.
It's Linux, Apache, MySQL, PHP, Perl, Python, et cetera.
And these technologies are what everyone is doing in
universities today.
All of the graduates coming out are
building on top of this.
This is a fundamental architectural shift, as
fundamental as the architectural shift that
occurred in client server computing that I was part of
20 years ago.
And the architecture that's used to build these
applications, generally known as AJAX which stands for
Asynchronous JavaScript with XML, is a way of having the
computers that are on the servers--
literally, the services and the data run on the service.
And then you can basically access them using your
ubiquitous connection, et cetera, et cetera.
All of those things that you were supposed to be doing when
you're having dinner instead.
And the fact of the matter is that that new architecture is
now sweeping through all of these industries and all of
these businesses.
And the simplest way of thinking about it is, anybody
here lost or dropped their computer recently and tubed
the hard drive?
It's a disaster.
So any rational person would put all of the personal stuff
somewhere else so that you can have lots
of different devices.
That transition is something that we have all talked about
and foreseen in the industry for many years, but it's
happening now.
And it's the underlying architecture that's driving so
many of these companies.

There's one other point that took me a very long time to
figure out.
That in addition to having the right technical architecture,
you also have to have the right business architecture.
Client server computing was populated and paid for by
enterprise hardware and software companies.
And I call this the Larry Ellison sales model.
Worked very, very well.
And Larry invented it, as best I could tell.
Direct sales force goes to a company, does a $1 million
software deal with recurring revenue.
And that pays for the software development.
It's busy building all of these client server
applications.
And I go to the engineers and I say, you should talk to the
sales people.
The engineers, typically clueless, go, why?
And I said, because they pay your salary.
If they don't exist, you can't do what you want to do.
Simple, easy to understand.
Be nice to the sales guys.
In the new model, a new version, a new way of
monetizing this world came along based on advertising.
In fact, based on targeted advertising.
So it is not just the architecture that I'm talking
about, but also the development of this new world
that occurred.

And I think I'll end with this because I think it's probably
more interesting to have Mary's comments
and you all's questions.
I want you all to think about the combination of the
technology trends that I talk about and the explosion in new
monetization as the metaphor for investing opportunities
for the next five, 10, 15 years.
It is that sum that I think makes this so
powerful and so effective.
Mary?
MARY MEEKER: Thank you, Eric.
Couple questions based on some comments earlier on in the day
to focus on high level stuff, and then drill
right down on Google.
You're on the board of Apple.
The iPhone captured people's imagination in a way that a
new product hasn't in a long time.
You talked about ubiquitous connectivity.
Could this be the thing that pushes the small form factor
device connectivity to a new level over the
next one to two years?
ERIC SCHMIDT: The iPhone is certainly going to be an
incredible product.
And part of the reason that it's been so incredible is
it's the first full-featured phone with a real user
interface designed around internet browsing.
So as Apple talks about the iPhone, they talk about it as
both a musical device, a communications device, as well
as a browsing device.
If you look at the history of these personal devices, they
were always closed and walled gardens.
They always did not have access.
But if you follow the metaphor that I was talking about,
people want access to everything.
They want access to all these powerful apps.
They need a device, of which the iPhone will be the first
of I think many.
And there will be many coming, I'm sure.
MARY MEEKER: So we should expect to see real impact from
mobile devices on a larger base of software companies,
not in the next six to 12 months but really
two to three years.
ERIC SCHMIDT: It's a function of how many devices there are.
One of the things to know about these devices is that
they're incredibly highly targetable.

Google's a strange company in a number of ways.
We, for example, don't have a revenue meeting.
We have a quality meeting.
We know that if we improve quality, revenue follows.
We do that because we do targeted ads.
And the more accurate the ads, the more highly targeted they
are, the value of the ads go up.
Very strange.
It's completely the inverse of how traditional advertising
works, where if you show more ads you make more revenue.
In a highly targeted model, if you show fewer ads that are
better ads, your revenue might go up.
Very interesting.
Phones turn out to be the first of a generation of
highly, highly targetable devices
because they're personal.
And they know where they are.
MARY: Yep.
If we turn to Google and look at revenue opportunities, one
of the things that has surprised people I think on
the upside is how you've been able to improve your targeting
and how you've been able to improve the monetization.
Can you give us a sense of where you are in the
evolution of that?
Are you more comfortable that you can continue to turn the
dials year in and year out.
If we look at 2007, our view is that targetability is the
big story that surprises people.
As receivers of advertisers, we start to get a lot more
targeted ads.

