February 12, 2024 • 43 mins
For a couple of years in Google's early history, it was a real search company. But that would change in 2000 as Google would embrace advertising. Learn how Google would shift to become a dominant force in online ads.
See omnystudio.com/listener for privacy information.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:04):
Welcome to tech Stuff, a production from iHeartRadio. Hey there,
and welcome to tech Stuff. I'm your host, Jonathan Strickland.
I'm an executive producer with iHeart Podcasts and how the
tech are you. I've often said that Google is not
(00:25):
a search engine company. It's an advertising company. And of
course these days, when I say Google, you could argue
that I really mean Alpha, that you know, the parent
company that owns all the various Google companies. But to
be clear, I am not the only person who says this, right,
I don't mean to say that I'm some sort of
(00:45):
guru who I did not come up with it. I'm
not the first person to say it. I also say it.
That's all. I agree with those who say this. But
here's something that I find interesting. So, once upon a time,
Google was a search company. It was not an advertising company.
In the very early days, Google didn't generate revenue through advertising.
(01:08):
So let's set the clocks back quite a bit. In
nineteen ninety five, a guy named Larry Page was mulling
over the idea of enrolling in Stanford University in order
to go to grad school. He was at the campus
in order to get the lay of the land, and
there was a student who was assigned to him to
show him around, a Stanford student named Sarah gay Brinn,
(01:29):
and the story goes that the two got along kind
of like oil and water. They apparently didn't agree on
practically anything, but a year later, when Larry Page was
in the graduate program, the two decided that they would
work together on what would turn out to be a
truly huge project. So their goal was to build a
(01:52):
new kind of web search engine. So at the time,
the dominant search engines that were still working off the
very young Worldwide Web primarily used keywords as their guide.
So you would pop into a search engine. The one
I used all the time was web Crawler, but there
were lots of them, and you would type whatever it
(02:12):
was you were searching for, like let's say it's the
word tractor, for example. The search engine would return results
of web pages that featured that keyword somewhere in the page,
and they might have different ways of prioritizing the web pages.
Maybe if the keyword appears in the title of the page,
it might rank higher.
Speaker 2 (02:32):
That kind of thing.
Speaker 1 (02:33):
But the problem was there was never a guarantee that
the search results you got would actually be relevant to
the search query that you put in. They might include
the search term, but it may not turn out to
be useful. Maybe the page just mentions tractors off handedly,
like as a joke or as a reference, and it's
(02:55):
not actually about tractors, it just happens to have the
word tractors in it. Or maybe it's a web page
run by a real scumbag who just crams the bottom
of the page with a ton of random keywords, ones
that they've determined are really popular, and it's on an
effort to capture as much traffic from search engines as possible.
(03:16):
They don't care that this isn't what you were looking for.
They just want to get you onto that page. The
search experience could be really a frustrating one. It often
involved users going to a page, taking a quick look,
saying oh, this isn't what I need, backing out, and
then doing it all over again. And it wasn't very
(03:37):
satisfying or helpful page. And Brend wanted to create an
engine that would return much better search results.
Speaker 2 (03:44):
So how would you do that?
Speaker 1 (03:45):
Well, their solution was to take a broader approach to search.
Their search engine would assign a rank a page rank
if you will to each result, and higher ranks would
appear higher in the search results page and lots of
factors would go into determining what page rank a page
would receive. For example, let's say that you've got a
(04:07):
web page and you notice that there are tons of
incoming links pointing to this web page. Well, that could
end up being a boost to that page's rank, because
the thought is, if a lot of other web pages
are pointing to this page, it must be good, right,
if you've got a lot of different pages. Of course,
you can game that too, right, You could create a
whole network of pages that just link to each other
(04:29):
in an effort to try and boost page eranks. So
another part of this ranking system would take into account
the quality of those incoming links, the idea being well,
if the incoming links are coming from really established websites,
maybe they belong to a known entity like an important company,
(04:49):
or maybe it's a web page that has been around
for a while and has sort of proven itself, any
links that that page has are going to be worth more.
Plus those pages themselves, the ones that been around or
from known entities, those would have a higher rank because
they were thought of as just being higher quality. Now,
this is just a couple of examples of factors that
would determine a web page's page erank.
Speaker 2 (05:11):
As it turned out, there were.
Speaker 1 (05:12):
Lots of them, and the formula would change frequently, and
it was always kept secret so that pages couldn't try
to game the system. Although they didn't stop anyone from trying,
it just meant it was hard to do because you
didn't exactly know which factors were the most important. The
PageRank approach meant that you were far more likely as
(05:32):
a user to come across a search result that was
relevant to whatever it was you were actually searching for.
Speaker 2 (05:38):
Now.
Speaker 1 (05:39):
Originally, the pair called their search engine BackRub. I think
we could all agree that it was a fortunate thing
that they decided to change the name to Google. The
two received a healthy one hundred thousand dollars check from
an investor in nineteen ninety eight, and they launched Google Incorporated,
and for the first couple of years, Google existed primarily
(06:01):
off of influxes of cash coming from investors. Everyone could
recognize that Google was providing a valuable service, one that
was becoming increasingly important as the web was getting more
complex and larger, and it was doing so at a
level that really set Google apart from the competition. Google
was quickly making a name for itself as being the
(06:22):
best of the web search engines at that time, but
investors were also starting to get a little concerned, and
the reason for that was that the powerful Google search
engine didn't have an equally powerful way to generate revenue.
Speaker 2 (06:35):
Now.
Speaker 1 (06:35):
Initially, one strategy was to try and form partnerships with
portal sites. And these were websites that acted kind of
like a homepage, like this is the page that you
would see when you started up your browser. This was
kind of a gateway to the rest of the web.
They might host a few articles. Maybe some of those
articles were written in house by an editorial staff that
(06:57):
worked for the website. Maybe some of the articles were
really generated by a third party and they were just
being linked to on this landing page. And many of
them would also include a search tool. So Google would
try and make deals with these different sites to be
that search tool. But Google with small potatoes in those days,
(07:18):
so these deals were few and far between, and they
certainly weren't enough to make Google look like it was
going to be arousing success financially, So clearly Google would
have to come up with a different approach if the
company was to stay in business, and someone else had
to come up with an idea that, with a little tweaking,
would work just fine, and then Google would take that idea.
(07:42):
To say they stole it would probably be too harsh,
but they certainly emulated an idea that someone else came
up with for a totally different site. Now that's someone
else was a guy named William Bill T.
Speaker 2 (07:56):
Gross.
Speaker 1 (07:57):
He graduated from the California Institute of Technology with the
degree in mechanical engineering, which would not make you think
of someone who would go on to create a web
search engine, but he became known as something of a
serial entrepreneur. He would launch lots of different companies across
his career as soon as he was graduating, essentially, including
(08:17):
a video arcade and a couple of software companies, and
an audio equipment company and more. But in the late nineties,
Gross wanted to solve the same problem that Page and
Brinn had been working on or would be working on, actually,
because Gross's solution would come out first. He wanted to
be able to create a search engine that would return
better search results than a lot of the competitors that
(08:40):
were on the market. But he took a different approach
from what Brin and Page were looking at. So instead
of building out a system meant to assess the value
of every page that's on the Internet, he figured he
could deliver better search results by having companies pay him
in order to have their search result rank higher on
(09:00):
the list. So here's the basic idea. Now, as I
mentioned earlier, one of the big problems with early search
engines is that a lot of web administrators were trying
to game the system. So it really wasn't unusual for
web designer to just stuff a page full of keywords
that didn't actually apply to the page's content itself. So
(09:20):
clever designers would do this in a way that wasn't
obvious to the visitor, right because otherwise you would just
like I said, you would scroll down a page and
you see like a word jumble of a bunch of
unrelated words at the end of a page, and it
just looked like a mess. Clever web administrators would choose
to make the text the same color as the page's
(09:41):
background at the bottom, so you wouldn't even see it
like it might be that the page seems unusually long
with no content toward the bottom of it, but in
fact there would be content. It's just it was the
same color as the background, so you couldn't see it.
Speaker 2 (09:55):
However, while you.
Speaker 1 (09:57):
Couldn't see it, a search engine could. The search engine
would still see the keywords there even though you wouldn't,
and that's why this page that doesn't seem to be
related at all to the thing you were searching for
would pop up in search results. Meanwhile, you had legitimate
sites out there, some connected to established brands, that would
very much like to rank high in search in order
(10:19):
to drive more traffic to their sites, but with all
these junk sites out there, that could be hard to do.
So they might be in a position where they're able
to and willing to pay to jump to the head
of the line if it meant that folks would actually
find their stuff online. And so Gross launched a company
called go to dot com Incorporated. His strategy included a
(10:42):
pay per click arrangement, which is exactly what it sounds like.
