This course provides a comprehensive introduction to programmatic advertising, explaining its fundamental concepts, ecosystem components, terminology, challenges, and various buying methods. It emphasizes the growing importance and efficiency of programmatic advertising for marketers and advertisers.
Mind Map
点击展开
点击探索完整互动思维导图
Hello and welcome to this course on programmatic
programmatic
advertising. My name is Rohit Otam
Chandani. I'm a digital marketing
consultant, speaker and trainer and I'm
really excited to be your trainer for this
this
course. I've been in the digital
marketing space for about 6 years now.
Worked with multiple clients across
sectors. Help them craft their digital
strategies, scale their online presence.
I've managed millions of dollars in
advertising spends across various
digital advertising channels and
platforms. I have a strong passion to
speak. So I go around speaking at
digital marketing conferences and events
across the globe. I also have a strong
passion to train and over the last 4
years I've trained over 4,000
professionals including entrepreneurs
and senior level marketing executives in
digital marketing and strategy. So now
that you know a little bit about who I
am, let's look at what we will cover in this
this
course. We'll begin with an
understanding of what programmatic
advertising really is. I'll then take
you through the fundamentals of the
digital advertising auction for those of
you who are not familiar with it. We
then look at the different components of
the programmatic advertising ecosystem.
We look at demand side platforms, supply
side platforms, ad networks, ad
exchanges, what they are and where they
fit into the ecosystem. I'll then take
you through some terminology that you
will keep hearing in the programmatic
advertising space. We look at frequency
cap, above the fold and below the fold
inventory, run on network and run on
site types of ad buying. We'll look at
the most common challenges that are
faced by advertisers in the programmatic
space. We'll understand brand safety,
advertising fraud, and the concept of
viewability. We'll then move on to
understanding the different types of
programmatic advertising. We look at
open auctions, private auctions,
preferred deals, and programmatic
guaranteed deals as well. I'll also tell
you what are the pros and cons of each
type, when you should pick which type of
programmatic advertising. I'll also take
you through the Google programmatic
ecosystem, all the components that are
part of it, and what are the functions
of each component. We look at the
difference between programmatic
advertising and standard display, what
are the advantages and disadvantages of
each and when you should pick one over
the other. And then we'll finish with
some success stories from brands in the
programmatic advertising space that will
show you the true power of programmatic
advertising. So now that you know what
we're going to be talking about, let me
take you through why this course will be
really important and useful to you,
especially if you're a marketer or an
advertiser. Well, programmatic
advertising is growing at a rapid pace.
If you look at the Zenith media
programmatic marketing forecast from
2018, about 65% of digital media spends
in 2019 will be programmatic. In fact,
in the United States, by the end of
2019, nearly 84% of digital display ad
spends will transact
programmatically. Programmatic is
becoming an increasingly important part
of the digital advertising ecosystem.
And as a marketer or an advertiser, it
is imperative that you fully understand
how it works. It's an in- demand skill
that will help you take your career to
the next
level. So, what are we waiting for?
Let's get started. I'm really excited to
take you around the world of
programmatic advertising and I look
forward to seeing you in the course.
Hello and welcome to the first lesson of
the programmatic advertising course. I
hope you're as excited about this course
as I am. In this lesson, we look at what
programmatic advertising really
is. It may seem like this complex fancy
term that people throw around in
conversations to sound a little more
intelligent, but in actuality,
programmatic advertising is nothing but
using technology to buy and sell media.
It's just a way of automating the media
buying process.
Programmatic advertising is marketing
data and technology working together to
make ad buying systems and processes
more efficient. The singular objective
of programmatic advertising is to help
advertisers reach the right audience at
the right time in the right context and
with the right message. As an
advertiser, if you can do this, if you
can ensure that you're reaching each
person you want to reach at the moment,
that's optimal. And with a message
that's personalized for them, your
advertising ROI would be at its best.
And that's exactly what technology,
data, and marketing are working together
here to do to increase the efficiency of
your digital
advertising and maximize the return
investment or ROI you get from it. So
that's in a nutshell what programmatic
advertising really is. and we'll look at
the ecosystem and how it works in more
detail as we go through this course.
We'll begin with understanding the
fundamentals of the digital advertising
auction especially for those of you who
are not familiar with how the auction
works and that's what the next video is
all about. So I look forward to seeing
you in that one. Welcome back to the
course. A good part of digital
advertising is bought using auctions.
And in this lesson, we'll understand the
fundamentals of how digital advertising
options really work. Now, some of you
who already have some knowledge or
experience in digital marketing might
find this lesson a little basic. You
might already know some of this stuff.
In that case, you can either skip this
lesson to the end, or you can skim
through it and watch it at twice the
speed to get a quick refresher of the
topic. Great. So let's get started and
understand how the auction in digital
advertising actually works. We'll first
look at the Google Ads auction and see
how does Google decide which ad to show
at which position. So let's go do a
Google search. So let's say I want to
buy a life insurance plan. So I've just
gone to Google and searched for the
keyword buy life insurance online. And
as you can see, I have four ads that
have shown up here on the top. And as I
scroll below the organic
results, I have three ads here at the
bottom as
well. So how does Google
decide which ads should show at what
position? Well, this happens through the
auction. So each advertiser who wants to
show up for this particular keyword here
would go to their Google Ads dashboards
and bid on this
keyword. So let's assume that those four
advertisers are advertisers A, B, C, and
D. And they all want to show their ads
for that specific keyword. So they would
go to that Google Ads dashboard and they
would enter how much they're willing to
bid for that query. Now, every auction
has a base price. Even in real life
auctions, for those of you who have been
to real life auctions or taken part in
general auctions where products are
being sold, there is always a base
price. And here as well, there is a base
price for this keyword. So, let's assume
that this base price, which is set
before the bidding begins, is a $5 cost
per click. So, for every click on that
keyword, the minimum that an advertiser
would bid would be $5. How is this base
price decided? Well, it depends on multiple
multiple
factors. Most importantly, the
competition in the auction. It depends
on how many advertisers are going after
that keyword and how much are they
willing to pay for it. The more
advertisers in the auction and the more
they are willing to pay for it, the
higher the base price would be. So,
let's say these four advertisers have
paid $7, $8,
$5.50, and $10.
So in this case, if the bid was the only
thing that was taken into account, it's
quite obvious that advertiser D would
win the auction and their ad would show
right on top. But the Google Ads auction
does not take just the bid into account
because in that case advertisers with
deep pockets could buy out smaller
advertisers and users would not get the
best experience as well. In order to
ensure that users get the best
experience and users get the most
relevant ad to show up for their query,
there's another factor that goes into
the auction and that factor is called a
quality score. Now, what is a quality
score? A quality score is a score on a
scale of 1 to 10, one being the lowest
and 10 being the highest that is given
to advertisers for every keyword that
they are bidding on. How is this quality
score calculated? Well, it takes three
factors into account. It takes the
expected clickthrough rate on an ad, the
relevance of the ad to the query being
searched for, and the experience that
the user would have on the landing page.
So going back to our example, let's say
advertiser A has a quality score of
three, advertiser B has a quality score
of four, advertiser C has a quality
score of 8 and advertiser D has a
quality score of two. Now in this case,
the position of the ad or something
called the ad rank is decided based on
both these factors. So ad rank is a
combination of the bit that the
advertiser has put into the system and
the quality score that has been assigned
to their keywords in this case based on
the three factors that we just
discussed. So advertisers who put in a
lot more effort to give the user a
better experience by making their ads
more relevant by making their landing
page experiences better would have a
higher quality score. So in this case
here, let's assume the ad rank is a
simple multiplier between the bid and
the quality score. Though in actuality,
it's quite complex. But for now, for the
sake of simplicity, let's assume that
it's a simple multiplier. The ad ranks
would then be 21, 32, 44, and 20. Going
by the ad rank, the auction would now be
won by advertiser C. So, an interesting
point for you to note here is that
although advertisers see had bid the
lowest, they still ended up winning the
auction. And that's because, like I said
earlier, they put in the effort to make
their ads more relevant and their
landing pages better from a user experience
experience
perspective. Let's take another example.
This time, let's look at a banner ad
being sold through a programmatic
advertising auction. Let's say you've
loaded a site on your laptop. And this
site has an ad spot there where ads can
possibly show. Now there are multiple
advertisers there who would want to
target you with their products and
services. So they would take part in a
similar auction to get their ads to
show. For this auction, just like any
other auction, let's assume that the
base price set before the auction begins
is a $2 CPM, which is $2 per,000
impressions. And let's say there are
four advertisers A, B, C, and D who have
their banner ads ready and they want and
they want these ads to be shown to you
in that ad spot. In order to enter the
auction, they would now have to bid over
that base price, which is $2 for,000
impressions. Let's say advertiser A has bid
bid
$2.20. Advertiser B has bid
$4.80. C has bid $2.80. 80 cents and D
has bid $35. Now, in this case, who do
you think is going to win the auction?
Well, it's pretty obvious. Advertiser B,
who's bid the highest, would win the
auction here, and his ad would end up
showing in that ad spot to you. Now, the
question is, he's won the auction, but
how much would the advertiser end up
paying? Would he pay
$4.80? Would he pay more? Would he pay
the base
price? How much would he end up paying?
The bid that he has set is $4.80, but
how much would he end up paying now that
the ad has been
shown? Well, the answer
is he would end up paying
$3.51. That's because all digital
advertising auctions, including the
Google auction that you just saw, are
second price auctions. When I say second
price auction, it means the winning
bidder would pay a price which is one
notch above the second highest bidder.
In this case here, the second highest
bidder is advertiser D who bid
$3.50 and advertiser B after winning the
auction would pay one notch higher which
is 1 cent higher. And that's how it
comes up to
$3.51. Like I said, all digital
advertising auctions are second price
auctions where the winning advertiser
pays a price which is one notch higher
than the second highest
advertiser. Now this process happens in
real time. It happens every time you do
a search on Google or every time you
load up a site. The auction is taking
place in the background at a speed which
is faster than the time it takes for you
to blink your eye. That's how tech
intensive the ecosystem
is. So if I go back to my Google search
here again and refresh the page, you
would see these positions would most
likely change. So if you know the
positions now, at position one, we have
this advertiser called policy bazar. At
position two, we have an advertiser
called ICICI. Position three is Alaba
Capital and four is Max Life. So I'm now
going to refresh this
page and you saw the positions
change. So position one which was
earlier Policy Bazar has now been taken
by Alia Capital and the other three
positions have changed too. That's
because the auctions are happening in
real time. Every time you refresh a
page, all these advertisers go back to
the auction. The entire process takes
place. Bids are checked. In this
particular case, the quality scores are
checked. The ad ranks are recalculated
and the ads then show up. The same thing
holds good when you visit a site on the
internet. So, I've just opened up the
Washington Post and you see an ad here.
This is an ad that you see on the side.
If I refresh this
page. So I just refreshed the page and
once I refreshed all advertisers who
wanted to bid for that particular spot
would have gone into the auction again
and you can see the ad has now
changed. Now some of you might have a
question on the Google ads auction which
is well advertiser see won the auction
and got their ad to show right on top
but how much would they end up paying?