ERIC SCHMIDT: Without quibbling with your question,
the same question could have been asked in 2002, 2003,
2004, 2005, 2006.
MARY MEEKER: That's a good thing.
ERIC SCHMIDT: The fact of the matter is, people are
constantly surprised about the compounding.
They, and I mean generally, always miss the fact that the
world is a very large place, and that very targeted ads are
very, very valuable.
We are just at the beginning of our ability to do targeted
ads, both as an industry and as a company.
There's a tremendous number of new technologies that we and
others are inventing to do more targeting.
I mention mobile devices because you asked about that.
It's possible to do it based on many, many factors, of
which location and the person is just one.
MARY MEEKER: Just to ask one revenue associated question
and then move on to some stuff on video, because it's
certainly topical.
It is, do you think that local, mobile, and the other
stuff you're doing off of the internet, i.e.
the tools-related stuff, radio, print, et cetera, do
you think those things are enough to add material revenue
in each of the next two to three years, i.e.
local in '07, mobile in '08, other in '09?
If you don't want to answer that, that's fine.
But any color on when we should think about the timing
of the new stuff?
ERIC SCHMIDT: Again, a simple way of thinking about that is
how big is the global advertising market?
Some number between 600 billion and 800 billion
depending on how you add, growing at roughly GDP.

What percentage of that advertising is essentially
untargeted?
90%, 95%.
So even small improvements in targeting, that is
personalization, if you will, of advertising, can drive
extremely large amounts of revenue.
The easiest one to use, for those of you who've not been
following this space, is television.
Most households have a couple televisions.
Most households see a constant barrage of ads that are not
appropriate for that household.
There's no baby in the house, but the television shows baby
diaper ads, and things like that.
Or there's no pet in the house and they show pet food ads.
That sort of thing.
That's an opportunity cost. That's an opportunity missed.
So even a small improvement in targeting of any of the
categories you've described is a very large impact in terms
of revenue.
So we do see that.
It's hard to call which year and what the rate is because
these are large industries with complicated
infrastructure issues, all of which we're working on.
MARY MEEKER: Moving to video.
It's gotten a lot of play in the media recently, in part
because the media owns the media.
And they're involved in negotiations with you.
And they certainly have agendas.
But YouTube traffic continues to grow.
The monetization opportunity is a very real one out there.
The percent of video that's actually monetized is low.
You're striking a lot of deals with some of the
non-mainstream video partners out there.
How do you think this market plays out over the
next six to 12 months?
How important are the big media deals?
How important is the timing of it?
It certainly hasn't slowed down the growth of
the medium at all.
ERIC SCHMIDT: Some time in the spring of last year, online
video took off.