The advertiser agrees to pay a certain amount every time
someone clicks on the ad, so typically you aggregate clicks
by the thousands, so each click is really just worth
you usually fractions of a penny. Gross's logic was that
the garbage sites out there couldn't and wouldn't pay for
(11:06):
positioning because their entire strategy relied on using keywords in
bulk that had no application to the actual site. So
if they wanted to try and do that the same
way with Gross's strategy, they would have to pay for
all these different keywords that didn't actually apply to them,
and it would bankrupt them because people would be clicking
on those sites over and over again on the search results,
(11:27):
and they'd have to pay every time that happened, or
at least every time every thousand times it happened. So
you would be paying go To for all this traffic.
You wouldn't be making that money back through the ad
revenue you generated on your site. There's no way you'd
be making more money through say, web advertising, than you
were paying out to go To. So whamo, you end
(11:48):
up with a search engine that delivers better results, or
at least more relevant results. Since you can't guarantee that
the entities that actually do pay for their positioning are
in fact the best ones out there, but chances are
there better than the junk sites. Gross launched goo to
dot com in nineteen ninety eight. In fact, he launched
it before Page and brend got Google off the ground.
And while go to only saw modest success in the
(12:10):
grand scheme of things, that revenue generation model stood out,
and when it came time to determine a way for
Google to make money, the go to became well, well,
it became the go to model. So Google would introduce
a pair of strategies in two thousand that would bring
(12:31):
the company into the advertising space. Now, according to a
Google SEC filing from two thousand and four, quote, in
the first quarter of two thousand, we introduced our first
advertising program through our direct sales force. We offered advertisers
the ability to place text based ads on our websites
targeted to our users search queries under a program called
(12:54):
Premium Sponsorships. Advertisers paid us based on the number of
times their ads were displayed on users search results pages,
and we recognized revenue at the time these ads appeared
end quote.
Speaker 2 (13:06):
So with this.
Speaker 1 (13:08):
Strategy, Google would collect a certain amount of money every
time someone saw a particular ad in their search results.
Clicking on the ad wasn't necessarily a requirement. Just getting
the ad in front of folks was enough to warrant
to the charge of the ad. As for how much
the advertisers were actually paying, I saw one site where
a client said their rate was around forty dollars per
(13:31):
one thousand impressions or views, so every thousand times this
ad was seen they would pay out forty dollars. And
Google was rapidly rising in popularity, and when you think
of all the different potential ads the company could sell
against different search terms, it really starts to add up.
(13:51):
The premium sponsorship ads would appear at the top of
the search results, so the very first search result would
actually be an AD, so there was a decent chance
that unobservant folks would just click on the ad, assuming
it was the best result for their query. Because it
was at the top of the search results page. Google
did have the foresight to place a shaded background behind
(14:13):
the link, which set it apart from the actual search results.
I suppose folks of the company understood that unless they
made the effort to differentiate the paid advertising spot from
the legitimate return results, they could get into some real
trouble there with regulators like the FTC, which absolutely demanded
that search engine companies make differentiation between sponsored results and
(14:38):
search results.
Speaker 2 (14:40):
All Right, we're gonna talk.
Speaker 1 (14:41):
More about how Google did that, plus other strategies Google
got into in the advertising space, but first we actually
need to take a break for our own advertisers.
Speaker 2 (14:51):
We'll be right back, okay.
Speaker 1 (15:02):
So when we broke off for ads, we were just
talking about how Google needed to separate their sponsored results
from their actual search results, and to follow along that
for a second.
Speaker 2 (15:16):
This is going to just be a little bit of
a tangent.
Speaker 1 (15:18):
Over the years, Google has changed up how it tags
or labels ads versus search results. So example, in two
thousand and seven, the company changed the background color from
being blue to yellow. In twenty ten, it went to
a very kind of light green background. Actually in the
example I saw it looked more teeled to me. But
then it went back to yellow. And going back to
(15:40):
yellow became kind of a thing. Over the following years,
Google would experiment and changed to different colors, but would
come back to yellow over and over again. In twenty thirteen,
Google went with a very pale yellow, which, depending upon
how your settings are on your mind or you may
not even be able to see, but that same year.
That was when the US Federal Trade Commission really issued
a war learning to search engines in general about a
(16:02):
quote decline in compliance end quote with those rules that
all search engines were supposed to follow, that they were
supposed to make it clear that certain entries were ads
versus legitimate search results. So Google would end up placing
an AD tag next to ads in search results to
make it more clear that they were in fact advertisings. However,
(16:24):
this AD tag actually appeared below the hyperlink text, so
you'd get the hyperlink text first, and then that little description,
that block of text and stuff that would describe the
link that appears below it. Next to that, you would
have this little AD tag. The company switched up how
that looked a few times. They would use different shaded
backgrounds on the tag and whatnot in an effort to
(16:45):
comply with the rules, while you know, not complying too hard,
because ideally you still get people clicking on those ads,
and it's easier to do that if the people don't
notice at first that it is an AD, which has
certainly happened to me. I know, I've been guilty of
it a few times. So I did a search just
now to see what looks like these days, And depending
(17:06):
on what I'm searching for, it may bring up different things.
Like I searched for a refrigerator, and in that case,
it brought up a bunch of sponsored links to where
I could buy a new refrigerator. But then I did
a different search and what I got were a couple
of search results at the top that had the label
sponsored appearing before the hyperlink. So you had a little
(17:29):
highlighted sponsored and then the hyperlink and then the text.
And I feel like that's fairly clear that it's an
advertisement as opposed to just a regular search result. So
props to Google in that case, because I think that's
a reasonable approach. It's not beating you over the head
with it, but I feel like it's noticeable. Anyway, Premium
(17:50):
sponsorships really started Google's ad revenue journey, but as mentioned,
Google's direct sales team was responsible at this time for
selling that adspace, and that means that the sales team
was reaching out to advertisers with this offer and trying
to convince them to agree to plunk down cash in
return for thousands of impressions per ad So it was
(18:12):
a strong start, but not nearly the strategy that would
propel Google into becoming one of the largest, most powerful
tech companies out there. Instead, that development would wait a
whole two or three months because that same year, in
two thousand, Google would not just introduce premium sponsorships, they
also introduced a little offering that at the time was
(18:33):
called AdWords. So unlike premium sponsorships, prospective advertisers could actually
sign up for AdWords at a link, you know, self
service style. They didn't have to be reached by a
direct sales representative. They would just go to a link
and say, okay, yes, I want to purchase some AdWords.
Speaker 2 (18:52):
And that wasn't the only.
Speaker 1 (18:52):
Difference between AdWords and premium sponsorships. Another was that adword
results appeared not at the top of search results. Instead,
it would appear to the right of search results. Originally
in what we used to call, or still call, I guess,
the right hand rail. I just don't hear much about
it anymore because I'm not in those meetings anymore, which
is fine by me. But the right hand rail is
(19:14):
you can think of that as like the right margin
of a web page. So in the old web banner days,
not that web banners are gone, but they certainly aren't
the dominant form of advertising anymore. In the old web
banner days, there were a few places where you would
typically put a web banner, and you might charge different
amounts depending upon the location. So the really popular one
(19:35):
was at the top of the page, right like at
the very top of the page you have a web
banner ad that advertises for something, But another would be
in the margins, either along the left or the right.
Very rarely you might put one at the bottom. Those
were often the least valuable ones, or at least the
least expensive ones, because the common thought was that people
just didn't scroll all the way to the bottom of
(19:56):
the page. But anyway, the right rail is where the
adword results originally would appear, So starting in August two
thousand or so, searching for something on Google would bring
up premium sponsorship ads at the top of your results
and a couple of AdWords ads set off to the right.
The AdWords ads also had a sponsored link label on them,
(20:17):
so they did indicate to the user that they were advertising.
They were not an official search result, so that label
was there from the beginning. From the advertiser side, buying
AdWords was pretty straightforward. The advertiser identified which keyword or
keywords they wanted, They then submitted the ad and bammed.
Their ad would show up wherever anyone searched for that
(20:39):
particular keyword. It was pretty simple stuff, and of course
they also paid a certain amount of money. They were
kind of doing a pay per click as opposed to
pay per impression, meaning that in this case, it wasn't
enough that someone saw the ad, they needed to have
clicked on it for it to count toward how much
Google would charge the advertiser. Things held steady for a
(21:00):
couple of years, but in two thousand and three, Google
made a decision that would shake things up. They decided
to phase out premium sponsorships and actually replace them with AdWords,
and the direct sales angle wouldn't be necessary anymore, at
least not for this, and AdWords would migrate from the
right hand rail to being above the search results. So essentially,
(21:24):
they were saying the AdWords approach seemed to be the
most successful, most lucrative, the one that was giving the
most satisfaction to advertisers as well as to Google, so
that was what they were going to switch to. They
were going to make that the dominant form of ad
revenue generation. In addition, adwards clients would receive charges based
on those number of clicks the ads got, Like I said,
(21:46):
not the number of impressions, which is nice, right, because
someone just seeing your ad that might not mean very much,
but if they click on your ad, that's a much
bigger deal. However, this also meant that Google would like
the cost of those ads. They wouldn't charge that forty
dollars per one thousand impressions. Now it'd be much higher now.