Well the fundamentals is the same. This
is also a second price auction. In this
case, the algorithm takes the ad rank
and the quality score into account and
calculates a normalized bid. And with
the normalized bid, the winning bidder
would pay one notch higher than the
second highest
bid. So, we've seen a couple of examples
and I hope you guys are now clear with
the concept of how digital advertising
auctions work. So let's recap some of
the key points that we just
learned. The first thing is just
advertising auctions are second price
auctions. So the winning advertiser pays
one notch higher than what the second
highest advertiser had
bid. They are blind auctions. So
advertisers don't know what other
advertisers are bidding. So in the
examples that I showed you, advertiser A
did not know what advertiser B is
bidding. B did not know what C is
bidding and C did not know what D was
bidding. When it comes to programmatic
advertising, the auctions are completely
automated. So in this case that you saw
right from the bids to how much should
be bid is all done through the system.
The bids would be changed in real time
as well. So when you refresh pages, the
bids that you see here for each
advertiser would not remain the same.
the system would figure out the best
match for that user who's loaded the
page and alter the bids accordingly. And
last is that these auctions happen in
real time. Every time you refresh a
page, all advertisers go back to the
auction and the system automatically
finds the best ad to
show. Great. So, that's the fundamentals
of how digital advertising auctions
work. I hope you guys had a great time
learning and I'll see you guys in the
next one. Welcome back. Before we dive
deeper into programmatic advertising,
let's go back to the beginning of time
in digital advertising and see what it
looked like then. So, let's say about 15
years ago, you were an advertiser. Let's
take an example. Let's say you were
BMW and you wanted to advertise on the
internet to audiences that were there.
You made a list of publisher sites that
you want to show your ads on. Publishers
are websites that have ad space to sell.
Let's say one of those publishers that
you wanted to advertise on was the
Washington Post. How do you think you
would have gone about it? Well, the only
way to do it was to get in touch with
them directly. So, you would speak to
the people in charge of ad sales at the
Washington Post and tell them that you
wanted to run a banner ad for BMW on
their site. They would revert to you
with a proposal that would have
different options to advertise on their
site with the prices for each option.
You would then negotiate with them, try
and get the best options at the best
price. You would close the deal,
finalize the period for which the ad
would show. Then a lot of things on the
operational front would happen manually.
You would have the ad being sent across,
tracking tanks being shared. It was
basically a lot of manual work. Right
from the stage where you got in touch to
closing the deal to executing it to
measuring the impact, everything was
being done manually. Apart from all the
manual work that went in, what's worse
is that you would end up showing your ad
to everyone on the Washington Post site
without being able to segment it to an
audience that you would want to go
after. This resulted in a lot of wasted
impressions and a lot of time and effort
that was being spent by teams on the
entire execution
process. From this scenario to where
digital advertising is today, it's
evolved a fair bit and that's what we'll
talk about in the next lesson. See you
in that one. Hello and welcome
back. Let's dive deeper into the world
of programmatic advertising.
In the previous lesson, we talked about
what digital advertising looked like in
the early
days. Well, today it has evolved from
being something as simple as this in the
beginning to something as complex as
this today with thousands of players in
the ecosystem between the advertisers
and the publishers helping match the
right ad impression to the right user.
their agencies, demand side platforms,
supply side platforms, ad networks, ad
exchanges, a whole lot of stuff. So
let's understand where these fit into
the ecosystem. So on one side you have
advertisers, people like us who want to
show our ads to our target audience,
consumers at the other end. In order to
reach consumers, we'll have to reach
them through the sites and apps that
they use online, which is basically what
we call
publishers. Now, as an advertiser, when
you want to buy ad inventory, you would
most likely hire an agency that would
help you buy and manage your ads across
multiple platforms. Now, the left side
of this chart is called the demand side
because this is where there's a demand
for ad impressions.
And the right side of the chuck is
called the supply side because that's
where the supply of the impressions are.
Now agencies do not directly buy from
publishers in the programmatic scenario.
They buy through a platform which is
called a demand site platform or a DSP.
Now, a DSP is nothing but a tool or
software that helps advertisers or their
agencies buy ad inventory that is in
sync with their marketing
objectives. The simplest and most basic
example of a DSP would be the Google Ads
platform itself. Because if you think
about it, that's where advertisers or
agencies go and buy ad impressions, be
it search ads, display or video advertising.
advertising.
A more advanced example of a DSP which
has a lot more features and options for
advertisers is something called
doubleclick bed manager or
DBM. Now these names have all changed
since July 2018 when Google rebranded
all its advertising platforms. So for
some of you who are familiar that's when
Google Adwords became Google ads and all
double click products had their names
changed. So Doubleclick is also owned by
Google and DoubleClick bid manager which
is DBM is today called Google Display
and Video 360 or just Google DV 360.
Don't worry about getting confused
between the old and the new names. In a
later section, I have some time
allocated to getting you familiarized
with these platforms, what their old
names were and what their new names are.
You'd need to know both because these
are used interchangeably in the industry
even today. So that's the demand side.
You have an advertiser who might hire an
agency to buy their ad impressions. The
agency would then interact with the
demand side platform or DSP which is
nothing but a tool or a software that
helps agencies and advertisers buy
relevant ad impressions. For smaller
advertisers who do not want to hire an
agency and pay the agency fees and
commissions, they could go directly to
DSPs as well. So you have advertisers
who use the Google Ads platform directly
for example.
Now on the supply side, just like you
have a demand side platform for
advertisers, you have a supply side
platform or an SSP for
publishers. So what does a supply side
platform do? Well, just like what a DSP
does, it helps publishers sell their
inventory to advertisers in the most
efficient manner possible.
When I say efficiency, what I mean is it
helps them liquidate most of their ad
inventory at the best price
possible. Again, the simplest example of
an SSP would be Google AdSense, which is
where publishers can go and sign up to
have their ad inventory sold through
Google. So let's say you're a publisher,
you have a blog or a website, you want
to monetize it by showing ads. You could
just go to Google AdSense, sign up, and
then Google would start showing ads on
your site and pay you a percentage of
the revenue that they make. Just like
Google Ads works as a great platform for
small to midsize advertisers and also to
an extent for large
advertisers, Google AdSense works as a
great platform for small to midsize
publishers. For publishers who are
really large, they would need a lot more
advanced functionality in the supply
side platform to help them liquidate the
massive inventory that they would have
at the best price. So an example of a
more advanced supply side platform, one
that is used by a large number of
publishers across the globe is double
click for publishers or DFP. Again, this
name has changed as well and today it's
called the Google Ad Manager. Like I
said, don't worry about the old and the
new names. We'll come back to it at a
later point. Now, the demand side and
the supply side interact with each other
through something called an ad network
or an ad exchange. Let's look at ad
networks first. Ad networks are
basically companies that have a bunch of
publishers partnered with them to show
ads on their site.
A classic example of an ad network is
the Google Display Network or the GDN.
As most of you might be aware, the GDN
is nothing but a network of sites and
apps that have partnered with Google to
show Google ads on them. So, Google owns
the network and shows ads on these publisher
publisher
sites. Publishers then make a percentage
of the revenue. So these ads are sold by
publishers through Google AdSense and
bought by advertisers through Google
ads. Another example of an ad network
which is quite popular is the Facebook
audience network or the
FAN. So these are ad networks which like
I said they're companies that have
partnered with a whole bunch of
publishers to show ads on those
publisher sites and apps.
The other way in which the demand side
and supply side interact is through
something called an ad
exchange. Ad exchanges tend to be much
larger than ad
networks. You could have one ad exchange
interacting simultaneously with multiple
ad networks.
Ad exchanges are also more transparent
when it comes to their options and the
pricing models because with ad networks
the prices are determined individually
by the company that owns it. With ad
exchanges, it's more real time and
decided by market dynamics through the
auction. A classic example of an ad
exchange is the doubleclick ad exchange
or adex as it's abbreviated.
This is one of the largest ad exchanges
globally. And when we come to the Google
programmatic ecosystem in a later
section of this course, we'll understand
the inventory that the doubleclick
exchange has and the other ad networks
and ad exchanges it has partnered with.
Great. So that's the programmatic
advertising ecosystem for you. Now, one
thing I want to clarify for you guys
here as advertisers and marketers is
that in order to operate a system
efficiently, you don't really need to
know the nitty-gritty aspects of how the
systems work in the back end. To give
you an analogy here, I'm sure most of
you can drive a car. But if I were to
ask you to tell me how the engine really
works, most of you wouldn't have clarity
on that aspect. But I'm sure you can
drive really well. The same thing
applies to the programmatic ecosystem as
well. The technical side of it can be
quite complex, but as advertisers and
marketers, you don't really need to
understand that side in order to be able
to operate it efficiently. All you need
to know is have a clear idea of what
these components are, what do they do,
and where they fit into the ecosystem.
And that's what we've done by looking at
this flowchart here. I hope you now have
more clarity on the basic functioning of
the programmatic advertising ecosystem.
In later sections, we'll go on to
understand some more terms that are
spoken in the programmatic advertising
space and we'll also look at the
different types of programmatic
advertising out there and how you can
use them as
advertisers. Thank you for being part of
the lesson. I hope you had a great time
learning and I will see you in the next
one. Welcome back. In this section, we
talk about some of the most common terms
that you would hear in the programmatic
advertising space. But before we get
there, a quick recap of what we've
covered so far. We've looked at what
programmatic advertising really is.
We've gone through the fundamentals of
the digital advertising auction. We've
seen how digital advertising has evolved
from the beginning of time to what it is
today. And we've also gone through the
entire programmatic advertising
ecosystem, seen how it works, understood
some of the most important components.
We've seen DSPs, SSPs, ad networks, ad
exchanges, where they fit into the
ecosystem and what function do they
perform. We now move on to our first
term in programmatic advertising,
something called a frequency cap. A
frequency cap is also used across
digital advertising, not just in the
programmatic space. But there is one
advantage when it comes to programmatic
advertising, which we'll talk about at
the end of this lesson. Before we go to
frequency capping, let's look at what
frequency is. This is something some of
you who are familiar with dish
advertising would already know. So you
could consider this as a quick
refresher. And for those of you who are
new, it's an important topic that you
must pay attention
to. Frequency is basically the average
number of times that your ad has been
shown to a unique user. Right? So if
your ad is being shown five times to a
single person, the frequency is five. If
it's been shown eight times to a single
person, the frequency is eight. So how
is this calculated when you're running
ad campaigns reaching a large number of
people where frequency which is
generally calculated for a specific time
frame? So it could be the frequency over
a week, over a month, it could be
frequency over a day. This is basically the
the
impressions that have been delivered in
that time frame divided by reach in the
same period. A quick refresher,
impressions are the number of times that
your ad has been shown to users and
reach is the number of people or unique
users who your ad has been shown to.
Impressions are always greater than
reach. You will never have a scenario
where impressions are less than reach.
So you take a couple of examples here.
Let's say that you ran a campaign, the
impressions were
5,000 and you reach 2,500 people. So in
this case, the frequency is two. That's
the average frequency. And let's assume
this is all over a week. So that's a
frequency of two over the last week
where your ad campaign got you 5,000
impressions and you reach 2,500 people.
If the impressions are 3,000 and the
reach is 3,000 as well, that means your
frequency is one. This means you reached
everyone in the audience just
once. Likewise, another example. Let's
say your campaign delivered 6,000
impressions and reached 1,000 people. In
this case, on an average, each person
was delivered the ad six times, right?
So, that's how frequency is
calculated. So, what's the significance
of this
metric? Well, it's something that's
really important when you're managing
campaigns. If you set your frequency too
high, you're going to be overexposing
your audience, right? This will result
in sub-optimal spending of your
marketing budget because beyond a point,
you're not really going to be
influencing your audience anymore. And
what's worse is you could even annoy
them. Nobody likes seeing the same ad
too many times. And if you overexpose
them, your ads could end up having
detrimental effect as well. And as an
advertiser, you wouldn't want that to
happen. You would want to show the ad to
your users an optimal number of times,
which is just enough to drive a recall
for you, right? Because if you show your
ads infrequently or where the frequency
is too low, then the opposite would
happen. You would underexpose your
audience and this would not drive strong
enough recall and visibility for you.