I don't know whether everyone got a camera at Easter or
something, a video camera.
But it happened in the United States.
It happened globally.
And everyone started posting everything that they could
think of online.
We saw this.
We had a product called Google Video.
YouTube of course saw this with an
incredible spike in growth.
Of course, that and a series of other steps took us to
acquire YouTube.
That growth rate is if anything accelerating.
We're seeing across the board huge growth in video, both in
terms of video search, video uploading, video content, and
so forth, as you said.
It started off as what I call my pet videos.
So, family videos, that sort of stuff.
But there's now an increasing number of people who are using
the internet for, as a short form, for entertainment and
eventually for monetization, I think.
We're talking to all of the companies that you named at
various stages.
And they're all trying to figure out how to move to this
new medium.
The difference between the old medium and the new medium is
the old medium is all on a per-view basis.
Sorry.
The new medium is and the old medium is not.
They pay, essentially.
So it's a difficult business model transition for them to
go through.
But they'll get through it in some form.
So our strategy is first and foremost to get as much
licensed content as possible onto YouTube and other sites,
to index everything, and to develop the advertising tools
that will enable people to actually make money on content
of this new medium.
The size of the business is very large.
It's unclear what the monetization will look like
because it's too early.
We're going to have to find that out.
It's the big project for this year.
MARY MEEKER: I hear that.
That said, you're starting to, for lack of a better
characterization, clean up the site a little bit to make sure
that you've got the right content on the site.
And the media companies have scrubbed YouTube a little more
than they had in the past. It is what it is.
You know how many views you have. It's not hard to wrap a
frame around a video.
You know you don't want pre-rolls on a lot of things
because your users don't want them.
You can measure the clicks.
The monetization opportunity seems to be pretty compelling.
And if I'm a traditional content company, I have
trouble getting traffic to my videos.
And you're the number one opportunity for
them in that way.
There's something missing in the puzzle.
ERIC SCHMIDT: There's an alternative model, which is an
explosion of a large number of specialized sites which host
and monetize their own video.
And that scenario is possible though not likely.
And the reason has to do with power laws.
And in the internet, the internet runs by power laws.
A power law essentially looks like what you think of as the
long tail curve.
And a small number of companies or institutions
and/or sites end up getting the
majority of the mine share.
And our strategy of course is to be the number one there.
And in that context, we're likely to have the most number
of viewers.
And we can monetize those viewers in many, many ways.
The scenarios that you describe, which were an iframe
as well as pre-roll, are two of many.
The pitch that we typically make to somebody who is a
content owner is that somebody who has gone through Google
and YouTube to find your content is very valuable to
you, Mr. Content Owner, because this is a fan.
And fan is a derivative of the word fanatic.
And you can monetize them.
You can sell things to them.
You can get them excited.
You can get them to buy more things.
You can get them to go to your movies.
You can get them to go to the basketball game,
or whatever it is.
This message is working.
People begin to understand it.
MARY MEEKER: Do you think we'll see an increase in
content companies purchasing keywords to drive traffic to
their sites?
ERIC SCHMIDT: We already are.
MARY MEEKER: OK.
You've spent a lot on capex.
I'll call it the server nation, for a lack of a better
characterization, or server globe.
One of the things we learned at your analyst day about six
to nine months ago was that you're actually
buying servers to demand.
You're not buying in anticipation of demand.
So clearly there's a lot of traffic on those servers.
Could you give us a sense of what the biggest incremental
drivers are to that demand?
And could you talk a little bit about WiMAX?
There's actually a connection there.

ERIC SCHMIDT: We have a number of
shortages inside the company.
One of the most acute over the last few years has been data
center capacity.
And this is after special designs of hardware and
special performance engineering that could only be
done at Google, which is highly proprietary and very
interesting from a computer science perspective.
Part of the reason we have so much demand is we have so much
more traffic than everyone else.
The other one, of course, is that we have much larger
indices than everyone else.
And we also do many large replicas.
So it's [UNINTELLIGIBLE].
We have a larger index.
We have more copies of it.
And we have more global reach.
It's the combination of that that drives the
higher capital spending.
Many people, and people in this audience, have asked me,
why do you need to spend so much money on capital?
And the true answer is that if we spend more, we will in fact
generate more traffic, more monetizable opportunities, and
more revenue.
So there's a direct connection between our capital spending
and our revenue results and our market
share growth globally.
Another way to say this is that two years ago right after
we went public, people were criticizing about this.
But the investments we made two years ago caused these
exceptional returns we had last year.
So there is in fact an implication.
Had we not spent that money, we would not have been able to
serve that traffic.
Those customers would have gone to competitors, who would
be doing better now than we are.
MARY MEEKER: WiMAX.
Paul Otellini was sort of optimistic
about it this morning.
And I bring it up in the context of increasing demand
and increasing access.
What's your view on your two to three year outlook?
ERIC SCHMIDT: We have a direct correlation between broadband
access and Google monetization.
Everyone who is a broadband user is much more likely to be
a Google user.
And they're much more likely to use
our advertising services.
So anything that we can do, we Google to encourage the
adoption of broadband in many forms, of which WiMAX is one.
We're reasonably indifferent as to whether it's WiMAX or
the other choices.
And it's clear to me that WiMAX will be one of the
successful approaches along with fiber to the home and the
others, which are very important from our growth path
perspective.
MARY MEEKER: One of the things that the media loves to do,
one of the things that investors love to do is pit
Google against eBay, and pit Google against Yahoo, and pit
Google against Microsoft.
What's your view of the outlook for those
companies in general?
And I say that because I think you're doing a lot of
different things and all help the entire internet show
relatively strong growth.
One might be excluded, but--
ERIC SCHMIDT: Microsoft is always a special case.