(22:08):
I'm not sure how much more Google was demanding, because
when I was looking at discussions of this online, no
one was really giving hard figures, but based on some
fairly obtuse articles from the time, it sounds like it
was a pretty significant increase, and I'm sure it more
than balanced out the shift from impressions to clicks. Also,
in two thousand and three, Google launched ad Sense. Now,
(22:32):
this product works within websites that belong on the Google network. So,
for example, a blog on Blogger, which Google also acquired
in two thousand and three, would count. It's part of
the Google network. So you create a blog on say Blogger,
your blog would belong to the Google network and you
could enroll in ad sense. And the way ad sense
(22:54):
works is that Google will place ads on websites and
those ads will be contextually relevant to whatever the content
of the site is itself. So let's say that you
write a blog about travel and it's a pretty popular blog. Well,
Google could serve up ads that relate to stuff like
travel agencies or airlines or hotels, and it would be
(23:18):
less likely to serve up ads about you know, I
don't know, underwear or kitchen appliances. Ad Sense was viewed
with some skepticism early on because a few advertisers voice
concerns that a consumer would be far more likely to
act on an AD if they're actively searching for information
through a web search rather than if they are just
casually visiting a website, Because, as it would turn out,
(23:41):
ad sense would end up making Google a hang on,
let me let me look up this figure. Yeah, a
butt load of money. So this was a really popular
offering once you get past the initial launch phase, and
content creators get a cut of that cash. By the way,
so choosing to monetize a site means the person creating
(24:04):
the content gets some of that money. Google gets some
of it and the advertisers get eyeballs. It's not that
different from the way advertising works on YouTube, which no
big surprise, you know Google owns YouTube as well, right Like,
you can choose to monetize your channel and you get
some of the money from the commercials that run on
(24:25):
your channel, and Google gets all the rest of it.
So very similar in that way. So Google was officially
in the ad business starting in two thousand, really ramping
up by the end of two thousand and three. Beginning
of two thousand and four. In two thousand and six,
it tried an experiment. Google made a proposal to newspapers
(24:46):
in the United States suggesting that Google would offer Internet
based clients the chance to bid on unsold newspaper ad space,
So the newspapers would be able to sell inventory that
otherwise would just go to way, and the ad clients
would potentially reach a new audience that traditionally they might
not have because they just weren't in the newspaper advertising business.
(25:10):
So around sixty six newspapers agreed to the pilot experiment.
Google would put up the ad space for each newspaper
and then open it up to bidding, and around one
hundred different advertisers would compete with one another to bid
on that ad space, and so the winning bid would
fork over the cash and then place whichever ad they
wanted in that slot. Google said the experiment was a
(25:32):
smashing success, but newspaper said, we're not really sure how
this really benefits us just yet. It seemed really good
for advertisers and their clients, It seemed great for Google,
but newspapers weren't totally sold on it. In fact, The
New York Times said that their plan was to forge
relationships with these smaller advertisers and ultimately to create a
(25:55):
direct relationship with them, eventually cutting Google out of the equation,
and that they were very upfront about this, like it
wasn't even on the slide. They're like, no, this was
our plan, We've told Google. Google was actually doing something
similar with radio stations. Like with the newspapers, the strategy
was to secure unsold ad inventory on radio stations and
(26:16):
then auction that space off to advertisers who otherwise wouldn't
have the opportunity to secure radio spots. And no big
surprise here. Google would then go on to try and
do the same thing with television. In two thousand and seven,
Google would secure an agreement with clear Channel Communications to
sell ads across their stations. And here's a fun fact.
(26:37):
Clear Channel Communications would eventually transform into iHeartMedia, So I
figured I should mention that since you know I worked
for iHeart podcasts. But the radio experiment would only last
for about three years because in two thousand and nine
Google pulled the plug on it, and the print advertising
experiment as well. The push to extend web advertising strees
(27:00):
to the quote unquote real World appeared to be mostly
a bust. It just didn't have a level that was sustainable.
The initial success was really promising, but it didn't last.
Fun side fact, audio ads would actually make a return
many many years later in twenty twenty, but this time
the audio ads were not designed.
Speaker 2 (27:22):
To play on the radio.
Speaker 1 (27:24):
Instead, they were designed to play against YouTube. The idea
was that more folks were listening to YouTube content while
not necessarily actively watching it. They might listen to it
while they do other tasks like cooking or cleaning, or
trying to find out exactly what is making that eerie
whaling sound beneath the floorboards. You hear it right anyway.
(27:48):
In twenty twenty two, Google would open up audio ads
to all advertisers who wanted to take advantage of listener
habits as they used YouTube to listen to stuff like
streaming music or podcasts. So audio ads are back. Maybe
they're just not the same thing as they were back
in two thousand and six. But speaking of that, let's
get back into the history of Google advertising. So in
(28:09):
two thousand and seven, Google would introduce pay per action ads.
So pay per click is a type of pay per action, right.
The action in that case is clicking on the ad.
The advertiser only has to pay when users take the
action of clicking on the ad itself. If they just
see it, there's no charge. However, that's not the only
(28:31):
type of action you could monetize. The action could be
something else. For example, maybe you set up an ad
deal where you will only pay if the user not
just clicks through, but signs up to say, subscribe to
a newsletter, or maybe you only pay if the user
will click through and make a purchase on the target site.
(28:52):
As you might imagine, the ad rates on those types
of ads can get pretty expensive because very few people
in general, like in the grand scheme of things, are
going to take those actions, but the ones who do
are going to be very valuable. So the AD rates
are going to be much higher for those kinds of actions,
but at the same time, they're not going to happen
as frequently as someone who just say clicks on an
(29:15):
AD or just looks in an AD. Google also experimented
with a tool that could display how many times users
search particular keywords. Obviously that's really important, right, Like, if
you're sitting there trying to determine which keywords you want
to target for your ads, it's helpful to know which
keywords are really popular versus which ones aren't. Let's say
(29:36):
you're an advertiser representing a sporting goods company, and so
you take a look to see which keywords are really trending,
and you're looking at phishing rod and you see a
lot of folks are searching for phishing rod, but very
few are actually searching for rod and real. So if
you're only going to target one of those sets of keywords,
(29:56):
you're probably going to want to target the phishing rod
one because those are the folks who are targeting more people.
More people are going to see her ad because that's
what most people are searching for You could argue that
if more people are serious about phishing and they're using
rod and Real, maybe you want to target that one instead. Yeah,
there are fewer people who are searching for it, but
(30:17):
the ones who are searching are really serious about getting
into fishing. There are a lot of decisions that had
to be made, but in order to make those decisions,
you need the analytical tools in the first place. Okay,
when we come back, we're going to talk about probably
the biggest change for Google's advertising strategy, or at least
the most impactful one.
Speaker 2 (30:38):
I would argue.
Speaker 1 (30:39):
But before we do that, let's take another break to
think our sponsors. Okay, I would argue that the biggest
move Google made in two thousand and seven was a
three point one billion with a B dollar acquisition of
(31:02):
a little company called double Click. So Dwight Merriman and
Kevin O'Connor came up with the idea for double Click
way back in the nineties, and it had its humble
beginnings as a web banner ad business. They figured that
the web would give them the chance to track ad
performance in a way that other media just couldn't do. Right,
Like you can buy ads for television or magazines or whatever,
(31:24):
but you're never really sure how much of your audience
actually saw or paid attention to the ad that you've
placed in front of them. With banner ads, you could
tell how many people actually took action, right, because you
could count how many people clicked on the ad. Plus
you would also still know how many impressions the ad
got whether anyone clicked on it or not, so you
(31:45):
would have that information too. The co creators met up
with a bunch of other folks who got involved, and
they merged their idea with another company that was kind
of struggling to do something similar, and they got to
work and double Click forged a path by using online
cookies to track user behaviors, which meant the company could
do better paid advertisements. They could pair ads with specific users.
(32:06):
So targeted advertising was starting to become a thing, and
it was really onto something. Double Click was really kind
of zeroing in on what the future of online advertising
was going to be. So another company called Hellman and
Friedman bought double Click in two thousand and five. They
spent a cool one point one billion dollars, and it
was from them that Google would purchase double Click in
(32:29):
two thousand and seven for the three point one billion dollars.
The FTC took a look at this deal to make
sure everything was copasetic, and they said, it doesn't look
like a problem to us.