Which is where, like I said, the
frequency needs to be optimal and that's
where a frequency cap comes in. Setting
a frequency cap helps you limit your
exposure to whatever you want, right?
So, how do you figure out what this
optimal frequency is? Well, the best way
to find the optimal frequency is to test
different frequency values over a
certain time period against your desired
marketing outcome. So, the outcome could
be a lift in brand recall, a lift in
sales. You would test different
frequencies and see if you're meeting
your end objective. So, are you getting
the conversions you want? Are you able
to generate the brand recall that you
want? That's how you would go about it.
A lot of people ask me if there is a
fixed frequency that they should go
after or is there a simple chart to
follow saying that well I should set a
frequency cap to five per week or two
per day or 25 in a month. Well, honestly
it varies widely from brand to brand,
business to business and marketing
objective that you want to go after. So
the best way honestly to go about it is
to test. When it comes to testing, I
have a great framework here for you that
you can keep in mind when you're testing
different frequencies. It's a great
frequency factors chart which has been
published by Facebook ads. It's one of
the most clear and concise frameworks
that I've seen when it comes to planning
your frequency cap. So, this chart gives
you multiple factors to look at when
you're deciding or testing your
frequency caps. The first bunch of
factors is what we call market factors.
So, let's say you're an established
brand, a brand that's been around for
ages. You've got a large market share.
Uh, you got a high share of voice. In
this case, you would rather use a low
frequency because you already have a
fair bit of brand recall and you don't
want to overexpose your audience or
waste your budget by reaching people who
already remember you well enough. On the
other hand, let's say you're a new
brand. You've just started out. You
don't have the kind of share of voice
that a bigger brand has. In this case,
you would need to test higher frequency
caps for the same time period. So,
market factors are one. The other is
what we call message factors. So, think
about your ad messaging. If your message
is unique and not very complex, fairly
easy to understand, then a lower
exposure would do because if people see
the message once or twice, they would
get it. On the other side, if you got an
ad messaging that's that's new, that's
not really unique, that's quite complex
to understand, you might need to plan a
higher cap because people would take
time to figure out your message and then
remember what you're trying to say. The
last one is media factors. So, you look
at your campaign duration, for example.
Is it a campaign that you're planning to
run for a long, long time? Are you going
to run this campaign for the next 3
months with similar messaging? Then you
might want to keep your caps lower
because in this case over time the
exposure would be high. So when the
weekly cap would be low, if you look at
the overall exposure over the 3 month
duration of your campaign, it would be
enough for you to drive the right amount
of recall. Likewise, if you're using
multiple channels, so let's say you're
doing ads across different publishers,
different networks, then you might want
to have a lower frequency across all
because you're going to have multiple
touch points with your audience and
you're going to be exposing your ads to
them across different media. On the
other hand, if your campaign duration is
short, let's say it's a short burst
campaign, you've got a launch and you're
doing a campaign just for 3 days or you
got an offer and you're running a
campaign just for those 3 days prior to
the offer, then you would use higher
frequency caps, right? So, this is a
great framework to use when you're
planning your frequency caps. Keep these
in mind and test your frequency. Find
what works best for specific brands and
specific scenarios. Now, when it comes
to programmatic advertising, a big
advantage that advertisers have is you
can use something called a global
frequency cap, which you can set across
publishers and networks. With normal
display advertising, you would not be
able to do this. Let's say you're
running ads on on Google Adwords, which
is now Google Ads, or on Facebook and a
bunch of other platforms. You can set
individual frequencies on each platform,
right? You can control your frequency
separately on Google, separately on
Facebook, separately on let's say Yahoo
or on LinkedIn, but you can't really
control the frequency or the exposure
globally. With programmatic advertising,
it makes your life much easier as an
advertiser because now you can control
frequency across platforms. So you can
work with multiple publishers, multiple
ad networks and set a global frequency
cap saying that well the moment my
frequency cap of let's say three is hit
I want my ads to stop showing to that
user across all
networks. So that's frequency cap. Like
I said it's a common term that you hear
in the programmatic space also use it
quite often and it's important to
understand what it is. How do you go
about setting the right frequency caps
and what is the advantage with frequency
capping when it comes to programmatic
advertising? Great. So that's all about
frequency cap. Hope you had a fun time
learning this concept and I will see you
in the next one. We're talking about
programmatic advertising terminology.
And in this lesson, we look at first,
second, and third party data. These are
common terms that you will hear across
the programmatic advertising industry
and you must know what they are. Let's
start with first party data. So what is
first party data? It is basically data
that a brand owns. So what does that
mean? Let's look at some examples.
Customer databases are an example of
first party data. So your customer email
ids, mobile numbers that you've
collected over time. This is data that
you own. You know how you've collected
them. you know where it came from.
Another example is data from your CRM
software. This could be your customer
purchase data. Which customers have
higher transaction volumes? How
frequently do they make a purchase? All
of this is again data that you own and
it's first party data. Data of your app
users is another example. People who use
your app, people who use your app less
frequently, people who use it more
frequently, people who make purchases in
the app. Another example is people who
visited your website and you're able to
track those visitors and remarket to
them later. You can even use this data
to build audiences similar to your
existing customers. Another example of
first party data is people who watched
your video. Let's say all the people who
watched your video on YouTube, you own
the data and you could target them with
another video. All the people who
watched a certain video on Facebook, you
could target them with another video. So
you own all this data. It's been created
through your marketing activities,
through the various activities that your
business does. This is extremely
valuable data and as a brand, it's
really important that you capture every
data point that you can. Of course, at
the same time, ensuring that you comply
with data privacy regulations.
One of the most powerful ways of using
first Py data apart from just
remarketing their existing customers or
reaching out to them and re-engaging
them is something called lookalike
modeling or like I said earlier using
the capabilities in different
advertising platforms to build audiences
that are similar to your existing
customer. These similar audiences can
then be very effective in helping you
achieve your business objective. So
that's first party data. Let's look at
the second type which is second party
data. So what is second party
data? It is data that is shared by a
partner. Let me explain this with an
example. Let's take an airline here, any
airline that has a frequent flyer
program and wants to promote it. And
let's take a bank that wants to promote
their travel credit card. Now the target
audience for these two companies would
have some degree of overlap. So they
could then agree to do some kind of a
cross promotion where the airline would
tell the bank that they would advertise
to existing customers of the travel
guide card and the bank would in turn
advertise to existing members of the
frequent flyer program. So this type of
data sharing is called second buy data.
When you partner with someone and use
their data for your marketing and
advertising purposes, it's called second
buy data. Here you know where the data
came from. You know it's from a credible
source and you can trust it. The third
type of data is something called third
party data. This is basically data
collected by a third party just as the
name indicates. Let's look at some
examples of this to understand what this
really is. Let's say you purchased a
database from a vendor. Now you have no
clue how the vendor collected that data.
You have very little information on the
quality of the data. So this is an
example of data that's been collected by
a third party and you don't really have
a clear idea of how the data was
collected. Another example of third
party data is your Google ads audience
data. So when you run display ads on the
Google ads platform, you have various
audiences that you can target. For
example, you can target audiences who
are interested in a certain topic,
audiences who have an affinity for a
certain category, audiences who are in
the market for a certain product or
service. Now again in this case these
are audiences that have been built by
Google based on the data that Google has
collected. Google does give you some
idea on how they've collected the data
but you don't have a clear picture of
how those audiences are actually built.
This is another example of third party
data. Only difference here is because
Google is an extremely credible player
in the online advertising ecosystem. you
would trust that data much more than you
would trust data from let's say
something you purchased from a random
vendor. Another example of third party
data is data collected by something
called DMPS or data management
platforms. So data management platforms
are companies that collect data from
multiple sources, put it together,
process it, segment it, and then package
it into specific buckets that can be
bought by advertisers. Now on
programmatic advertising platforms, you
have access to buy third party data from
data management platforms. You can't do
this with Google Ads. In fact, this is
something we'll also talk about in a
later lesson. The key advantage of
programmatic advertising is that you
have access to a lot of third party
data. Now, not every third party data
set might be reliable or credible, but
it's up to you to ensure you target
third party data sets from more reliable
data management platforms. Some of the
more reliable data management platforms
are BlueAI which is owned by Oracle,
Lot, Adobe Audience Manager, and the
Google Audience Center 360. So when you
buy data from DMPS, ensure the DMP is
credible and you can trust that the
third party data would be worth
targeting in campaigns that you're
running. Using credible third party
audience data can help you reach out to
newer audiences through program
platforms who you might not be able to
really reach
otherwise. Great. So we've seen the
three different types of data. We've
looked at first, second, and third party
data. Understood what they are and
looked at examples of each. I hope
you're now clear with this concept and I
look forward to seeing you in the next
lesson. We've been discussing
programmatic advertising terminology.
And in this lesson, we talk about four
terms, ATF and VTF and RO and
ROS. Now, there's this old joke that
programmatic advertising is full of what
we call TLAs or threeletter acronyms.
You'll keep seeing a lot of these in the
programmatic space. So there's DSPs,
SSPS, RO, ROS, ATF, BDF, there's RTB,
DBM, DMP, DFP, DCM. Well, sounds like a
mouthful and it might get a little
overwhelming when you're just starting
on programmatic, but don't worry. By the
end of this course, you'll be more
comfortable with all these TLAs. And as
you get into the programmatic space
yourself, after a point, you'll just get
very comfortable with all of them. And
these are just a few that you see on the
screen. There are a lot more that we'll
keep talking about as we go along. I
will also include a resource for you
which would be a ready recorder of all
these TLAs in one place towards the end
of the
course. So let's look at ATF and BTF
first. These are common terms that you
would see when you're in the
programmatic space. You'd see these in
proposals. You'd see these in reports.
So let's look at what this really means.
Let's look at this publisher site here.
Now, above the fold or ATF, it refers to
the area on the screen that you see when
you load a site. So, when you first load
a site, the area that you see on the
screen here, this is what we call above
the fold. And all the ads that would
show in this area is called above the
fold inventory or ATF inventory. So in
this case, the ad that you're seeing
here on the top left corner from
HostGator and this ad that you're seeing
here for this live webinar are above the
fold ads. Likewise with this ad here
which says become a
subbroker. So these ads are all above
the fold. As I scroll
below, there's this ad here from
LinkedIn which says introducing video
ads which I'm getting targeted for
because it's relevant to me. So this ad
was just at the fold and then as you
keep scrolling
below you'll see more ads as well. So
let's keep scrolling here. You see this
ad here. So this ad is below the fold.
In fact this is me getting remarketed
for something I searched for. If you
guys remember in the session on dish
advertising auction fundamentals, I
searched for life insurance plans and I
accidentally happened to click on one of
those links at that point and now I'm
getting retargeted or remarketed here.
So this is an ad that's below the fold.
Likewise, this ad that you see here,
this is also a below the fold ad. So
when you're buying ad
inventory, inventory that is above the
fold and just at or below the fold is
inventory that's considered much more
valuable. Inventory that is below the
fold and the more and more below the
fold you go, the less valuable the
inventory becomes. So that's something
to keep in mind when you're buying
inventory is to check whether it's above
the fold or below the fold. And if you
specifically want above the fold
inventory, you can mention that while
buying your ads as well. Moving on to RO
and ROS which are types of ad buying.