The companies you named are critically dependent upon the
build out of the internet globally.
So we share a set of political agendas and public policy
concerns, regulatory concerns, and so forth.
I'm sure that broadband rollout is also good for them.
Each of the companies you named has different
strategies, and all are riding at various stages of the wave
that I talked about.
I don't really think it's appropriate for me to talk
about them directly except to say that, for example, many
people are concerned about
competition in a market where--
because they assume that these are zero sum games.
And an awful lot of people have said, well, if such and
such a company enters the market, that will affect your
advertising revenues.
But if you think about it, advertising is
not a zero sum game.
If you have a product you want to advertise, you're going to
advertise on more than one of those choices.
So the fact that one product is doing better or not does
not necessarily mean that it would affect us negatively.
In fact, it might grow the market.
And that's what we've been seeing so far.
MARY MEEKER: On the competitive front, you're
pushing Checkout very hard.
Why not just use PayPal?
ERIC SCHMIDT: Checkout is actually a little different.
PayPal is a stored value system for person to person.
And Checkout is a way to get--
the moment you see a product to purchase it and cause it to
show up at your home or your office.
And so far we're seeing a direct correlation between
advertiser satisfaction, quick checkout times and, we
believe, bidding results.
So we believe that the Checkout is central to causing
the value of our ads to go greater.
Because if you think about it, the
advertiser's trying to sell.
Using Checkout, they can sell that thing that much quicker.
So Checkout can best be understood as a
tactic around velocity.
Just a tactic.
MARY MEEKER: User-generated content has a lot of momentum.
YouTube is clearly at the forefront of that.
So is Wikipedia.
Do you look at them as a competitor or a partner?
They've shown tremendous growth, lot of momentum,
answering questions, and are becoming a resource for a lot
of-- and you're providing them with a huge number of links.
ERIC SCHMIDT: Wikipedia is an incredible global success.
And a lot of traffic through Google ends up at Wikipedia.
So we think it's absolutely wonderful.
When you think about user-generated content, it's
important to remember that people have a lot to say.
And the new generation of internet technologies is
enabling them to say exactly what they think, whether you
like it or not.
So the rise of social networks, which are baffling
to many people of a certain age, all of the--
MARY MEEKER: People of a certain age.
ERIC SCHMIDT: People of a certain age.
MARY MEEKER: I like that.
ERIC SCHMIDT: At least me sometimes.
MARY MEEKER: I don't think you're baffled.
ERIC SCHMIDT: It's a new phenomena.
And it is very real.
And it's very, very large.
And it's going to be with us for a very, very long time,
like the next 1,000 years.
So people have a lot to say.
And from our perspective, that's good because it means
more search.
But anybody who thinks that the world consists of the
model that I grew up in, which is a relatively closed network
of communications models, relatively traditional modes
of communication, that's all gone.
And it's gone because everyone's going to have a
mobile phone.
Everybody's going to be taking pictures.
Everybody's going to be talking to each other.
And they're going to be doing exactly the same thing.
And if you don't believe that, talk to your teenager.
Or more likely, watch them talking to everyone else while
they ignore you.
MARY MEEKER: I'm going to ask a few more questions, then
turn it over to the audience.
But Google gets a rap for being arrogant, again, in the
media with some of the partner companies that are out there.
The irony is that you take 30% of your revenue, your gross
revenue, and hand it back to partners.
You have hundreds of thousands, maybe 1 million,
close to-- whatever it is, a large number of advertisers.
You're making a very large number of people very happy.
Are you [INAUDIBLE] use that effectively as ammo in the
negotiations with some of the people that are trying--
I guess the word frenemy, which was new to me, has been
created around Google?
And the reality is, you're one of the best partner companies
I think the world has ever seen, if one looks at volume
and revenue sharing.
ERIC SCHMIDT: Historically, the company has been certainly
arrogant in a number of ways.
And it was not until a few years ago
that we started focusing--
and this was an error which I'll take responsibility for--
on partnerships.
We didn't tell the story.
We didn't talk about people.
We didn't have the people inside the company to talk to
the other partners, et cetera, et cetera.
That has largely now been remedied.
As you pointed out, we share an enormous amount of our
gross revenue in the food chain throughout the
organization.
And it has worked well.
The media are obsessed with Google on these issues for a
number of reasons.
The internal joke is that it's 70/20/10 at Google.
70% of the press attention is focused on 10% of Google, 20%
on the 20%, and 10% on the 70%.
And maybe that's the nature of how we're
going to get covered.
But the fact of the matter is that the model that we are
seen as iconic with respect to a set of model transitions
that are very powerful, that would be occurring whether
Google was part of it or not.
So we've taken the position that
these models are shifting.
We will work with you to help you get there.
And we will furthermore share in many cases the majority of
the revenue with you to make that.
And that works after the initial conversation.