Speaker 2 (32:40):
Now. I seriously wonder.
Speaker 1 (32:41):
If the deal were to happen today, if our current
FTC would have objected to it, because I bet they
would have. But that was not how the world turned
back in two thousand and seven. Two thousand and seven,
if you wanted to get a massive merger through chances
are the FTC would be like, Yeah, it's fine. These days,
they FTC is more like, eh, that looks a little
(33:02):
scary to me. So double click became part of the
Google Empire, and Google began to take advantage of double
clicks technologies, and that would really accelerate Google's ability to
leverage ads. In two thousand and eight, Google introduced a
new dashboard to allow advertisers to buy ad space across
different types of media. Keep in mind, this was still
(33:23):
when Google was in print and radio and TV before
they would pull the plug on all that, but at
this time they were really selling ad space through AdWords
across multiple platforms, whether that was the Web or newspapers
or magazines. All the magazine thing really only lasted a
brief time, or radio or television. And we know that
(33:45):
this would start to collapse in two thousand and nine
because that's when Google would shut down its print and
radio ad operations. But in two thousand and eight, the
plan was to create a kind of integrated user interface
for advertisers, to give them all the options in the
world to have their ad place wherever they felt would
work best. So it was a neat idea. It just
turned out that Google really had not nailed how to
(34:08):
integrate with these other pre existing forms of media. So
Google also opened up the opportunity for advertisers to use
third party advertising tags to track their ad performance across
Google's content network. So again, this would only apply to
ad Sense ads. You know, ad sets only works on
Google's network of content sites. They weren't selling ad Sense
(34:32):
for just generic sites that weren't part of Google's network.
But it meant that advertisers didn't have to rely solely
upon Google's own integrated internal analytics to see how their
ads were doing. So that was good, right because you know,
you worry when Google is the only source of information
(34:52):
about how well your ads are doing, because you don't
know if they're telling the truth. But then with these
third party tags, suddenly the advertisers were able to to
monitor ad performance independently of Google and make sure that
everything is going well, or if it's something's not going well,
that they can swap things out. In two thousand and nine,
while shutting down print and radio advertising projects, Google made
(35:14):
another acquisition. This time it was a company called ad
Mob or ad Mob as it might turn out to be,
and this specialized in ads on mobile platforms. Now, in
my opinion, this was an incredibly savvy move on Google's
part because just keep in mind, in two thousand and seven,
Apple proved that there was a market for a consumer smartphone,
(35:37):
particularly in the United States, they launched an iPhone. It
starts to sell incredibly well. Google follows suit by introducing
the Android operating system several months later, and then we
would see habits shift dramatically as people began to use
smartphones more frequently to access the web. So entire websites
were scrambling to adopt new mobile friendly strategies they never
(36:00):
had to worry about before. So it stands to reason
that Google wanted to leverage a company that specialized in
mobile advertising, rather than having to reinvent the wheel themselves,
they went and got a company that was already focused
on this sort of thing, and they could kind of
get ahead of this growing trend of the mobile web
becoming the dominant form of people accessing the web. Google
(36:23):
would continue to refine its ad tools over the years, so,
for example, in twenty eleven, the company gave advertisers the
chance to optimize which ads to display based on conversion
rates rather than through click through rates. So if you're
wondering what does that mean, Well, first, most advertisers will
have a whole bunch of ads for the same client, right.
(36:43):
They might have twenty or thirty different ads they're running
that are all about trying to move the same product
or service for the same client. But they have different images,
or different wording or different calls to action, but they're
all designed to try and get the same resist right
to sell.
Speaker 2 (37:01):
This product or service.
Speaker 1 (37:03):
Ideally, as an advertiser, you want to see which of
your ads are having the best impact, and when you
see that, you might want to start shifting to rely
more on that ad versus all the other ones you're
running for the same product or service, and you could
optimize it a couple different ways. If you optimize via
(37:23):
click through rate, what you're prioritizing are the ads that
get the most clicks on them. But click through isn't
the same thing as conversion. A conversion is when you
get a user to become a customer, So it's when
you convince your user to take a further action, such
as to make a purchase. The ads that lead to
conversions might be the ones you want to emphasize. You know,
(37:47):
maybe they don't get as many clicks as the highest
click through rate ones do, but the important thing is
the people who are clicking on that ad are the
ones who are buying what you're selling. So Google gave
advertisers the chance to automatically optimize based on whatever criteria
they found to be the most important for their clients.
So the advertisers could say, all right, well, if these
(38:08):
twenty ads are running and these three ads are getting
the most conversions, automatically, those three will get prioritized over
the other seventeen or whatever I think I said twenty.
So that would be the way that would kind of
work in a broad, high level sense. Now, I'm not
going to go over all the different features and changes
(38:30):
in Google's advertising business. That would require a mini series
of episodes, and I'm sure it would be interesting to
some listeners, but personally, I think I would find it
a little on the dull side to really go into
that deep detail. But we can hit a couple of
important points. For example, on July twenty fourth, twenty eighteen,
AdWords got a makeover, so Google rebranded AdWords and renamed
(38:54):
it Google Ads. So it's pretty much the same product,
just with a new name. If you're Cynicle, you could
say Google Ads is the product that lets an advertiser
buy the top spot in search results for particular keywords.
You just got to spend the money and suddenly you're
at the top of the search results. You do still
have that little tag that if people are paying attention,
(39:15):
they'll see it's an ad as opposed to the top
search result. But that same year, in twenty eighteen, Google
grouped its Google Analytics three sixty platform with most of
its double click properties and rename them the Google Marketing Platform,
the double Click Ad Exchange, and the Double Click for
Publishers products. Those became the Google Ad Manager so everything
(39:38):
was sort of getting rebranded and to new names and
collected together to have stuff that complemented each other become
a single platform. So let's finish up with a quick
talk about how much money all this advertising actually brings in. First,
keep in mind how Google has an incredible grip on
market share, particularly when it comes to stuff like online
(40:01):
video and online search. So I've pulled some stuff from
various statistic sites. Keep in mind everyone has their own
version of analytics, and there's probably a pretty wide range
of error margins here as well, but it gives us
an idea, right, So let's start with desktop search. Now
(40:22):
we're talking about laptops and like desktop computers. According to Statista,
Google search takes up more than eighty percent of the
market share for that, so eight out of ten web
search is being done on computers are using Google. That's
more than just dominant. But on mobile it gets even
(40:43):
more dramatic. Stat Counter estimates Google to have a ninety
five percent market share for mobile search. But even that
gets left behind when we look at online video like
streaming video. According to six Sense, YouTube has a ninety
seven point seven to one percent market share. The largest
(41:04):
competitor is JW Player, which has one point three five
percent market share, So no wonder the US government suspects
Google to have an anti competitive advantage. They do have
an insane market share in these areas. Keep in mind
that all of these are really just platforms upon which
(41:24):
Google can serve advertising. So all that market share means
that Google has ample opportunity to give ads to us.
And in twenty twenty three, the company made a whopping
two hundred thirty seven point eighty six billion with a
B dollars in AD revenue. That was about seventy seven
percent of all the revenue that the company generated last year.
(41:47):
That's a lot of princely sums, and it is why
I and lots of other people say that Google is
ultimately an advertising company, not a search company, search video.
These things are just methods by which Google can bring
advertising to us. And obviously Google's ad strategy goes well
(42:10):
beyond what I've covered today. As I said, to really
get into detail, we would have to do a full
mini series of it, but I thought it'd be interesting
to just kind of take a quick look and get
a surface level understanding of what's going on. It's also
a reminder that if the service you're using is free,
chances are you're the product, right. So like, if it's
(42:34):
free for you to use web search, well, the reason
it's free is because there's all this advertising that's playing
against that web search. Your behaviors are shaping what ads
are going to be sent to you, and the fact
that you are spending time using these services means the
services get to know more and more about you, which
means they can sell that information to advertisers or leverage
(42:57):
that information with advertisers in order to serve quote unquote
better ads to you. It just becomes this system in
which we, the users, are actually the products. So good
thing to recall, good thing to keep in mind. Doesn't
mean that you shouldn't use these things, but it's good
to be aware of it and not to just be
(43:19):
naive and think, oh, isn't it nice that Google offers
this service for free. Anyway, I hope you are all well.
I'm getting better every day, which is a relief, and I'll.
Speaker 2 (43:32):
Talk to you again really soon.
Speaker 1 (43:41):
Tech Stuff is an iHeartRadio production. For more podcasts from iHeartRadio,
visit the iHeartRadio app, Apple Podcasts, or wherever you listen
to your favorite shows,
Welcome to tech Stuff, a production from iHeartRadio. Hey there,
and welcome to tech Stuff. I'm your host, Jonathan Strickland.