ROS or runoff site is an ad buying
option where your ad placements would
appear on any pages of a target site. So
on the site that you just saw which was
moneycontrol.com. If I buy ads from them
on an ROS model, that means my ads could
show on any page of their site and any
position on those pages, right? I don't
control which pages I want to show my
ads on. If I want to show my ad only on
their homepage, then I would choose a
different ad buying option. I would not
choose an ROS option. RO on the other
hand is runoff network. This is an ad
buying option in which your ad
placements would appear on a wide
collection of sites in a specific ad
network without having the ability to
choose specific sites. So to give you an
example, let's say you're running ads on
the Google Display Network and you're
targeting a specific audience. Let's say
you're targeting an audience aged 25 to
45 in California who is in the market
for a new car. Your ads would show to
that audience anywhere and everywhere on
the network where the audience is
present. Likewise with different ad
networks, they might give you an option
saying you could run ads across our
network and reach specific audiences
without having to worry about choosing
specific sites. So these are different
options. You would choose these options
depending on the objective of the
campaign that you're running at that
point. In some cases, you might choose
RO. In some cases, you might choose ROS.
In some cases, you might not choose any
of these. And you might choose to
advertise only on specific areas of a
specific site. So, that's the end of
this lesson. I hope you're now clear on
what above the fold and below the fold
is and what runoff site and runoff
network types of ad buying are. I look
forward to seeing you in the next one.
Hello and welcome to this lesson on
brand safety.
Brand safety has been a key concern for
marketers and advertisers
today, especially when it comes to programmatic
programmatic
advertising. So, what really is brand
safety? Well, brand safety is ensuring
that an online ad does not appear in a
context that could potentially damage
the advertiser's brand. In a recent
study done by the integral ad science,
which is one of the most credible
companies in the programmatic
advertising space, they found that over
40% of display ad impressions on
desktops served next to violent content
and about 18% of impressions served next
to adult content. Now, as brands, you
don't want your ads to show in
environments that are not congenial to
your brand. These are not just limited
to violent nut content, but there's a
whole list of toxic content categories
that a brand would not want to associate
itself with. It could be content around
military conflict, obscenity, illegal
drugs, adult content, content around
weaponry and arms, hate speech, crime,
terrorism, and other spammy and harmful sites.
sites.
In the recent past, especially over the
last couple of years, there's been a
spurt in the number of reported
incidents of brand ads being shown next
to content which brands would not want
themselves to be associated with. Just
Google for brand safety examples and
you'll find a whole bunch of them. I've
just pulled up some to show you which
are safe to show in a course. So, this
is an example of a brand ad showing on a
YouTube video from Combat 18, which is a
violent pronabsi group. Here's another
example. This is from Walmart where we
have Walmart's ad showing on a video of
this organization called the Animal
Liberation Front, which is an extremist
organization that goes to the extent of
killing humans to protect animals.
Another problem with ads like these,
apart from the fact that a brand would
not want to associate themselves with
this content, is that to a lot of users,
it might appear that Walmart is
endorsing this group. And that's not
something the brand would want. Not just
that, since Walmart is paying YouTube
for this ad and YouTube is going to
share that revenue with the video on
which the ad was displayed indirectly,
Walmart is also helping the extremist
group to monetize. And most brands, like
I said, would not want to do this. You
can find a whole lot of examples. This
is another one. This is an ad for this
brand called Sandals, which is a luxury
holiday operator appearing on a video
that is promoting
jihadis. Another one. This is an ad for
the Guardian in a content environment,
which is not the most ideal. So brand
safety is a serious concern for
marketers and of late YouTube has been
under the scanner for a lot of this
video content on the platform where
brands ads have been showing. In fact
some big advertisers in the United
States like on AT&T and Johnson and
Johnson also pull their ads from YouTube
for a while over the content environment
that their ads were showing in. These
are all news articles that have been
published in leading online publishers
and I will share the links to these
articles in the resources section of
this lesson. So you can check them out
and read them yourselves as well. This
is another article which is from N 2017
where in the UK brands found that their
ads were showing on video sexualizing
kids and this caused a good number of
brands including IDAS and Dodgebank to
pull out as well. This was reported in
the times in the UK too.
Well, some of you might be wondering why
isn't YouTube pulling down this content
in the first place. Well, YouTube is
doing its best to do that, but it's not
always possible at the scale at which
YouTube is to police every video that
comes in. And especially when some
videos are subjective, because what
might seem extremist to me might not
seem extremist to somebody else. In
cases where the content is subjective,
policing it becomes all the more
difficult. It's much easier to remove
content that's explicit, adult content,
content that has nudity, because that
can even be done through advanced
artificial intelligence algorithms. But
content that is, like I said, subjective
becomes quite difficult to police.
Though, YouTube is doing its best to at
least not help these channels monetize.
In fact, in response to these brand
safety concerns that advertisers had
raised, YouTube had changed its
monetization guidelines for publishers
on the platform where starting last
year, publishers would need to have at
least,000 subscribers on their channel
and 4,000 hours of watch time a year.
It's not just YouTube, but as recent as
about a couple of weeks back,
advertisers found their ads on Instagram
showing against content that was
connected to or promoting suicide. This
was reported across the media as well,
and I'll share this article too in the
resources section so you can go and take
a look. Apart from these cases, there
could also be very subtle cases where
brands would not want the ads to show.
For example, it could be a news site
where there's a report of a car crash
and next to that you have an ad for a
Honda car. That's not going to be ideal.
You don't want to see an ad to buy a car
on a page that's talking about a car
crash. Now, brand safety is a major
problem, especially with programmatic
buys and open auction programmatic buys.
The reason I'm explaining this concept
now is because when we move on to types
of programmatic advertising in a later
session, I will talk about the open
auction type and because the open
auction type ars advertisers to go after
audiences anywhere on the internet
irrespective of what site they are,
brand safety becomes a major problem. In
the lesson where we talk about the pros
and cons of each type of programmatic
advertising and we talk about the
disadvantages of the open auction brand
safety is a key disadvantage there. So
I'll come back to this in that lesson
and talk about this challenge. But for
now it's important for you to understand
the concept of brand safety and why it
has become a key concern for advertisers
today. In order to help advertisers
tackle this challenge, there are some
platforms and tools out there that help
verify brand safety, help in verifying
if your ads showed in the right places.
Of course, this would be after the fact,
but at least you would know and you can
take corrective action for the future.
So, they also categorize sites based on
their brand safety risk and you could
choose to avoid sites where the risk is
high or moderate. So, these are two
tools that you can use. So I'd suggest
you go to their websites, check them
out, see what they do. The first one is
double verify. The second one is
integral ad science. So that brings us
to the end of this lesson on brand
safety. Hope you're now clear with that
concept and I will see you guys in the
next video. Welcome back. Apart from
brand safety which we covered in the
last lesson, another important topic
that you need to understand is ad fraud
and bot traffic. Sometimes also called
NHT. So that's another threeletter
acronym that you need to remember. NHT
is non-human traffic. It's also called
IVT sometimes which is invalid traffic.
So you can remember both those terms.
NHT which is non-human traffic and IVT
which is invalid traffic. So let's begin
with what is ad fraud in the first
place? Well, ad fraud causes advertisers
to pay for fake worthless traffic. So
where does this fake worthless traffic
come from?
It comes from bots. And as most of you
might be aware, bots are computer
programs that mimic human actions on the
internet. Now, most of you would be
surprised to know that over 50% of
internet traffic is bots. And this is
from the Encapsula Bot Traffic Reports
2016, which analyze data from over 16.5
billion visits to about 100,000 randomly
selected domains on their
network. There are multiple reports that
have been published with similar stats
where it states anywhere between 40 to
50% of web traffic is bots. Now, if you
look at that chart there, you've got
something called good bots and bad bots.
So what are these good
bots? Good bots are bots that make the
internet a better place. So Google
search engine bot for example. So how do
you think Google shows you the best
results from websites across the
internet? Do you think they have
somebody sitting and reading all that
stuff out there? Well, it's all done by
their bots which go around crawling
every website on the internet indexing
it and then giving you the best results.
So, similar to Google search engine bot,
there are a whole bunch of bots out
there that help making our lives on the
internet easier. The remaining bots or
the bad bots are basically all those
impersonators, scrapers, spammers, and
hackers that try and gather information
for harmful reasons. So, these are the
bots that are responsible for all that
online fraud that happens. If you look
at the numbers in 2017 about $14
billion was lost to ad fraud and it's
estimated that by 2022 about $44 billion
of advertiser money would be lost to the
ad fraud industry. Now that's massive
and it's a big problem at
hand. Again, if you look at ad fraud,
it's being covered in the mainstream
media as well. This is an article from
CNBC talking about the amount of money
being wasted on fake websites and
fraudulent ad networks. In fact, just a
couple of months back, the United States
and the UK have joined up to tackle ad
fraud, which they see growing into a $50
billion problem. But a lot of you might
be wondering, how does this ad fraud thing work? What exactly happens? So,
thing work? What exactly happens? So, I'm going to show you how this really
I'm going to show you how this really works and how does it
works and how does it happen. So, the first step in creating
happen. So, the first step in creating ad fraud is turning on a fake website.
ad fraud is turning on a fake website. So, what these hackers and spamsters do
So, what these hackers and spamsters do is create websites that are fake. They
is create websites that are fake. They then get their websites listed on ad
then get their websites listed on ad exchanges and networks so that their
exchanges and networks so that their websites would not be eligible to show
websites would not be eligible to show ads. Note that these ad exchanges would
ads. Note that these ad exchanges would be smaller ones and not the large ones
be smaller ones and not the large ones because the larger and the credible ones
because the larger and the credible ones have a whole long list of guidelines
have a whole long list of guidelines before they take on a site. For example,
before they take on a site. For example, if you want to join the Google Display
if you want to join the Google Display Network, you have to meet their site
Network, you have to meet their site monetization guidelines. So these
monetization guidelines. So these spamsters would go on and join smaller
spamsters would go on and join smaller ad networks and ad exchanges and there
ad networks and ad exchanges and there are thousands of them out there
are thousands of them out there especially in the programmatic
especially in the programmatic ecosystem. So once they've created a
ecosystem. So once they've created a fake website and joined an ad network or
fake website and joined an ad network or an ad exchange, they now get to load ads
an ad exchange, they now get to load ads on their site. Once ads start showing on
on their site. Once ads start showing on their site, they would need users to
their site, they would need users to view those ads, right? Only when ads are
view those ads, right? Only when ads are viewed and ads are clicked would they
viewed and ads are clicked would they make money. So how do they get these
make money. So how do they get these viewers? Well, they use bots. So these
viewers? Well, they use bots. So these bots now view those websites and result
bots now view those websites and result in those ad impressions being triggered.
in those ad impressions being triggered. Now, let's assume that the exchange pays
Now, let's assume that the exchange pays them a $5 CPM for all the ad impressions
them a $5 CPM for all the ad impressions they generated, and it probably cost
they generated, and it probably cost them 50 cents to a dollar to drive that
them 50 cents to a dollar to drive that bot traffic to their site. They would
bot traffic to their site. They would then pocket the difference, which is
then pocket the difference, which is $4. So, I'll repeat the process so it
$4. So, I'll repeat the process so it becomes clear again. So, these hackers
becomes clear again. So, these hackers turn on fake websites, get those fake
turn on fake websites, get those fake websites to join ad networks and
websites to join ad networks and exchanges. So now these sites are
exchanges. So now these sites are eligible to show ads. Ads start showing
eligible to show ads. Ads start showing on those sites. They then buy bot
on those sites. They then buy bot traffic, let's say at a $1 CPM. These
traffic, let's say at a $1 CPM. These bots view those ads. Since the ads are
bots view those ads. Since the ads are being viewed, the ad exchange would pay
being viewed, the ad exchange would pay them, let's say, a $5 CPM, and they
them, let's say, a $5 CPM, and they would then pocket the difference and
would then pocket the difference and make a profit on it. So, they spent $1
make a profit on it. So, they spent $1 to get the traffic, earn $5 on the
to get the traffic, earn $5 on the impressions, $4 is their profit, which
impressions, $4 is their profit, which is about 80% operating margin. Right?