MARY MEEKER: A couple financial related questions.
Google has a lot of cash, very high operating margins.
How worried should investors be that you just may turn on a
dime and do something to radically change those
positive attributes?
ERIC SCHMIDT: It's highly unlikely.
One of the problems in the high tech industry, and a
number of you have commented on this, is that successful
companies tend to generate cash pretty liberally.
And they don't tend to have good and rational
places to put it.
And Google is careful with these things.
So it's not obvious to me where it would go.

MARY MEEKER: The stock market hasn't been performing very
well of late.
And the market is oftentimes an indicator of what may
happen in the economy six to 12 months down the road.
Any general comments on Google's resiliency in an
economic downturn, having not lived through one yet as a
company but certainly having lived through a few yourself?
ERIC SCHMIDT: Of course, we lived through 2001, 2002 as a
private company.

MARY MEEKER: That was fine.
ERIC SCHMIDT: It was a terrible situation globally,
terrible in the United States.
And of course the company didn't have any cash at the
time, a small private company.
I was terrified that somehow the company would-- people
would stop using our product.
And the curious thing was that people accelerated their
spending on Google and decreased their spending on
everything else.
And the reason was because it was more measurable.
So I can't say for sure, of course.
But we seem to be more sensitive to the transition
from the rate at which people transition from untargeted to
targeted mechanisms, and less sensitive to the kind of
question that you're describing.
The company has many, many other ways in which it can
insulate itself from these things.
As you know, for example, our European business is
incredibly strong and extremely
profitable, et cetera.
We're expanding dramatically and quickly in Asia.
So if there were a global downturn, it may very well be
that the principal that we saw in 2001, 2002 would recur.
Maybe it would be a new thing.
We don't really know.
MARY MEEKER: Except I don't think you'll be low on cash.
ERIC SCHMIDT: That's correct.
MARY MEEKER: Come on.
This is a tough crowd.
But I have two more questions about other industries and
then turn it over to other questions, or I'll keep going.
And I'm going to read these so I get them right.
In one to three years, do you think there's a better than
50% chance we'll see a resurgence in TV advertising
as broadcasters and cable operators get smarter about
working with internet marketing,
especially in local markets?
ERIC SCHMIDT: If you go back to the earlier example I was
using about television, and I was saying the television set
that's showing the wrong kind of ads.
There is a solution to this, which is many people have IP
addressable set top boxes, whether they're from the cable
company or satellite based.
And a new generation of such boxes are highly addressable.
With that addressability, it should be possible with
television to show much more targeted ads.
And we can debate the mechanisms of targeting.
So we at Google have certainly looked at this.
We have a whole bunch of partners, and we're
exploring that area.
So the answer is yes to your question, but only because the
ads are targeted.
MARY MEEKER: Yep, gotcha.
Similar question.
In one to three years, do you think there's a better than
50% chance that we see a resurgence in revenue from
leading video content providers that aggressively
pursue the internet as a distribution channel?
You've spent time with these folks over
the last six months.
They are clearly dealing with some channel challenges.
As you know from the history of technology, it's always
hard to make the bet.
You can make it up on volume.
But this is one of these opportunities in front of us.
You've gotten probably a little closer to looking at
their challenges and opportunities.
Do you think that that is an opportunity for them from a
holistic level two to three to four years down the road?
ERIC SCHMIDT: Every one of the large media companies either
has to or already has an internet strategy.
And they face some daunting issues.
They face issues of piracy.
They face issues of monetization.
And they also face dissatisfaction from their
existing licensees.
Plus, they have this enormous rights rats nest--
sorry about the alliteration--
where they have very complicated set of rights, but
which don't necessarily allow them to do the things that
they would like to do.
So it is the sum of those issues that I think slows down
progress in that area.
It should be the case that monetizable viewing habits on
the internet should eventually catch up to where television,
for example, is.
And today there are a number of studies that indicate that
the time spent, if you will, doing video and audio and so
forth on the internet is monetized as a rate of a
factor of five or a factor of 10 less, on
a per minute basis.
So it looks like with new tools and technology, as the
media companies shift, we can replace that revenue.
As to whether it will grow, we'll see how big those
markets are.
MARY MEEKER: Yep.
OK.
Any questions from--
go ahead.
Fire away.
AUDIENCE: A question here on Apple.
You are on the board of Apple.
And they have a closed model of video delivery.
And you have an open model video delivery.
And how do you manage the conflict of being on the board
and understanding their strategy?
Was yours quite different?
And the next question is about the mobile device, the iPhone.
There have been rumors that Google might also introduce an
internet communicator, which also is another
[UNINTELLIGIBLE PHRASE].