I'm an executive producer with iHeart Podcasts and how the
tech are you. I've often said that Google is not
(00:25):
a search engine company. It's an advertising company. And of
course these days, when I say Google, you could argue
that I really mean Alpha, that you know, the parent
company that owns all the various Google companies. But to
be clear, I am not the only person who says this, right,
I don't mean to say that I'm some sort of
(00:45):
guru who I did not come up with it. I'm
not the first person to say it. I also say it.
That's all. I agree with those who say this. But
here's something that I find interesting. So, once upon a time,
Google was a search company. It was not an advertising company.
In the very early days, Google didn't generate revenue through advertising.
(01:08):
So let's set the clocks back quite a bit. In
nineteen ninety five, a guy named Larry Page was mulling
over the idea of enrolling in Stanford University in order
to go to grad school. He was at the campus
in order to get the lay of the land, and
there was a student who was assigned to him to
show him around, a Stanford student named Sarah gay Brinn,
(01:29):
and the story goes that the two got along kind
of like oil and water. They apparently didn't agree on
practically anything, but a year later, when Larry Page was
in the graduate program, the two decided that they would
work together on what would turn out to be a
truly huge project. So their goal was to build a
(01:52):
new kind of web search engine. So at the time,
the dominant search engines that were still working off the
very young Worldwide Web primarily used keywords as their guide.
So you would pop into a search engine. The one
I used all the time was web Crawler, but there
were lots of them, and you would type whatever it
(02:12):
was you were searching for, like let's say it's the
word tractor, for example. The search engine would return results
of web pages that featured that keyword somewhere in the page,
and they might have different ways of prioritizing the web pages.
Maybe if the keyword appears in the title of the page,
it might rank higher.
Speaker 2 (02:32):
That kind of thing.
Speaker 1 (02:33):
But the problem was there was never a guarantee that
the search results you got would actually be relevant to
the search query that you put in. They might include
the search term, but it may not turn out to
be useful. Maybe the page just mentions tractors off handedly,
like as a joke or as a reference, and it's
(02:55):
not actually about tractors, it just happens to have the
word tractors in it. Or maybe it's a web page
run by a real scumbag who just crams the bottom
of the page with a ton of random keywords, ones
that they've determined are really popular, and it's on an
effort to capture as much traffic from search engines as possible.
(03:16):
They don't care that this isn't what you were looking for.
They just want to get you onto that page. The
search experience could be really a frustrating one. It often
involved users going to a page, taking a quick look,
saying oh, this isn't what I need, backing out, and
then doing it all over again. And it wasn't very
(03:37):
satisfying or helpful page. And Brend wanted to create an
engine that would return much better search results.
Speaker 2 (03:44):
So how would you do that?
Speaker 1 (03:45):
Well, their solution was to take a broader approach to search.
Their search engine would assign a rank a page rank
if you will to each result, and higher ranks would
appear higher in the search results page and lots of
factors would go into determining what page rank a page
would receive. For example, let's say that you've got a
(04:07):
web page and you notice that there are tons of
incoming links pointing to this web page. Well, that could
end up being a boost to that page's rank, because
the thought is, if a lot of other web pages
are pointing to this page, it must be good, right,
if you've got a lot of different pages. Of course,
you can game that too, right, You could create a
whole network of pages that just link to each other
(04:29):
in an effort to try and boost page eranks. So
another part of this ranking system would take into account
the quality of those incoming links, the idea being well,
if the incoming links are coming from really established websites,
maybe they belong to a known entity like an important company,
(04:49):
or maybe it's a web page that has been around
for a while and has sort of proven itself, any
links that that page has are going to be worth more.
Plus those pages themselves, the ones that been around or
from known entities, those would have a higher rank because
they were thought of as just being higher quality. Now,
this is just a couple of examples of factors that
would determine a web page's page erank.
Speaker 2 (05:11):
As it turned out, there were.
Speaker 1 (05:12):
Lots of them, and the formula would change frequently, and
it was always kept secret so that pages couldn't try
to game the system. Although they didn't stop anyone from trying,
it just meant it was hard to do because you
didn't exactly know which factors were the most important. The
PageRank approach meant that you were far more likely as
(05:32):
a user to come across a search result that was
relevant to whatever it was you were actually searching for.
Speaker 2 (05:38):
Now.
Speaker 1 (05:39):
Originally, the pair called their search engine BackRub. I think
we could all agree that it was a fortunate thing
that they decided to change the name to Google. The
two received a healthy one hundred thousand dollars check from
an investor in nineteen ninety eight, and they launched Google Incorporated,
and for the first couple of years, Google existed primarily
(06:01):
off of influxes of cash coming from investors. Everyone could
recognize that Google was providing a valuable service, one that
was becoming increasingly important as the web was getting more
complex and larger, and it was doing so at a
level that really set Google apart from the competition. Google
was quickly making a name for itself as being the
(06:22):
best of the web search engines at that time, but
investors were also starting to get a little concerned, and
the reason for that was that the powerful Google search
engine didn't have an equally powerful way to generate revenue.
Speaker 2 (06:35):
Now.
Speaker 1 (06:35):
Initially, one strategy was to try and form partnerships with
portal sites. And these were websites that acted kind of
like a homepage, like this is the page that you
would see when you started up your browser. This was
kind of a gateway to the rest of the web.
They might host a few articles. Maybe some of those
articles were written in house by an editorial staff that
(06:57):
worked for the website. Maybe some of the articles were
really generated by a third party and they were just
being linked to on this landing page. And many of
them would also include a search tool. So Google would
try and make deals with these different sites to be
that search tool. But Google with small potatoes in those days,
(07:18):
so these deals were few and far between, and they
certainly weren't enough to make Google look like it was
going to be arousing success financially, So clearly Google would
have to come up with a different approach if the
company was to stay in business, and someone else had
to come up with an idea that, with a little tweaking,
would work just fine, and then Google would take that idea.
(07:42):
To say they stole it would probably be too harsh,
but they certainly emulated an idea that someone else came
up with for a totally different site. Now that's someone
else was a guy named William Bill T.
Speaker 2 (07:56):
Gross.
Speaker 1 (07:57):
He graduated from the California Institute of Technology with the
degree in mechanical engineering, which would not make you think
of someone who would go on to create a web
search engine, but he became known as something of a
serial entrepreneur. He would launch lots of different companies across
his career as soon as he was graduating, essentially, including
(08:17):
a video arcade and a couple of software companies, and
an audio equipment company and more. But in the late nineties,
Gross wanted to solve the same problem that Page and
Brinn had been working on or would be working on, actually,
because Gross's solution would come out first. He wanted to
be able to create a search engine that would return
better search results than a lot of the competitors that
(08:40):
were on the market. But he took a different approach
from what Brin and Page were looking at. So instead
of building out a system meant to assess the value
of every page that's on the Internet, he figured he
could deliver better search results by having companies pay him
in order to have their search result rank higher on
(09:00):
the list. So here's the basic idea. Now, as I
mentioned earlier, one of the big problems with early search
engines is that a lot of web administrators were trying
to game the system. So it really wasn't unusual for
web designer to just stuff a page full of keywords
that didn't actually apply to the page's content itself. So
(09:20):
clever designers would do this in a way that wasn't
obvious to the visitor, right because otherwise you would just
like I said, you would scroll down a page and
you see like a word jumble of a bunch of
unrelated words at the end of a page, and it
just looked like a mess. Clever web administrators would choose
to make the text the same color as the page's
(09:41):
background at the bottom, so you wouldn't even see it
like it might be that the page seems unusually long
with no content toward the bottom of it, but in
fact there would be content. It's just it was the
same color as the background, so you couldn't see it.
Speaker 2 (09:55):
However, while you.
Speaker 1 (09:57):
Couldn't see it, a search engine could. The search engine
would still see the keywords there even though you wouldn't,
and that's why this page that doesn't seem to be
related at all to the thing you were searching for
would pop up in search results. Meanwhile, you had legitimate
sites out there, some connected to established brands, that would
very much like to rank high in search in order
(10:19):
to drive more traffic to their sites, but with all
these junk sites out there, that could be hard to do.
So they might be in a position where they're able
to and willing to pay to jump to the head
of the line if it meant that folks would actually
find their stuff online. And so Gross launched a company
called go to dot com Incorporated. His strategy included a
(10:42):
pay per click arrangement, which is exactly what it sounds like.
The advertiser agrees to pay a certain amount every time
someone clicks on the ad, so typically you aggregate clicks
by the thousands, so each click is really just worth
you usually fractions of a penny. Gross's logic was that
the garbage sites out there couldn't and wouldn't pay for
(11:06):
positioning because their entire strategy relied on using keywords in
bulk that had no application to the actual site. So
if they wanted to try and do that the same
way with Gross's strategy, they would have to pay for
all these different keywords that didn't actually apply to them,
and it would bankrupt them because people would be clicking
on those sites over and over again on the search results,
(11:27):
and they'd have to pay every time that happened, or
at least every time every thousand times it happened. So
you would be paying go To for all this traffic.