is about 80% operating margin. Right? So, this is just a simple flowchart to
So, this is just a simple flowchart to help you understand how this really
help you understand how this really works. It's actually much more complex
works. It's actually much more complex than this. In fact, in the following
than this. In fact, in the following lesson, I will show you some of the
lesson, I will show you some of the largest ad frauds that have happened in
largest ad frauds that have happened in the last few years, and you'll see how
the last few years, and you'll see how complex it can actually get.
complex it can actually get. But before we get there, I want you guys
But before we get there, I want you guys to watch this really interesting video
to watch this really interesting video created by the team at Integral Ad
created by the team at Integral Ad Science, which shows you a realtime bot
Science, which shows you a realtime bot demo and it'll give you a better
demo and it'll give you a better understanding of how these bots view ads
understanding of how these bots view ads and how ad fraud really happens. So in
and how ad fraud really happens. So in your next lesson, I've shared the link
your next lesson, I've shared the link to this video. You guys can watch this
to this video. You guys can watch this video and once you come back, we'll
video and once you come back, we'll discuss this and look at ad fraud in
discuss this and look at ad fraud in more detail. See you in the next lesson.
more detail. See you in the next lesson. Hello and welcome back. Hope you guys
Hello and welcome back. Hope you guys enjoyed watching that video on the live
enjoyed watching that video on the live bot
bot demonstration. So, in the video, he
demonstration. So, in the video, he infects one laptop with a bot and then
infects one laptop with a bot and then uses another laptop to monitor the bot's
uses another laptop to monitor the bot's activities. And you could see how the
activities. And you could see how the bot was viewing ads that were meant to
bot was viewing ads that were meant to be shown to human beings. And in a short
be shown to human beings. And in a short period of time, it had viewed thousands
period of time, it had viewed thousands of ads. So, you can imagine what
of ads. So, you can imagine what millions of bots could possibly do. One
millions of bots could possibly do. One of the largest ad fraud scams to date
of the largest ad fraud scams to date has been the method scam which hit in
has been the method scam which hit in December 2016. This was covered widely
December 2016. This was covered widely in the mainstream media as hackers made
in the mainstream media as hackers made about $5 million a day by faking about
about $5 million a day by faking about 300 million video views and this to
300 million video views and this to about $180 million in online ads. Again,
about $180 million in online ads. Again, I'll share the links to these articles
I'll share the links to these articles in the resources section of this lesson.
in the resources section of this lesson. Now, $3 to5 million in fraudulent dish
Now, $3 to5 million in fraudulent dish ad revenue per day is massive. So, how
ad revenue per day is massive. So, how did they manage to do this? Well, here's
did they manage to do this? Well, here's the modisa brandy. The hackers first
the modisa brandy. The hackers first created about 6,000 domains. If you
created about 6,000 domains. If you remember the flowchart that we discussed
remember the flowchart that we discussed sometime back in the course where we
sometime back in the course where we talked about how ad fraud works, you'd
talked about how ad fraud works, you'd see the fundamental process is pretty
see the fundamental process is pretty much the same, right? So after they
much the same, right? So after they created these domains, they then created
created these domains, they then created about 250,000 distinct URLs and made
about 250,000 distinct URLs and made them appear that they belong to
them appear that they belong to prestigious publishers like ESPN or
prestigious publishers like ESPN or Woke. This is something we call domain
Woke. This is something we call domain spoofing in the industry where you use
spoofing in the industry where you use technology to make a domain or a URL
technology to make a domain or a URL appear to belong to another publisher.
appear to belong to another publisher. Next, they set up fraud farms to produce
Next, they set up fraud farms to produce fake traffic from more than half a
fake traffic from more than half a million bots, which watched about 300
million bots, which watched about 300 million video ads per day. The average
million video ads per day. The average payout being $13 per thousand fake
payout being $13 per thousand fake views. This was a fairly sophisticated
views. This was a fairly sophisticated operation done at scale. And in order to
operation done at scale. And in order to bypass anti fraud detection mechanisms,
bypass anti fraud detection mechanisms, they obtained hundred thousands of IP
they obtained hundred thousands of IP addresses, associating them with major
addresses, associating them with major United States providers, making them
United States providers, making them look like they come from American
look like they come from American households. This is a classic example of
households. This is a classic example of how sophisticated these ad fraud
how sophisticated these ad fraud operations have
operations have become. Another large scam to hit
become. Another large scam to hit recently in October 2017 was the
recently in October 2017 was the Sportbot scam which affected most of the
Sportbot scam which affected most of the top sports sites across the globe and
top sports sites across the globe and cost advertisers about $250
cost advertisers about $250 million. So why does ad fraud exist in
million. So why does ad fraud exist in the first place? It's out in the open.
the first place? It's out in the open. It's being covered in the mainstream
It's being covered in the mainstream media. People talk about it. So why is
media. People talk about it. So why is no one doing anything about it? And why
no one doing anything about it? And why is it a problem that just seems to be
is it a problem that just seems to be growing? Well, the biggest reason is the
growing? Well, the biggest reason is the open nature of programmatic advertising
open nature of programmatic advertising marketplaces. We talked about the
marketplaces. We talked about the programmatic ecosystem. We said there
programmatic ecosystem. We said there are thousands of ad networks, ad
are thousands of ad networks, ad exchanges, demand and supply side
exchanges, demand and supply side platforms where impressions are bought
platforms where impressions are bought and sold. The ecosystem is massive and
and sold. The ecosystem is massive and there's a lot of money being spent
there's a lot of money being spent there. As per remarketer, about $280
there. As per remarketer, about $280 billion was spent on digital advertising
billion was spent on digital advertising in 2018. With such a large ecosystem and
in 2018. With such a large ecosystem and so much money being spent, we'd expect
so much money being spent, we'd expect some regulation. But unfortunately, as
some regulation. But unfortunately, as of today, the entire programmatic
of today, the entire programmatic ecosystem is completely unregulated. And
ecosystem is completely unregulated. And this is what is helping driving the
this is what is helping driving the growth of ad fraud. Another equally
growth of ad fraud. Another equally important reason why ad fraud exists and
important reason why ad fraud exists and is growing today is that it is
is growing today is that it is technically not illegal in any country.
technically not illegal in any country. That's why a lot of the mafia across the
That's why a lot of the mafia across the globe has started indulging in ad fraud.
globe has started indulging in ad fraud. However, as ad fraud has been growing to
However, as ad fraud has been growing to massive numbers of late, impacting the
massive numbers of late, impacting the entire ecosystem and the money that is
entire ecosystem and the money that is being generated is being used by the
being generated is being used by the mafia to fund other illegal activities.
mafia to fund other illegal activities. A lot of countries have started taking
A lot of countries have started taking note. In fact, an article that I showed
note. In fact, an article that I showed you earlier talked about how the United
you earlier talked about how the United States and the UK governments are coming
States and the UK governments are coming together to combat ad fraud in 2019. So,
together to combat ad fraud in 2019. So, we should see things improve over time.
we should see things improve over time. Another reason that ad fraud continues
Another reason that ad fraud continues to exist is that most digital marketers
to exist is that most digital marketers are completely unaware of how the ad
are completely unaware of how the ad fraud economy really works. And I can
fraud economy really works. And I can tell you this sort of pure personal
tell you this sort of pure personal experience because I train a lot of
experience because I train a lot of digital marketers across the globe. and
digital marketers across the globe. and those who are aware are driven by
those who are aware are driven by misaligned incentives across the entire
misaligned incentives across the entire supply chain leading to inertia and
supply chain leading to inertia and passive treatment of the problem. For
passive treatment of the problem. For years in the digital advertising
years in the digital advertising ecosystem, brands have focused on
ecosystem, brands have focused on measuring the wrong metrics. Measuring
measuring the wrong metrics. Measuring clicks, impressions, clickthrough rates,
clicks, impressions, clickthrough rates, which are not really business metrics.
which are not really business metrics. Of late, this trend is changing and
Of late, this trend is changing and brands today are measuring sales,
brands today are measuring sales, conversions, and brand recall. But for a
conversions, and brand recall. But for a long time, employee incentives were
long time, employee incentives were linked to the amount of traffic they
linked to the amount of traffic they would get or the clickthrough rates they
would get or the clickthrough rates they would drive on their ads. Which is why
would drive on their ads. Which is why most of the digital marketers who were
most of the digital marketers who were aware of this didn't really want to take
aware of this didn't really want to take it up because they were still getting
it up because they were still getting their traffic. They were still getting
their traffic. They were still getting their incentives. However, this is
their incentives. However, this is changing as well and a lot of brands
changing as well and a lot of brands today are focusing more on metrics that
today are focusing more on metrics that impact the business like sales and brand
impact the business like sales and brand recall and incentives are being aligned
recall and incentives are being aligned to these business metrics as well. So,
to these business metrics as well. So, now that we've looked at why ad fraud
now that we've looked at why ad fraud exists in the first place, let's see how
exists in the first place, let's see how we can detect and prevent ad fraud. In
we can detect and prevent ad fraud. In the early days of ad fraud, detecting it
the early days of ad fraud, detecting it was much easier. All you needed to do
was much easier. All you needed to do was go to your Google Analytics reports
was go to your Google Analytics reports and look for traffic sources that had a
and look for traffic sources that had a time on site of 1 second or a bounce
time on site of 1 second or a bounce rate of 100%. And you would know that
rate of 100%. And you would know that those sources were driving bot traffic.
those sources were driving bot traffic. Of late though, bots have become much
Of late though, bots have become much much smarter. Today, you have bots that
much smarter. Today, you have bots that mimic human behavior on a page. They
mimic human behavior on a page. They scroll pages. They click around. They
scroll pages. They click around. They can sign up for email newsletters and
can sign up for email newsletters and they can even make calls. In fact, of
they can even make calls. In fact, of late, I had a client where we ran a
late, I had a client where we ran a massive campaign for them. They got a
massive campaign for them. They got a few hundred calls and a lot of their
few hundred calls and a lot of their calls, almost about 90% from one
calls, almost about 90% from one particular source, had a duration of
particular source, had a duration of just about a second. So the moment the
just about a second. So the moment the call was answered, the line immediately
call was answered, the line immediately got cut. So how do we combat fraudsters
got cut. So how do we combat fraudsters and bots that are becoming smarter and
and bots that are becoming smarter and smarter by the day? Well, here are a
smarter by the day? Well, here are a couple of things that as an advertiser
couple of things that as an advertiser you can do. First thing is stick to
you can do. First thing is stick to trusted ad networks and exchanges. Like
trusted ad networks and exchanges. Like I said, in the programmatic ecosystem,
I said, in the programmatic ecosystem, there are thousands of ad networks, ad
there are thousands of ad networks, ad exchanges. Stick to the ones that you
exchanges. Stick to the ones that you trust. Stick to the ones that are more
trust. Stick to the ones that are more credible. Take references from people
credible. Take references from people you know in the industry. See what's
you know in the industry. See what's worked and stick to the top ones. Avoid
worked and stick to the top ones. Avoid working with smaller exchanges and
working with smaller exchanges and networks that you're not really sure of
networks that you're not really sure of how they're getting their traffic. Next
how they're getting their traffic. Next is to monitor your campaigns closely
is to monitor your campaigns closely with a lot of advanced analytics tools.