ERIC SCHMIDT: It's important to understand the proper role
of a board member.

Reading the press, it seems like everybody's confused
about this.
And a board represents shareholders to make sure that
proper governance and so forth and so on.
The management team runs the company.
So I'll let Apple speak for why their
strategy makes sense.
I think Apple's success in the last few years shows what an
incredible management team and what a clear strategy they
have.
I don't want to comment on rumors.
I will tell you that Google and Apple are doing more and
more things together through the normal course of
communications.
Apple remains one of the great innovators in the areas that
you all know.
And we hope Google is the same.
We have similar goals and similar competitors.

AUDIENCE: Yeah.
Good afternoon.
Two questions, one non-search based and one search based.
Could you put together a reasonable approximation of
what non-click revenues will be in '07?
And at the end of the year, if you choose not to break that
out, why not?
And then the second question, in terms of--
[UNINTELLIGIBLE]
on the last quarter said that your quarterly growth, I
believe, was up 61% on a year to year basis.
In your K, it was up 65% for the entire year.
So I was very surprised at the slight attenuation as the year
progressed, in that it's extremely healthy.
Could you go behind why it's so sustainable, as robust as
it is for the whole year?

ERIC SCHMIDT: First place, two separate questions.
With respect to non-search based revenue, we don't break
those things out.
They're not material yet, from a financial sense of
materiality.
And we're still exploring how to make them much larger.
The enterprise business has done very well for us with the
search appliance.
And if they were a standalone business, it
would be a huge success.
Course, it's buried inside of this huge advertising engine
that is Google.
The next really big one is actually an extension of
Google Apps.
And the reason is it's an extension of
that enterprise business.
And we've had a series of announcements there.
The other non-search businesses are too early to
forecast. When they hit, they're
likely to move quickly.
In which case, we'll let you know.
So I don't want to give you a precise number for 2007.
With respect to query growth, Google's query growth has
continued, as we said in the 10K and
10Q, to be very strong.
And we attributed this to a number of factors.
The first is a focus on international.
Mary, who's one of the experts on the international world,
has not actually asked any questions about
international focus.
But the quickest--
MARY MEEKER: I already know everything.
What can I say?
Joking.
ERIC SCHMIDT: Maybe you covered it earlier.
But it turns out that her team actually has
mapped all this out.
And you can study it.
And it's very, very interesting.
The quickest way to grow queries, and in fact
advertising revenue and traffic, is simply take the
current product and make it available outside of your
current markets, and deepen your market share in every
country around the world.
And that's been our number one focus.
It turns out that even in developing countries where the
monetization is small, there are so many people that the
multiplication of the number of people who are doing
queries times the low monetization
equals a large number.
So we're very, very focused on delivering services worldwide.
The second is the expansion of our content, and in particular
of comprehensiveness.
And we've also seen significant increases in
search quality.
So again, the media and everybody's always excited
about this acquisition or this deal or so forth.
The core business of Google, which is how we spend 70% of
our time because it's the 70% part of the business,
continues to grow well because of those factors.

MARY MEEKER: Any other questions?
Just an international question while we're waiting.
Or I'll keep going, but I assume there'll
be some more questions.
Any interesting observations about YouTube usage in
developing countries?
ERIC SCHMIDT: Follows the same rules as everyone else.
Remember what I said, was that everybody has
something to say.
And it's amazing to see in countries where people have
never really had a voice, how powerful their voice is.
And one of the things that's really exciting about the next
five to 10 years is, what will happen when the next billion
people come online?
And they're going to come online because of inexpensive
mobile phones, which very wonderful companies are
pushing very hard.
They're going to come online because they're going to want
to have the same kind of information that we all have
the benefit of.
And they're going to come online because they can get it
in their own language.
So the general adoption rate is a function of access.
There's a basic level that you have to have
some electric power.
And you have to have a computer or some kind of a
mobile device.
People will do it.
And their voices will be found.
MARY MEEKER: We had a little bit of an epiphany about six
months ago when we realized that there will be more mobile
phones in the next six months with recording devices or
cameras than there are internet users today.
ERIC SCHMIDT: That's right.
MARY MEEKER: Any indication of how relevant that will be for
video uploads on the internet?
ERIC SCHMIDT: We, of course, view this is as a
great search problem.