You wouldn't be making that money back through the ad
revenue you generated on your site. There's no way you'd
be making more money through say, web advertising, than you
were paying out to go To. So whamo, you end
(11:48):
up with a search engine that delivers better results, or
at least more relevant results. Since you can't guarantee that
the entities that actually do pay for their positioning are
in fact the best ones out there, but chances are
there better than the junk sites. Gross launched goo to
dot com in nineteen ninety eight. In fact, he launched
it before Page and brend got Google off the ground.
And while go to only saw modest success in the
(12:10):
grand scheme of things, that revenue generation model stood out,
and when it came time to determine a way for
Google to make money, the go to became well, well,
it became the go to model. So Google would introduce
a pair of strategies in two thousand that would bring
(12:31):
the company into the advertising space. Now, according to a
Google SEC filing from two thousand and four, quote, in
the first quarter of two thousand, we introduced our first
advertising program through our direct sales force. We offered advertisers
the ability to place text based ads on our websites
targeted to our users search queries under a program called
(12:54):
Premium Sponsorships. Advertisers paid us based on the number of
times their ads were displayed on users search results pages,
and we recognized revenue at the time these ads appeared
end quote.
Speaker 2 (13:06):
So with this.
Speaker 1 (13:08):
Strategy, Google would collect a certain amount of money every
time someone saw a particular ad in their search results.
Clicking on the ad wasn't necessarily a requirement. Just getting
the ad in front of folks was enough to warrant
to the charge of the ad. As for how much
the advertisers were actually paying, I saw one site where
a client said their rate was around forty dollars per
(13:31):
one thousand impressions or views, so every thousand times this
ad was seen they would pay out forty dollars. And
Google was rapidly rising in popularity, and when you think
of all the different potential ads the company could sell
against different search terms, it really starts to add up.
(13:51):
The premium sponsorship ads would appear at the top of
the search results, so the very first search result would
actually be an AD, so there was a decent chance
that unobservant folks would just click on the ad, assuming
it was the best result for their query. Because it
was at the top of the search results page. Google
did have the foresight to place a shaded background behind
(14:13):
the link, which set it apart from the actual search results.
I suppose folks of the company understood that unless they
made the effort to differentiate the paid advertising spot from
the legitimate return results, they could get into some real
trouble there with regulators like the FTC, which absolutely demanded
that search engine companies make differentiation between sponsored results and
(14:38):
search results.
Speaker 2 (14:40):
All Right, we're gonna talk.
Speaker 1 (14:41):
More about how Google did that, plus other strategies Google
got into in the advertising space, but first we actually
need to take a break for our own advertisers.
Speaker 2 (14:51):
We'll be right back, okay.
Speaker 1 (15:02):
So when we broke off for ads, we were just
talking about how Google needed to separate their sponsored results
from their actual search results, and to follow along that
for a second.
Speaker 2 (15:16):
This is going to just be a little bit of
a tangent.
Speaker 1 (15:18):
Over the years, Google has changed up how it tags
or labels ads versus search results. So example, in two
thousand and seven, the company changed the background color from
being blue to yellow. In twenty ten, it went to
a very kind of light green background. Actually in the
example I saw it looked more teeled to me. But
then it went back to yellow. And going back to
(15:40):
yellow became kind of a thing. Over the following years,
Google would experiment and changed to different colors, but would
come back to yellow over and over again. In twenty thirteen,
Google went with a very pale yellow, which, depending upon
how your settings are on your mind or you may
not even be able to see, but that same year.
That was when the US Federal Trade Commission really issued
a war learning to search engines in general about a
(16:02):
quote decline in compliance end quote with those rules that
all search engines were supposed to follow, that they were
supposed to make it clear that certain entries were ads
versus legitimate search results. So Google would end up placing
an AD tag next to ads in search results to
make it more clear that they were in fact advertisings. However,
(16:24):
this AD tag actually appeared below the hyperlink text, so
you'd get the hyperlink text first, and then that little description,
that block of text and stuff that would describe the
link that appears below it. Next to that, you would
have this little AD tag. The company switched up how
that looked a few times. They would use different shaded
backgrounds on the tag and whatnot in an effort to
(16:45):
comply with the rules, while you know, not complying too hard,
because ideally you still get people clicking on those ads,
and it's easier to do that if the people don't
notice at first that it is an AD, which has
certainly happened to me. I know, I've been guilty of
it a few times. So I did a search just
now to see what looks like these days, And depending
(17:06):
on what I'm searching for, it may bring up different things.
Like I searched for a refrigerator, and in that case,
it brought up a bunch of sponsored links to where
I could buy a new refrigerator. But then I did
a different search and what I got were a couple
of search results at the top that had the label
sponsored appearing before the hyperlink. So you had a little
(17:29):
highlighted sponsored and then the hyperlink and then the text.
And I feel like that's fairly clear that it's an
advertisement as opposed to just a regular search result. So
props to Google in that case, because I think that's
a reasonable approach. It's not beating you over the head
with it, but I feel like it's noticeable. Anyway, Premium
(17:50):
sponsorships really started Google's ad revenue journey, but as mentioned,
Google's direct sales team was responsible at this time for
selling that adspace, and that means that the sales team
was reaching out to advertisers with this offer and trying
to convince them to agree to plunk down cash in
return for thousands of impressions per ad So it was
(18:12):
a strong start, but not nearly the strategy that would
propel Google into becoming one of the largest, most powerful
tech companies out there. Instead, that development would wait a
whole two or three months because that same year, in
two thousand, Google would not just introduce premium sponsorships, they
also introduced a little offering that at the time was
(18:33):
called AdWords. So unlike premium sponsorships, prospective advertisers could actually
sign up for AdWords at a link, you know, self
service style. They didn't have to be reached by a
direct sales representative. They would just go to a link
and say, okay, yes, I want to purchase some AdWords.
Speaker 2 (18:52):
And that wasn't the only.
Speaker 1 (18:52):
Difference between AdWords and premium sponsorships. Another was that adword
results appeared not at the top of search results. Instead,
it would appear to the right of search results. Originally
in what we used to call, or still call, I guess,
the right hand rail. I just don't hear much about
it anymore because I'm not in those meetings anymore, which
is fine by me. But the right hand rail is
(19:14):
you can think of that as like the right margin
of a web page. So in the old web banner days,
not that web banners are gone, but they certainly aren't
the dominant form of advertising anymore. In the old web
banner days, there were a few places where you would
typically put a web banner, and you might charge different
amounts depending upon the location. So the really popular one
(19:35):
was at the top of the page, right like at
the very top of the page you have a web
banner ad that advertises for something, But another would be
in the margins, either along the left or the right.
Very rarely you might put one at the bottom. Those
were often the least valuable ones, or at least the
least expensive ones, because the common thought was that people
just didn't scroll all the way to the bottom of
(19:56):
the page. But anyway, the right rail is where the
adword results originally would appear, So starting in August two
thousand or so, searching for something on Google would bring
up premium sponsorship ads at the top of your results
and a couple of AdWords ads set off to the right.
The AdWords ads also had a sponsored link label on them,
(20:17):
so they did indicate to the user that they were advertising.
They were not an official search result, so that label
was there from the beginning. From the advertiser side, buying
AdWords was pretty straightforward. The advertiser identified which keyword or
keywords they wanted, They then submitted the ad and bammed.
Their ad would show up wherever anyone searched for that
(20:39):
particular keyword. It was pretty simple stuff, and of course
they also paid a certain amount of money. They were
kind of doing a pay per click as opposed to
pay per impression, meaning that in this case, it wasn't
enough that someone saw the ad, they needed to have
clicked on it for it to count toward how much
Google would charge the advertiser. Things held steady for a
(21:00):
couple of years, but in two thousand and three, Google
made a decision that would shake things up. They decided
to phase out premium sponsorships and actually replace them with AdWords,
and the direct sales angle wouldn't be necessary anymore, at
least not for this, and AdWords would migrate from the
right hand rail to being above the search results. So essentially,
(21:24):
they were saying the AdWords approach seemed to be the
most successful, most lucrative, the one that was giving the
most satisfaction to advertisers as well as to Google, so
that was what they were going to switch to. They
were going to make that the dominant form of ad
revenue generation. In addition, adwards clients would receive charges based
on those number of clicks the ads got, Like I said,
(21:46):
not the number of impressions, which is nice, right, because
someone just seeing your ad that might not mean very much,
but if they click on your ad, that's a much
bigger deal. However, this also meant that Google would like
the cost of those ads. They wouldn't charge that forty
dollars per one thousand impressions. Now it'd be much higher now.