with a lot of advanced analytics tools. That way you know if a certain traffic
That way you know if a certain traffic source is bots or real human beings.
source is bots or real human beings. Move to measuring more actionable
Move to measuring more actionable metrics like conversions and leads, not
metrics like conversions and leads, not just clicks or clickthrough rates.
just clicks or clickthrough rates. Because though bots may view ads, click
Because though bots may view ads, click on them, might even fill forms, they
on them, might even fill forms, they wouldn't really buy your product. So
wouldn't really buy your product. So look at the quality of conversions
look at the quality of conversions you're getting from different traffic
you're getting from different traffic sources and take action based on that.
sources and take action based on that. If you're doing digital advertising and
If you're doing digital advertising and programmatic advertising at scale, you
programmatic advertising at scale, you would need to use advanced ad fraud
would need to use advanced ad fraud verification tools like the IAS, White
verification tools like the IAS, White Ops or Pixelate. These would help you
Ops or Pixelate. These would help you figure out which ad networks and
figure out which ad networks and exchanges to work with and also detect
exchanges to work with and also detect if any of your traffic sources have been
if any of your traffic sources have been affected by bots. You can then pause
affected by bots. You can then pause those networks and prevent more money
those networks and prevent more money from going down the drain. Great guys,
from going down the drain. Great guys, so I hope you're much more informed on
so I hope you're much more informed on how ad fraud works, what you can do to
how ad fraud works, what you can do to detect and prevent it. In the next
detect and prevent it. In the next lesson, we'll talk about another digital
lesson, we'll talk about another digital advertising challenge, which is ad
advertising challenge, which is ad viewability. See you in that one.
viewability. See you in that one. Welcome to the lesson on viewability.
Welcome to the lesson on viewability. Let me explain the concept of
Let me explain the concept of viewability with an
viewability with an example. Let's say I loaded this
example. Let's say I loaded this publisher
publisher page, scrolled a little, found an
page, scrolled a little, found an article that I wanted to read, clicked
article that I wanted to read, clicked on that article and left the
on that article and left the page. Now on this page below where I
page. Now on this page below where I clicked on this article, there are
clicked on this article, there are multiple
multiple ads. This one for example. Now when this
ads. This one for example. Now when this advertiser looks at their ad reports,
advertiser looks at their ad reports, would this impression for the ad be
would this impression for the ad be counted? Think about it. What do you
counted? Think about it. What do you think? Would this impression be counted
think? Would this impression be counted even though it had never come into my
even though it had never come into my view? I loaded the page. I came here,
view? I loaded the page. I came here, clicked on article and left. This ad was
clicked on article and left. This ad was way below. Would the impression be
way below. Would the impression be counted? Well, the answer is yes. It
counted? Well, the answer is yes. It might sound ridiculous, but yes. All the
might sound ridiculous, but yes. All the ads on this page that were triggered the
ads on this page that were triggered the moment I loaded this page would have the
moment I loaded this page would have the impression counts go up by one. That's
impression counts go up by one. That's how impressions are measured in digital
how impressions are measured in digital advertising. So if you look at your
advertising. So if you look at your reports for digital ad and if you see
reports for digital ad and if you see 100,000 impressions, well that doesn't
100,000 impressions, well that doesn't mean that the ad even came into view. It
mean that the ad even came into view. It could very well be that the ad did not
could very well be that the ad did not come into view at all and the impression
come into view at all and the impression was still counted.
was still counted. In fact, a very interesting statistic
In fact, a very interesting statistic published by e-arketer and app nexus
published by e-arketer and app nexus recently states that about 46% of all
recently states that about 46% of all digital advertising is never seen.
digital advertising is never seen. That's massive and that's exactly why
That's massive and that's exactly why viewability of ads is a big pain point
viewability of ads is a big pain point for marketers. For a long time, we
for marketers. For a long time, we looked at our impression reports and a
looked at our impression reports and a lot of us thought that our ads were
lot of us thought that our ads were being shown that many times to people.
being shown that many times to people. Well, they were being served, yes. But
Well, they were being served, yes. But were they being seen? That we had no
were they being seen? That we had no clue. In order to make impression
clue. In order to make impression measurement better for marketers and
measurement better for marketers and advertisers, the digital advertising
advertisers, the digital advertising industry came up with its definitions
industry came up with its definitions for something called
for something called viewability. So when is an ad viewable
viewability. So when is an ad viewable or when would an impression be counted
or when would an impression be counted as a viewable impression? Well, there
as a viewable impression? Well, there are standard definitions for it. So for
are standard definitions for it. So for a banner ad, the ad is considered
a banner ad, the ad is considered viewable when at least 50% of the ads
viewable when at least 50% of the ads pixels are in view for at least 1
pixels are in view for at least 1 second. I'll repeat that at least 50% of
second. I'll repeat that at least 50% of the ads pixels or the ads area is in
the ads pixels or the ads area is in view for at least 1 second. When it's a
view for at least 1 second. When it's a video ad, the definition changes a
video ad, the definition changes a little bit. Here 50% of the video's
little bit. Here 50% of the video's pixels or the area needs to be in view
pixels or the area needs to be in view and playing for at least 2
and playing for at least 2 seconds. If these conditions are met,
seconds. If these conditions are met, the impression becomes a viewable
the impression becomes a viewable impression. So at least you know that
impression. So at least you know that your ads have been shown and this many
your ads have been shown and this many impressions have been delivered as per
impressions have been delivered as per these
these definitions. Let me give you an example
definitions. Let me give you an example of a viewability report from Google Ads.
of a viewability report from Google Ads. So, those of you who work on Google Ads
So, those of you who work on Google Ads or run Google Ads can go to your Google
or run Google Ads can go to your Google Ads dashboard, go to your display or
Ads dashboard, go to your display or video campaigns, modify your columns and
video campaigns, modify your columns and add these columns for viewable
add these columns for viewable impressions, non-viewable impressions,
impressions, non-viewable impressions, viewable rate, and viewable CPM. You
viewable rate, and viewable CPM. You look at what percentage of your ads
look at what percentage of your ads actually met the globally accepted
actually met the globally accepted definition of viewability that we just
definition of viewability that we just discussed. So if you look at my
discussed. So if you look at my campaigns here, I blurred out the
campaigns here, I blurred out the campaign names for client
campaign names for client confidentiality reasons. If you look at
confidentiality reasons. If you look at the first campaign, 60 million
the first campaign, 60 million impressions were delivered out of which
impressions were delivered out of which only 21 million met with the definition
only 21 million met with the definition of viewability. The rest did not.
of viewability. The rest did not. There's also a column for measurable
There's also a column for measurable impressions and non-measurable
impressions and non-measurable impressions.
impressions. So if 21 million impressions met the
So if 21 million impressions met the definition of viewability that means the
definition of viewability that means the balance which is 60 minus 21 which is
balance which is 60 minus 21 which is about 39 million impressions should have
about 39 million impressions should have been non-viewable. However this number
been non-viewable. However this number is only about 23 million. Why is that?
is only about 23 million. Why is that? That's because there are two more
That's because there are two more columns here. One which says measurable
columns here. One which says measurable impressions and one which says
impressions and one which says non-measurable impressions. Measurable
non-measurable impressions. Measurable impressions imply partner sites in this
impressions imply partner sites in this case on the Google Display Network where
case on the Google Display Network where they're open to having their viewability
they're open to having their viewability measured. About 15 million impressions
measured. About 15 million impressions were on sites where viewability could
were on sites where viewability could not even be measured. So out of the
not even be measured. So out of the measurable impressions which were 44
measurable impressions which were 44 million, 21 million were viewable and 23
million, 21 million were viewable and 23 million were non-viewable. You can also
million were non-viewable. You can also see what was the viewable rate. It says
see what was the viewable rate. It says 47%. That's because it takes the
47%. That's because it takes the viewable impressions as a percentage of
viewable impressions as a percentage of the measurable impressions. If you take
the measurable impressions. If you take the viewable impressions as a percentage
the viewable impressions as a percentage of the total impressions, this number
of the total impressions, this number would be even lower. So you can imagine
would be even lower. So you can imagine these are ads served on the Google
these are ads served on the Google display network through the Google Ads
display network through the Google Ads platform. And even here barely half the
platform. And even here barely half the impressions were actually viewed as per
impressions were actually viewed as per the definition of
the definition of viewability. If you look at the CPM you
viewability. If you look at the CPM you can see a difference in the CPM as well.
can see a difference in the CPM as well. The average CPM or the cost per thousand
The average CPM or the cost per thousand impressions was 28 in this case while
impressions was 28 in this case while the viewable CPM taking into account the
the viewable CPM taking into account the viewable
viewable impressions was 54. So that's a good
impressions was 54. So that's a good amount of money being spent on ads that
amount of money being spent on ads that are not even coming into the user's
are not even coming into the user's view. So again, for those of you who are
view. So again, for those of you who are running Google Ads campaigns or working
running Google Ads campaigns or working with Google Ads, I suggest you go and
with Google Ads, I suggest you go and test this out on your own campaigns and
test this out on your own campaigns and you'll see the numbers and some of you
you'll see the numbers and some of you don't be surprised at the amount of
don't be surprised at the amount of money you would spend on impressions
money you would spend on impressions that were never in view.
that were never in view. A very interesting resource that I
A very interesting resource that I wanted to share with you is this
wanted to share with you is this measuring ad viewability report
measuring ad viewability report published on the think with Google site.
published on the think with Google site. I'll share the link with you in the
I'll share the link with you in the resources to this lesson. It's a very
resources to this lesson. It's a very comprehensive resource on understanding
comprehensive resource on understanding viewability. It's got a very interactive
viewability. It's got a very interactive demo as well to check how viewability
demo as well to check how viewability works. So if you look at the demo, it
works. So if you look at the demo, it shows you viewability in action. So if I
shows you viewability in action. So if I scroll
scroll below it shows me a site with ads on it.
below it shows me a site with ads on it. So the moment the ad turns green it
So the moment the ad turns green it meets the definition of viewability. So
meets the definition of viewability. So as I
as I scroll you can see this timer that has
scroll you can see this timer that has come on top here. There are four ads on
come on top here. There are four ads on this page. This is the first banner ad
this page. This is the first banner ad that you see which was right above the
that you see which was right above the fold. So it says here if you're seeing
fold. So it says here if you're seeing green the ad is in view. A display ad is
green the ad is in view. A display ad is considered viewable if 50% of the ad's
considered viewable if 50% of the ad's pixels are in view for at least 1
pixels are in view for at least 1 second. So here we've been viewing this
second. So here we've been viewing this ad for the last 36 seconds and counting.
ad for the last 36 seconds and counting. You can see the timer here. It's running
You can see the timer here. It's running live and the ad is 100% in view. As I
live and the ad is 100% in view. As I scroll
below that's the second ad coming in view now. You can see that's ad two.
view now. You can see that's ad two. It's now 38% in view as you can see
It's now 38% in view as you can see there. So it's not yet viewable. The
there. So it's not yet viewable. The moment I scroll
moment I scroll further, it's now 74% in view for more
further, it's now 74% in view for more than 1 second, it now turned viewable.