We don't know what people are going to do with these phones.

It's worth noting that in the United States, the internet in
the 1990s when it came out gave a voice to people who had
never been heard before.
That voice has had its problems.
So for example, everyone's excited about criticizing
various forms of polarizing bloggers.
But they existed 10 years ago, but they didn't have a voice.
And now they have a voice.
So you should expect the same phenomena occur globally.
And this is a good thing.
It's ultimately about user empowerment.
It also puts a burden on citizens of any country to try
to sort out, are they being manipulated?
Are they being disinformed?
There's another phenomena that we worry about, which is
essentially--
think of it as political spin applied to
non-political worlds.
You'll see, for example, people say, well, my website
got 10,000 hits or whatever.
What happens if 9,999 of those were paid for by somebody?
So it's possible that in this new world that new forces will
emerge that will try to shift opinion by various forms of
marketing disinformation.
And we don't really know.
MARY MEEKER: Any other questions?
Fire away.
AUDIENCE: It's a question about broadband rollouts.
I guess Korea and Japan [UNINTELLIGIBLE]
the fiber at home customers today, primarily because it's
been centrally directed, the rollouts, whereas the US and
Europe continues to be delayed.
And you see a lot of articles about it, but not much new
subscribers, really.
So what are you doing?
What do you feel like can be done to push the FCC to be
more forceful with these monopolies to rollout fiber
more rapidly?
ERIC SCHMIDT: The American system, as you know, is
different from in those countries.

We have always pushed for more broadband access.
And I think the countries that you named see broadband
leadership as competitive leadership for their country.
It would be great if the United States saw this as a
threat, and invested the money and the subsidies and so forth
to essentially close that last model.
It's harder in the United States because of distances.
But it's also good in the United States because we have
so much great technology.
Many people have argued that there's some sort of internet
bandwidth shortage.
And the only internet bandwidth shortage that exists
is the last mile, the last kilometer, or whatever.
There's so much dark fiber put in the ground over the last
five or 10 years, especially during the bubble and the
bust, that as long as you can get to one
of them you're fine.
And indeed, those prices continue to fall very
dramatically.
Google is a large consumer and owner of a whole bunch of that
for the way our data centers work, so we
know this to be true.
We at Google have been sponsoring a number of things,
including various forms of free WiFi and so forth, to try
to raise awareness of this.
Ultimately, because of the scale involved, it will
require some form of government collaboration with
the key leading broadband providers in each of the
countries, because they're going to need some help coming
up with the financials and the necessary wherewithal to spend
all of that money.
We don't ask private citizens to pay for
the highway up front.
They pay it in the form of taxes.
So there must be some way where we can make an analogy
to the interstate system, and shows how nationally
competitive it is.
People may not remember that in the 1950s, the US
Interstate System was built under the rubric of nuclear
war as a national defense issue.
And it may get to the point where we will feel that a
national broadband policy is central to the competitiveness
of the United States.
And I certainly encourage that.
MARY MEEKER: Any other questions?
ERIC SCHMIDT: Over here.
MARY MEEKER: Go ahead.
AUDIENCE: Could you comment on what your thoughts were on
Yahoo's Panama, to the extend that increased monetization
for Yahoo and looking at competition in a different
way, competition for affiliate deals, which is obviously an
important part of both of your strategies?
And then secondly, speaking of affiliate deals, maybe you can
give us some color around monetization around MySpace
deal that you guys have?
ERIC SCHMIDT: The MySpace deal is doing extremely well.
MySpace traffic is doing very, very well.
And our revenue and monetization is correlated
with traffic, as I was describing earlier.
So again, looking at the likely age of the audience,
you all may not be huge MySpace users every day.
And I can assure you that MySpace is taking off and
doing very, very well.
With respect to Panama, Panama as I understand it is Yahoo's
advertising product that they've been rolling out.
And as I said earlier, the fact that there is a new and
interesting advertising system does not mean that it affects
ours in a negative way.
It may, in fact, be positive.
So we've not seen an impact of that so far.
You never know, of course.
And the reason is that advertising systems--
people don't advertise on just one.
They will put their money in multiple places.
With respect to affiliate deals, and so far we've been
able to grow our monetization and profitability on affiliate
deals faster than any of the competitors, which is why
we've been able to keep them.
We don't see any reason to believe that
that's going to change.
MARY MEEKER: One more question?