(22:08):
I'm not sure how much more Google was demanding, because
when I was looking at discussions of this online, no
one was really giving hard figures, but based on some
fairly obtuse articles from the time, it sounds like it
was a pretty significant increase, and I'm sure it more
than balanced out the shift from impressions to clicks. Also,
in two thousand and three, Google launched ad Sense. Now,
(22:32):
this product works within websites that belong on the Google network. So,
for example, a blog on Blogger, which Google also acquired
in two thousand and three, would count. It's part of
the Google network. So you create a blog on say Blogger,
your blog would belong to the Google network and you
could enroll in ad sense. And the way ad sense
(22:54):
works is that Google will place ads on websites and
those ads will be contextually relevant to whatever the content
of the site is itself. So let's say that you
write a blog about travel and it's a pretty popular blog. Well,
Google could serve up ads that relate to stuff like
travel agencies or airlines or hotels, and it would be
(23:18):
less likely to serve up ads about you know, I
don't know, underwear or kitchen appliances. Ad Sense was viewed
with some skepticism early on because a few advertisers voice
concerns that a consumer would be far more likely to
act on an AD if they're actively searching for information
through a web search rather than if they are just
casually visiting a website, Because, as it would turn out,
(23:41):
ad sense would end up making Google a hang on,
let me let me look up this figure. Yeah, a
butt load of money. So this was a really popular
offering once you get past the initial launch phase, and
content creators get a cut of that cash. By the way,
so choosing to monetize a site means the person creating
(24:04):
the content gets some of that money. Google gets some
of it and the advertisers get eyeballs. It's not that
different from the way advertising works on YouTube, which no
big surprise, you know Google owns YouTube as well, right Like,
you can choose to monetize your channel and you get
some of the money from the commercials that run on
(24:25):
your channel, and Google gets all the rest of it.
So very similar in that way. So Google was officially
in the ad business starting in two thousand, really ramping
up by the end of two thousand and three. Beginning
of two thousand and four. In two thousand and six,
it tried an experiment. Google made a proposal to newspapers
(24:46):
in the United States suggesting that Google would offer Internet
based clients the chance to bid on unsold newspaper ad space,
So the newspapers would be able to sell inventory that
otherwise would just go to way, and the ad clients
would potentially reach a new audience that traditionally they might
not have because they just weren't in the newspaper advertising business.
(25:10):
So around sixty six newspapers agreed to the pilot experiment.
Google would put up the ad space for each newspaper
and then open it up to bidding, and around one
hundred different advertisers would compete with one another to bid
on that ad space, and so the winning bid would
fork over the cash and then place whichever ad they
wanted in that slot. Google said the experiment was a
(25:32):
smashing success, but newspaper said, we're not really sure how
this really benefits us just yet. It seemed really good
for advertisers and their clients, It seemed great for Google,
but newspapers weren't totally sold on it. In fact, The
New York Times said that their plan was to forge
relationships with these smaller advertisers and ultimately to create a
(25:55):
direct relationship with them, eventually cutting Google out of the equation,
and that they were very upfront about this, like it
wasn't even on the slide. They're like, no, this was
our plan, We've told Google. Google was actually doing something
similar with radio stations. Like with the newspapers, the strategy
was to secure unsold ad inventory on radio stations and
(26:16):
then auction that space off to advertisers who otherwise wouldn't
have the opportunity to secure radio spots. And no big
surprise here. Google would then go on to try and
do the same thing with television. In two thousand and seven,
Google would secure an agreement with clear Channel Communications to
sell ads across their stations. And here's a fun fact.
(26:37):
Clear Channel Communications would eventually transform into iHeartMedia, So I
figured I should mention that since you know I worked
for iHeart podcasts. But the radio experiment would only last
for about three years because in two thousand and nine
Google pulled the plug on it, and the print advertising
experiment as well. The push to extend web advertising strees
(27:00):
to the quote unquote real World appeared to be mostly
a bust. It just didn't have a level that was sustainable.
The initial success was really promising, but it didn't last.
Fun side fact, audio ads would actually make a return
many many years later in twenty twenty, but this time
the audio ads were not designed.
Speaker 2 (27:22):
To play on the radio.
Speaker 1 (27:24):
Instead, they were designed to play against YouTube. The idea
was that more folks were listening to YouTube content while
not necessarily actively watching it. They might listen to it
while they do other tasks like cooking or cleaning, or
trying to find out exactly what is making that eerie
whaling sound beneath the floorboards. You hear it right anyway.
(27:48):
In twenty twenty two, Google would open up audio ads
to all advertisers who wanted to take advantage of listener
habits as they used YouTube to listen to stuff like
streaming music or podcasts. So audio ads are back. Maybe
they're just not the same thing as they were back
in two thousand and six. But speaking of that, let's
get back into the history of Google advertising. So in
(28:09):
two thousand and seven, Google would introduce pay per action ads.
So pay per click is a type of pay per action, right.
The action in that case is clicking on the ad.
The advertiser only has to pay when users take the
action of clicking on the ad itself. If they just
see it, there's no charge. However, that's not the only
(28:31):
type of action you could monetize. The action could be
something else. For example, maybe you set up an ad
deal where you will only pay if the user not
just clicks through, but signs up to say, subscribe to
a newsletter, or maybe you only pay if the user
will click through and make a purchase on the target site.
(28:52):
As you might imagine, the ad rates on those types
of ads can get pretty expensive because very few people
in general, like in the grand scheme of things, are
going to take those actions, but the ones who do
are going to be very valuable. So the AD rates
are going to be much higher for those kinds of actions,
but at the same time, they're not going to happen
as frequently as someone who just say clicks on an
(29:15):
AD or just looks in an AD. Google also experimented
with a tool that could display how many times users
search particular keywords. Obviously that's really important, right, Like, if
you're sitting there trying to determine which keywords you want
to target for your ads, it's helpful to know which
keywords are really popular versus which ones aren't. Let's say
(29:36):
you're an advertiser representing a sporting goods company, and so
you take a look to see which keywords are really trending,
and you're looking at phishing rod and you see a
lot of folks are searching for phishing rod, but very
few are actually searching for rod and real. So if
you're only going to target one of those sets of keywords,
(29:56):
you're probably going to want to target the phishing rod
one because those are the folks who are targeting more people.
More people are going to see her ad because that's
what most people are searching for You could argue that
if more people are serious about phishing and they're using
rod and Real, maybe you want to target that one instead. Yeah,
there are fewer people who are searching for it, but
(30:17):
the ones who are searching are really serious about getting
into fishing. There are a lot of decisions that had
to be made, but in order to make those decisions,
you need the analytical tools in the first place. Okay,
when we come back, we're going to talk about probably
the biggest change for Google's advertising strategy, or at least
the most impactful one.
Speaker 2 (30:38):
I would argue.
Speaker 1 (30:39):
But before we do that, let's take another break to
think our sponsors. Okay, I would argue that the biggest
move Google made in two thousand and seven was a
three point one billion with a B dollar acquisition of
(31:02):
a little company called double Click. So Dwight Merriman and
Kevin O'Connor came up with the idea for double Click
way back in the nineties, and it had its humble
beginnings as a web banner ad business. They figured that
the web would give them the chance to track ad
performance in a way that other media just couldn't do. Right,
Like you can buy ads for television or magazines or whatever,
(31:24):
but you're never really sure how much of your audience
actually saw or paid attention to the ad that you've
placed in front of them. With banner ads, you could
tell how many people actually took action, right, because you
could count how many people clicked on the ad. Plus
you would also still know how many impressions the ad
got whether anyone clicked on it or not, so you
(31:45):
would have that information too. The co creators met up
with a bunch of other folks who got involved, and
they merged their idea with another company that was kind
of struggling to do something similar, and they got to
work and double Click forged a path by using online
cookies to track user behaviors, which meant the company could
do better paid advertisements. They could pair ads with specific users.
(32:06):
So targeted advertising was starting to become a thing, and
it was really onto something. Double Click was really kind
of zeroing in on what the future of online advertising
was going to be. So another company called Hellman and
Friedman bought double Click in two thousand and five. They
spent a cool one point one billion dollars, and it
was from them that Google would purchase double Click in
(32:29):
two thousand and seven for the three point one billion dollars.
The FTC took a look at this deal to make
sure everything was copasetic, and they said, it doesn't look
like a problem to us.
Speaker 2 (32:40):
Now. I seriously wonder.