than 1 second, it now turned viewable. Right? So this was the second ad. So it
Right? So this was the second ad. So it says the most viewable position is right
says the most viewable position is right above the fold, not at the top of the
above the fold, not at the top of the page. Think about it. That's a very
page. Think about it. That's a very interesting insight. When you load a
interesting insight. When you load a website, right at the top of the page,
website, right at the top of the page, which is this one, might not be the best
which is this one, might not be the best position because you might load the page
position because you might load the page and quickly scroll below.
and quickly scroll below. But this position which is just at the
But this position which is just at the fold is a position that would be
fold is a position that would be probably the most viewable as I scroll
probably the most viewable as I scroll below. So we've seen two banner ads so
below. So we've seen two banner ads so far and we've viewed both of
far and we've viewed both of them. Next this is another ad that's
them. Next this is another ad that's come to view. It says 13% in view. I
come to view. It says 13% in view. I scroll below it's now 28% in view 43%
scroll below it's now 28% in view 43% now. And now the moment it was more than
now. And now the moment it was more than 50% in view more for more than 1 second,
50% in view more for more than 1 second, it became
it became viewable. And as it says, you can look
viewable. And as it says, you can look at the insight here as well. It says
at the insight here as well. It says vertical units are the most viewable ad
vertical units are the most viewable ad sizes as they stay on the screen longer.
sizes as they stay on the screen longer. Then as we scroll below, there's a video
Then as we scroll below, there's a video now at the bottom. This is the video ad
now at the bottom. This is the video ad format which is now 10% in view. If I
format which is now 10% in view. If I quickly scroll past the
quickly scroll past the video, you see that I quickly scroll
video, you see that I quickly scroll past it because the time was less than 2
past it because the time was less than 2 seconds. Even though it was 100% in
seconds. Even though it was 100% in view, at some point, it did not count.
view, at some point, it did not count. So, I scroll back up again. Quickly
So, I scroll back up again. Quickly scroll
scroll past, you see it says 10% in view for
past, you see it says 10% in view for 0.1 seconds. If I scroll and scroll back
0.1 seconds. If I scroll and scroll back again, you see it says 7% in view for
again, you see it says 7% in view for 0.9 seconds until it hits the 2C counter
0.9 seconds until it hits the 2C counter for a video. The viewable impression
for a video. The viewable impression number will not go up. So now when I
number will not go up. So now when I scroll up, you can see the moment it hit
scroll up, you can see the moment it hit 2
seconds, it became viewable. Right? So as it says video ads are considered
as it says video ads are considered viewable if 50% of the video ad is in
viewable if 50% of the video ad is in view and playing for at least 2 seconds.
view and playing for at least 2 seconds. Right? So this is a great viewability
Right? So this is a great viewability demo tool. I suggest you guys go around
demo tool. I suggest you guys go around and play with it. That'll help you
and play with it. That'll help you understand viewability
understand viewability better. Apart from this realtime
better. Apart from this realtime viewability demo, there's also an
viewability demo, there's also an interesting section in this report which
interesting section in this report which talks about the state of ad
talks about the state of ad viewability. So they conducted research
viewability. So they conducted research using their measurement technology to
using their measurement technology to understand what drives viewability
understand what drives viewability around the world. And if you click on
around the world. And if you click on this, it will take you to this state of
this, it will take you to this state of ad viewability page where you can look
ad viewability page where you can look at
at viewability across any country that's
viewability across any country that's listed here. So if I wanted to look at
listed here. So if I wanted to look at the current viewability benchmarks in
the current viewability benchmarks in the United
the United States, it says for YouTube ads is
States, it says for YouTube ads is 94%. Web and app video ads is 66% and
94%. Web and app video ads is 66% and display ads the benchmark is 49%. So if
display ads the benchmark is 49%. So if you're getting if you're in the United
you're getting if you're in the United States and your display ads have a
States and your display ads have a viewability of 49%. You know you're at
viewability of 49%. You know you're at the benchmark because that's what is the
the benchmark because that's what is the current state in the United States.
current state in the United States. Likewise, let's say you're in Europe and
Likewise, let's say you're in Europe and you want to check for let's say the
UK. You can see here it says 94% on YouTube, 50% for display and 67% for web
YouTube, 50% for display and 67% for web and app video. Let's say you're in the
and app video. Let's say you're in the Middle East. So you could do the same
Middle East. So you could do the same thing for three countries available
thing for three countries available here. So in the UAE you can look at your
here. So in the UAE you can look at your benchmarks. If you're in AP pack again
benchmarks. If you're in AP pack again you can look at let's say India and you
you can look at let's say India and you can see what the benchmarks are here.
can see what the benchmarks are here. You can
You can also look at what categories have the
also look at what categories have the highest viewability rates. So it says
highest viewability rates. So it says that categories that create captivating
that categories that create captivating content have higher
content have higher viewability. You can look at which
viewability. You can look at which categories have the highest display and
categories have the highest display and ad viewability rates.
ad viewability rates. And it also tells
And it also tells you what you can do to create more
you what you can do to create more viewable ads. Talks about optimizing for
viewable ads. Talks about optimizing for mobile
mobile devices
devices using ad sizes that have highest
using ad sizes that have highest viewability. So you can take a look at
viewability. So you can take a look at which ad sizes have the highest
which ad sizes have the highest viewability rates for video as well as
viewability rates for video as well as for display and focus more on those
for display and focus more on those sizes. And it also tells you or gives
sizes. And it also tells you or gives you an interesting insight on ads being
you an interesting insight on ads being viewable below the fold as well. So this
viewable below the fold as well. So this is the fold. You can see the screen fold
is the fold. You can see the screen fold here. Just below the fold is where a
here. Just below the fold is where a good chunk of inventory is viewable as
good chunk of inventory is viewable as well. So it's not that you'd want to
well. So it's not that you'd want to pick only above the fold inventory. You
pick only above the fold inventory. You could even experiment with below the
could even experiment with below the fold inventory because even below the
fold inventory because even below the fold inventory has a viewability rate of
fold inventory has a viewability rate of about
about 47%. Great guys. So that's an awesome
47%. Great guys. So that's an awesome resource on viewability. Go ahead and
resource on viewability. Go ahead and explore it. I will see you guys in the
explore it. I will see you guys in the next lesson. Welcome to the lesson on
next lesson. Welcome to the lesson on deal ID. So what is a deal ID? A deal ID
deal ID. So what is a deal ID? A deal ID or a deal identifier is the unique
or a deal identifier is the unique number of an automated ad buy in the
number of an automated ad buy in the programmatic advertising ecosystem. It's
programmatic advertising ecosystem. It's a unique code that matches buyers with
a unique code that matches buyers with sellers individually based on a variety
sellers individually based on a variety of criteria that they've negotiated
of criteria that they've negotiated beforehand. The criteria could include
beforehand. The criteria could include the minimum price the advertisers
the minimum price the advertisers allowed to bid, type of ad units they
allowed to bid, type of ad units they bought, which section of the site the
bought, which section of the site the ads would show on amongst a bunch of
ads would show on amongst a bunch of other
other things. The reason we are discussing
things. The reason we are discussing deal ID here is because it is used in
deal ID here is because it is used in three types of programmatic advertising.
three types of programmatic advertising. It's used in private auctions, preferred
It's used in private auctions, preferred deals, and programmatic guaranteed deals
deals, and programmatic guaranteed deals as well.
as well. In the next section when we talk about
In the next section when we talk about the different types of programmatic
the different types of programmatic advertising, you'll understand how this
advertising, you'll understand how this deal ID fits into the scheme of things.
deal ID fits into the scheme of things. So that's in a nutshell what deal ID is.
So that's in a nutshell what deal ID is. Thank you and I will see you in the next
Thank you and I will see you in the next lesson. Hello and welcome back. Let's
lesson. Hello and welcome back. Let's dive deeper into programmatic
dive deeper into programmatic advertising. Before we go further, this
advertising. Before we go further, this is what we've covered so far. Just a
is what we've covered so far. Just a quick recap. We began with understanding
quick recap. We began with understanding what programmatic advertising is. We
what programmatic advertising is. We looked at the fundamentals of the risk
looked at the fundamentals of the risk advertising option, how risk advertising
advertising option, how risk advertising has evolved over time. We then went
has evolved over time. We then went through the entire programmatic
through the entire programmatic advertising ecosystem, understood each
advertising ecosystem, understood each component. We looked at DSPs, SSPs, ad
component. We looked at DSPs, SSPs, ad networks, and ad exchanges and learned
networks, and ad exchanges and learned what their role is in the ecosystem. We
what their role is in the ecosystem. We then understood various programmatic
then understood various programmatic advertising terminology. We looked at
advertising terminology. We looked at frequency cap above the fold and below
frequency cap above the fold and below the fold inventory, run off site and run
the fold inventory, run off site and run off network types of ad buying. We also
off network types of ad buying. We also looked at important concepts of brand
looked at important concepts of brand safety, ad fraud and viewability as
safety, ad fraud and viewability as well. In this lesson, we look at the
well. In this lesson, we look at the four different types of programmatic
four different types of programmatic advertising. The first type is something
advertising. The first type is something called the open auction. It's also
called the open auction. It's also sometimes called RDB, which is realtime
sometimes called RDB, which is realtime bidding. The second type is a private
bidding. The second type is a private auction sometimes called a PMP which is
auction sometimes called a PMP which is a private marketplace or an IOA which is
a private marketplace or an IOA which is an invitationon auction. So I'll repeat
an invitationon auction. So I'll repeat that. PMP is private marketplace and IOA
that. PMP is private marketplace and IOA is invitationonly auction. That's a few
is invitationonly auction. That's a few more threeletter acronyms or TLAs for
more threeletter acronyms or TLAs for you to remember. Then we've got
you to remember. Then we've got preferred deals, sometimes just called
preferred deals, sometimes just called PD or sometimes also referred to as UFR,
PD or sometimes also referred to as UFR, which is unreserved fixed rate. And then
which is unreserved fixed rate. And then we've got programmatic guarantee deals,
we've got programmatic guarantee deals, which most of time are just referred to
which most of time are just referred to as PG deals. So let's look at each one
as PG deals. So let's look at each one of these, understand what they are. The
of these, understand what they are. The first one is the open auction or RTB,
first one is the open auction or RTB, which is realtime bidding. We've already
which is realtime bidding. We've already discussed how digital advertising
discussed how digital advertising auctions work in an earlier lesson. What
auctions work in an earlier lesson. What happens in this case is a user visits a
happens in this case is a user visits a site. The website has inventory
site. The website has inventory available. The website then communicates
available. The website then communicates with its SSP or its supply side
with its SSP or its supply side platform. So I want you guys to remember
platform. So I want you guys to remember that flowchart of the programmatic
that flowchart of the programmatic ecosystem that we talked about. So
ecosystem that we talked about. So publishers get in touch with their
publishers get in touch with their supply side platform. The supply side
supply side platform. The supply side platform then passes on the information
platform then passes on the information through the exchange to the demand side
through the exchange to the demand side platforms. On individual demand side
platforms. On individual demand side platforms, advertisers then place their
platforms, advertisers then place their bids. The auction takes place and the
bids. The auction takes place and the highest bid wins the impression. This
highest bid wins the impression. This process is called realtime bidding
process is called realtime bidding because every time you refresh a
because every time you refresh a website, the auction happens. This
website, the auction happens. This entire process takes place. Like I've
entire process takes place. Like I've told you earlier, the process takes
told you earlier, the process takes place really fast at a speed faster than
place really fast at a speed faster than the time you take to blink your eye and
the time you take to blink your eye and the right impressions are delivered to
the right impressions are delivered to the user who's loaded the site. So
the user who's loaded the site. So that's what an open auction is. It's
that's what an open auction is. It's very similar to the auction that we
very similar to the auction that we discussed earlier in this
discussed earlier in this course. Next is a private auction or a
course. Next is a private auction or a private marketplace or PMP. What happens
private marketplace or PMP. What happens here is the publisher instead of giving
here is the publisher instead of giving all their inventory out into the open
all their inventory out into the open auction hosts a private auction where
auction hosts a private auction where only certain players are invited to bid.