AUDIENCE: You mentioned earlier about capacity growth,
and data centers, and servers, and such.
Can you talk about what that's doing for power costs?
And do you have any perspective on the electric
utility grid, and what type of growth needs to happen there
in order to satisfy yourselves and others?
ERIC SCHMIDT: We of course want green power as well as
inexpensive power.
And so we have teams looking around for hydroelectric dams
that have excess capacity.
And there's this very complicated set of criteria
about power usage, the amount of power we actually use, and
local utility rates.
It's been widely reported that we are in discussions with a
number of sites around the United States, and we also
have them globally as well.
So those negotiations continue.
We will be able to get enough power by following the rules
and by doing the kinds of things that I just described
for the foreseeable future.
There is enough power.
Again, we would prefer and in fact willing to pay a moderate
premium in order to actually have it be the right kind of
power with the right kind of reliability.
And we're negotiating that.
But the fact of the matter is, there is enough power
available to us.
The broader power issue is really a question of what are
the best sources for energy and so forth, which is an
issue we can discuss another time.
MARY MEEKER: Anything we've missed you'd
like to address, Eric?
ERIC SCHMIDT: There was a question.
MARY MEEKER: We have one more question, and then I'll ask a
final question.
And then we need to close.
AUDIENCE: Can you just comment on your thoughts on changes in
the competitive landscape with Microsoft to just to bite the
bullet and essentially buy Yahoo for the extra footprint
to compete with you?
ERIC SCHMIDT: There's just no way I
can answer your question.
I can't think of a legal way to answer that question.

If the market has gotten to the point where it's going to
consolidate in a way like that-- let me make up an
answer that's kind of near your question.
Then it would be early for the market to consolidate.
There are markets in technology--
Oracle is an example of a company that's busy
consolidating for growth, and doing I think a very good job
of it as best I can tell.
I don't think we're at that stage in the market.
There are so many new places where monetization can occur.
There are so many new initiatives, so many new
places where targeted advertising, information, et
cetera, can be done.
It does not strike me as the right time to be consolidating
the whole market and becoming normal.
This is a market that's still explosive in innovation.
It's still a market where new things can be invented this
year that we'll be talking about next year as though we'd
had them for years.
We're still in the early part of the S curve
of innovation cycle.
And I think that's going to continue for some years.
To answer your--
MARY MEEKER: One more or are we good?
One back, far back.
But we do need to--
AUDIENCE: Yes.
I was very interested by your comment about the relationship
between the value of advertising and the quality of
the consumer experience as it pertains to the Checkout.
What are your thoughts, if you can share them, around trust
and safety issues in e-commerce as
it pertains to Checkout?
ERIC SCHMIDT: All of the companies that I've named have
big initiatives around trust and safety.
eBay is one of the best examples, in the positive
sense of having--
all of them have pioneered.
Amazon, et cetera.
So we have similar, similar policies with the branding,
consumer fraud, and so forth and so on.
So far it's not been an issue.
From the stand point of a consumer,
consumers trust Google.
Consumers have been willing to give us their personal
information, and then we store it.
And with respect to Checkout, one of the boxes that you can
establish is that we do not give all that information to
the consumer.
Excuse me, to the store.
And so this is important because it means that the
consumer using Checkout may believe, and indeed it's true,
that it's a safer transaction.
And we think that that's a competitive benefit that we
have.
MARY MEEKER: Anything we missed, Eric,
you'd like to address?
ERIC SCHMIDT: I think that the most important thing that I
may or may not have talked about here very well is how
much bigger this opportunity is than we think.
The reason I went through the Moore's Law example and I use
the examples is that people don't sit down on a piece of
paper and they don't plot it out.
They don't see how large these markets are.
This is a mistake I make all the time.
And I've committed myself this year to not make it again.
I'm sitting there plotting it out over a two
or three year period.
How big of this market?
How much of this particular market could Google get?
How much could this market could Google get?
How much could a competitor get?
And I would encourage you when you're doing your analysis,
which I know is very rigorous, take a longer view and
understand the underlying change and the rate in which
these things are being adopted.
And then try to figure out if that's likely to where all the
money's going to get made.
MARY MEEKER: It's a good way to end.
Thank you, Eric.
I appreciate it.
ERIC SCHMIDT: Thank you very much.