Speaker 1 (32:41):
If the deal were to happen today, if our current
FTC would have objected to it, because I bet they
would have. But that was not how the world turned
back in two thousand and seven. Two thousand and seven,
if you wanted to get a massive merger through chances
are the FTC would be like, Yeah, it's fine. These days,
they FTC is more like, eh, that looks a little
(33:02):
scary to me. So double click became part of the
Google Empire, and Google began to take advantage of double
clicks technologies, and that would really accelerate Google's ability to
leverage ads. In two thousand and eight, Google introduced a
new dashboard to allow advertisers to buy ad space across
different types of media. Keep in mind, this was still
(33:23):
when Google was in print and radio and TV before
they would pull the plug on all that, but at
this time they were really selling ad space through AdWords
across multiple platforms, whether that was the Web or newspapers
or magazines. All the magazine thing really only lasted a
brief time, or radio or television. And we know that
(33:45):
this would start to collapse in two thousand and nine
because that's when Google would shut down its print and
radio ad operations. But in two thousand and eight, the
plan was to create a kind of integrated user interface
for advertisers, to give them all the options in the
world to have their ad place wherever they felt would
work best. So it was a neat idea. It just
turned out that Google really had not nailed how to
(34:08):
integrate with these other pre existing forms of media. So
Google also opened up the opportunity for advertisers to use
third party advertising tags to track their ad performance across
Google's content network. So again, this would only apply to
ad Sense ads. You know, ad sets only works on
Google's network of content sites. They weren't selling ad Sense
(34:32):
for just generic sites that weren't part of Google's network.
But it meant that advertisers didn't have to rely solely
upon Google's own integrated internal analytics to see how their
ads were doing. So that was good, right because you know,
you worry when Google is the only source of information
(34:52):
about how well your ads are doing, because you don't
know if they're telling the truth. But then with these
third party tags, suddenly the advertisers were able to to
monitor ad performance independently of Google and make sure that
everything is going well, or if it's something's not going well,
that they can swap things out. In two thousand and nine,
while shutting down print and radio advertising projects, Google made
(35:14):
another acquisition. This time it was a company called ad
Mob or ad Mob as it might turn out to be,
and this specialized in ads on mobile platforms. Now, in
my opinion, this was an incredibly savvy move on Google's
part because just keep in mind, in two thousand and seven,
Apple proved that there was a market for a consumer smartphone,
(35:37):
particularly in the United States, they launched an iPhone. It
starts to sell incredibly well. Google follows suit by introducing
the Android operating system several months later, and then we
would see habits shift dramatically as people began to use
smartphones more frequently to access the web. So entire websites
were scrambling to adopt new mobile friendly strategies they never
(36:00):
had to worry about before. So it stands to reason
that Google wanted to leverage a company that specialized in
mobile advertising, rather than having to reinvent the wheel themselves,
they went and got a company that was already focused
on this sort of thing, and they could kind of
get ahead of this growing trend of the mobile web
becoming the dominant form of people accessing the web. Google
(36:23):
would continue to refine its ad tools over the years, so,
for example, in twenty eleven, the company gave advertisers the
chance to optimize which ads to display based on conversion
rates rather than through click through rates. So if you're
wondering what does that mean, Well, first, most advertisers will
have a whole bunch of ads for the same client, right.
(36:43):
They might have twenty or thirty different ads they're running
that are all about trying to move the same product
or service for the same client. But they have different images,
or different wording or different calls to action, but they're
all designed to try and get the same resist right
to sell.
Speaker 2 (37:01):
This product or service.
Speaker 1 (37:03):
Ideally, as an advertiser, you want to see which of
your ads are having the best impact, and when you
see that, you might want to start shifting to rely
more on that ad versus all the other ones you're
running for the same product or service, and you could
optimize it a couple different ways. If you optimize via
(37:23):
click through rate, what you're prioritizing are the ads that
get the most clicks on them. But click through isn't
the same thing as conversion. A conversion is when you
get a user to become a customer, So it's when
you convince your user to take a further action, such
as to make a purchase. The ads that lead to
conversions might be the ones you want to emphasize. You know,
(37:47):
maybe they don't get as many clicks as the highest
click through rate ones do, but the important thing is
the people who are clicking on that ad are the
ones who are buying what you're selling. So Google gave
advertisers the chance to automatically optimize based on whatever criteria
they found to be the most important for their clients.
So the advertisers could say, all right, well, if these
(38:08):
twenty ads are running and these three ads are getting
the most conversions, automatically, those three will get prioritized over
the other seventeen or whatever I think I said twenty.
So that would be the way that would kind of
work in a broad, high level sense. Now, I'm not
going to go over all the different features and changes
(38:30):
in Google's advertising business. That would require a mini series
of episodes, and I'm sure it would be interesting to
some listeners, but personally, I think I would find it
a little on the dull side to really go into
that deep detail. But we can hit a couple of
important points. For example, on July twenty fourth, twenty eighteen,
AdWords got a makeover, so Google rebranded AdWords and renamed
(38:54):
it Google Ads. So it's pretty much the same product,
just with a new name. If you're Cynicle, you could
say Google Ads is the product that lets an advertiser
buy the top spot in search results for particular keywords.
You just got to spend the money and suddenly you're
at the top of the search results. You do still
have that little tag that if people are paying attention,
(39:15):
they'll see it's an ad as opposed to the top
search result. But that same year, in twenty eighteen, Google
grouped its Google Analytics three sixty platform with most of
its double click properties and rename them the Google Marketing Platform,
the double Click Ad Exchange, and the Double Click for
Publishers products. Those became the Google Ad Manager so everything
(39:38):
was sort of getting rebranded and to new names and
collected together to have stuff that complemented each other become
a single platform. So let's finish up with a quick
talk about how much money all this advertising actually brings in. First,
keep in mind how Google has an incredible grip on
market share, particularly when it comes to stuff like online
(40:01):
video and online search. So I've pulled some stuff from
various statistic sites. Keep in mind everyone has their own
version of analytics, and there's probably a pretty wide range
of error margins here as well, but it gives us
an idea, right, So let's start with desktop search. Now
(40:22):
we're talking about laptops and like desktop computers. According to Statista,
Google search takes up more than eighty percent of the
market share for that, so eight out of ten web
search is being done on computers are using Google. That's
more than just dominant. But on mobile it gets even
(40:43):
more dramatic. Stat Counter estimates Google to have a ninety
five percent market share for mobile search. But even that
gets left behind when we look at online video like
streaming video. According to six Sense, YouTube has a ninety
seven point seven to one percent market share. The largest
(41:04):
competitor is JW Player, which has one point three five
percent market share, So no wonder the US government suspects
Google to have an anti competitive advantage. They do have
an insane market share in these areas. Keep in mind
that all of these are really just platforms upon which
(41:24):
Google can serve advertising. So all that market share means
that Google has ample opportunity to give ads to us.
And in twenty twenty three, the company made a whopping
two hundred thirty seven point eighty six billion with a
B dollars in AD revenue. That was about seventy seven
percent of all the revenue that the company generated last year.
(41:47):
That's a lot of princely sums, and it is why
I and lots of other people say that Google is
ultimately an advertising company, not a search company, search video.
These things are just methods by which Google can bring
advertising to us. And obviously Google's ad strategy goes well
(42:10):
beyond what I've covered today. As I said, to really
get into detail, we would have to do a full
mini series of it, but I thought it'd be interesting
to just kind of take a quick look and get
a surface level understanding of what's going on. It's also
a reminder that if the service you're using is free,
chances are you're the product, right. So like, if it's
(42:34):
free for you to use web search, well, the reason
it's free is because there's all this advertising that's playing
against that web search. Your behaviors are shaping what ads
are going to be sent to you, and the fact
that you are spending time using these services means the
services get to know more and more about you, which
means they can sell that information to advertisers or leverage
(42:57):
that information with advertisers in order to serve quote unquote
better ads to you. It just becomes this system in
which we, the users, are actually the products. So good
thing to recall, good thing to keep in mind. Doesn't
mean that you shouldn't use these things, but it's good
to be aware of it and not to just be
(43:19):
naive and think, oh, isn't it nice that Google offers
this service for free. Anyway, I hope you are all well.
I'm getting better every day, which is a relief, and I'll.
Speaker 2 (43:32):
Talk to you again really soon.
Speaker 1 (43:41):
Tech Stuff is an iHeartRadio production. For more podcasts from iHeartRadio,
visit the iHeartRadio app, Apple Podcasts, or wherever you listen
to your favorite shows,
TechStuff News
Host
Jonathan Strickland
Popular Podcasts
1. The Podium
The Podium: An NBC Olympic and Paralympic podcast. Join us for insider coverage during the intense competition at the 2024 Paris Olympic and Paralympic Games. In the run-up to the Opening Ceremony, we’ll bring you deep into the stories and events that have you know and those you'll be hard-pressed to forget.
2. In The Village
In The Village will take you into the most exclusive areas of the 2024 Paris Olympic Games to explore the daily life of athletes, complete with all the funny, mundane and unexpected things you learn off the field of play. Join Elizabeth Beisel as she sits down with Olympians each day in Paris.
3. iHeartOlympics: The Latest
Listen to the latest news from the 2024 Olympics.