only certain players are invited to bid. So let's say the publisher has 10
So let's say the publisher has 10 premium advertisers. It could invite
premium advertisers. It could invite those 10 premium advertisers into a
those 10 premium advertisers into a private auction and tell them that well
private auction and tell them that well this is inventory that only 10 of you
this is inventory that only 10 of you can bid on. You guys can bid on it. The
can bid on. You guys can bid on it. The auction takes place. Whoever wins the
auction takes place. Whoever wins the auction will get their ad to show. Since
auction will get their ad to show. Since it's a private auction, the publisher
it's a private auction, the publisher here gets to offer premium inventory to
here gets to offer premium inventory to its select premium advertisers and also
its select premium advertisers and also charge them a slightly higher price. How
charge them a slightly higher price. How do the systems know which advertisers
do the systems know which advertisers are eligible to take part in the private
are eligible to take part in the private auction? Well, this happens through the
auction? Well, this happens through the deal ID. So, the deal ID that we talked
deal ID. So, the deal ID that we talked about earlier is given by the publisher
about earlier is given by the publisher to those select advertisers. They would
to those select advertisers. They would enter the deal ID into their DSPs. The
enter the deal ID into their DSPs. The deal ID would then identify the deal and
deal ID would then identify the deal and help them bid in that invitation only
help them bid in that invitation only auction. Right? So that's a private
auction. Right? So that's a private marketplace or a private auction. As of
marketplace or a private auction. As of now, I'm just giving you the
now, I'm just giving you the fundamentals of how it works. In future
fundamentals of how it works. In future lessons, we'll talk about the pros and
lessons, we'll talk about the pros and cons of each one of these from an
cons of each one of these from an advertiser perspective. Right? For now,
advertiser perspective. Right? For now, let's just stick to understanding the
let's just stick to understanding the concepts of what these are and how they
concepts of what these are and how they work.
work. So now that we looked at private
So now that we looked at private auctions, let's see how they compare
auctions, let's see how they compare with the third type of programmatic
with the third type of programmatic advertising which is a preferred deal.
advertising which is a preferred deal. So the private auction as we discussed
So the private auction as we discussed is a publisher inviting a select group
is a publisher inviting a select group of advertisers into an invitationonly
of advertisers into an invitationonly auction where they bid on that inventory
auction where they bid on that inventory and whoever wins the auction gets their
and whoever wins the auction gets their ad to show. In a preferred deal, which
ad to show. In a preferred deal, which is the third type of programmatic
is the third type of programmatic advertising, what happens is the
advertising, what happens is the publisher strikes a deal with just one
publisher strikes a deal with just one advertiser here as against a private
advertiser here as against a private auction where they invite multiple
auction where they invite multiple advertisers. Here the deal is struck
advertisers. Here the deal is struck with just one
with just one advertiser. So this is a onetoone deal
advertiser. So this is a onetoone deal with no auction happening. A fixed price
with no auction happening. A fixed price for the inventory is set. So let's say
for the inventory is set. So let's say they set a fixed $5 CPM. The advertiser
they set a fixed $5 CPM. The advertiser is then issued a deal ID. The advertiser
is then issued a deal ID. The advertiser can go to their DSP. Enter the deal ID
can go to their DSP. Enter the deal ID and buy the inventory for a fixed $5 CPM
and buy the inventory for a fixed $5 CPM without any competition in an auction.
without any competition in an auction. The only thing to note here is that the
The only thing to note here is that the inventory is not guaranteed. The price
inventory is not guaranteed. The price is guaranteed, the volume is not. So
is guaranteed, the volume is not. So though the advertiser has access to the
though the advertiser has access to the inventory at a $5 CPM, there is no
inventory at a $5 CPM, there is no guarantee from either side. The
guarantee from either side. The advertiser is not guaranteeing to buy a
advertiser is not guaranteeing to buy a fixed set or number of impressions. The
fixed set or number of impressions. The publisher too has not guaranteed that
publisher too has not guaranteed that they would receive a fixed number of
they would receive a fixed number of impressions which is why it's sometimes
impressions which is why it's sometimes called a UFR or an unreserved fixed
called a UFR or an unreserved fixed rate. So the rate is fixed but the
rate. So the rate is fixed but the inventory is not reserved or guaranteed.
inventory is not reserved or guaranteed. Which brings us to the fourth type of
Which brings us to the fourth type of programmatic advertising which is a PG
programmatic advertising which is a PG deal or a programmatic guaranteed deal.
deal or a programmatic guaranteed deal. In this case, the publisher and the
In this case, the publisher and the advertiser strike a onetoone deal. And
advertiser strike a onetoone deal. And here both the price and the volume are
here both the price and the volume are fixed. So they strike a deal saying the
fixed. So they strike a deal saying the advertiser would buy 1 million
advertiser would buy 1 million impressions for the publisher in this
impressions for the publisher in this time period at a fixed rate of let's say
time period at a fixed rate of let's say a $15 CPM. Again, the publisher creates
a $15 CPM. Again, the publisher creates the deal, gives the deal ID to the
the deal, gives the deal ID to the advertiser. The advertiser will put this
advertiser. The advertiser will put this deal ID into their demand side platform
deal ID into their demand side platform and have access to the inventory which
and have access to the inventory which is both a fixed price and volume. Right?
is both a fixed price and volume. Right? So let's summarize what we've just seen.
So let's summarize what we've just seen. Let's compare the open auction or
Let's compare the open auction or realtime bidding versus private auction
realtime bidding versus private auction or a private marketplace versus a
or a private marketplace versus a preferred deal versus a programmatic
preferred deal versus a programmatic guaranteed deal. So in the open auction
guaranteed deal. So in the open auction you have hundreds of buyers competing at
you have hundreds of buyers competing at an auction the dynamics of which we
an auction the dynamics of which we discussed in earlier lesson. The
discussed in earlier lesson. The auctions are second price auctions. The
auctions are second price auctions. The winning bidder would get their ad to
winning bidder would get their ad to show. Private auctions are invitationon
show. Private auctions are invitationon auctions where a certain number of
auctions where a certain number of advertisers are invited by a publisher
advertisers are invited by a publisher to have access to their inventory. The
to have access to their inventory. The price is negotiated. There's a minimum
price is negotiated. There's a minimum price or what we call a floor price
price or what we call a floor price that's decided. So the publisher would
that's decided. So the publisher would tell all advertisers that the floor
tell all advertisers that the floor price for the auction is $3 for CPM.
price for the auction is $3 for CPM. Each advertiser would bid above that.
Each advertiser would bid above that. The auction would take place the same
The auction would take place the same second price auction and the winning
second price auction and the winning bidder would get their ad to show.
bidder would get their ad to show. Volumes here are non-g guaranteed. With
Volumes here are non-g guaranteed. With preferred deals, there is no auction.
preferred deals, there is no auction. It's a one-on-one deal struck between
It's a one-on-one deal struck between the publisher and the advertiser where
the publisher and the advertiser where the price is negotiated and fixed, but
the price is negotiated and fixed, but the volume is non-g guaranteed. The last
the volume is non-g guaranteed. The last one which is programmatic guaranteed
one which is programmatic guaranteed deals. Again, it's a one-on-one deal
deals. Again, it's a one-on-one deal between the publisher and the advertiser
between the publisher and the advertiser where both the price is fixed and the
where both the price is fixed and the volume is reserved or guaranteed. The
volume is reserved or guaranteed. The last three are deal ID based. In all the
last three are deal ID based. In all the last three types, the publisher would
last three types, the publisher would give a deal ID or a deal identifier to
give a deal ID or a deal identifier to the advertiser which they can then enter
the advertiser which they can then enter into their DSP and buy the inventory.
into their DSP and buy the inventory. Right? Great guys. So, that's the four
Right? Great guys. So, that's the four different types of programmatic
different types of programmatic advertising. I hope you enjoyed learning
advertising. I hope you enjoyed learning about them. In the following lessons, I
about them. In the following lessons, I will take you through, like I said, the
will take you through, like I said, the pros and cons of each type and as an
pros and cons of each type and as an advertiser, when should you pick what
advertiser, when should you pick what type. See you in the next lesson.
type. See you in the next lesson. Welcome back. In the previous lesson, we
Welcome back. In the previous lesson, we talked about the different types of
talked about the different types of programmatic advertising. Now, before we
programmatic advertising. Now, before we go ahead and understand the pros and
go ahead and understand the pros and cons of each, it's important for you to
cons of each, it's important for you to know the concept of something called the
know the concept of something called the publisher waterfall. The publisher
publisher waterfall. The publisher waterfall is a term for how publishers
waterfall is a term for how publishers prioritize their ad inventory. Before we
prioritize their ad inventory. Before we understand that prioritization model,
understand that prioritization model, it's important for you to know what
it's important for you to know what premium inventory is. Now, premium
premium inventory is. Now, premium inventory and nonpremium inventory are
inventory and nonpremium inventory are terms that you'd keep hearing in the
terms that you'd keep hearing in the programmatic advertising space. So, when
programmatic advertising space. So, when publishers mention the term premium
publishers mention the term premium inventory, what does that really mean?
inventory, what does that really mean? Well, so premium inventory is basically
Well, so premium inventory is basically publisher inventory that would drive
publisher inventory that would drive high impact and be extremely valuable to
high impact and be extremely valuable to advertisers. Common examples of this
advertisers. Common examples of this premium inventory would be roadblocks,
premium inventory would be roadblocks, mast heads, homepage takeovers, and
mast heads, homepage takeovers, and other similar high impact units. Now,
other similar high impact units. Now, some of you might not be familiar with
some of you might not be familiar with what these roadblocks, mast heads, and
what these roadblocks, mast heads, and homepage takeovers are. So, let me show
homepage takeovers are. So, let me show you some examples of these. The best
you some examples of these. The best example of a master would be the YouTube
example of a master would be the YouTube master. So if you open YouTube either on
master. So if you open YouTube either on a desktop or on your mobile devices and
a desktop or on your mobile devices and if you're on a country where YouTube
if you're on a country where YouTube sells its master head inventory, you'll
sells its master head inventory, you'll see an ad here which would remain the
see an ad here which would remain the same for a 24-hour period. So this ad
same for a 24-hour period. So this ad unit is sold to just one brand for a
unit is sold to just one brand for a period of 24 hours, generally from
period of 24 hours, generally from midnight to midnight. Now this is truly
midnight to midnight. Now this is truly high impact inventory. The reach would
high impact inventory. The reach would vary from country to country, but in
vary from country to country, but in India where I live, the reach is about
India where I live, the reach is about 100 million people in a day and it's
100 million people in a day and it's about 1 billion impressions. That's how
about 1 billion impressions. That's how big it is and that's why it's called a
big it is and that's why it's called a high impact ad unit. So this is an
high impact ad unit. So this is an example of premium inventory. So let's
example of premium inventory. So let's take another example. Let's go to this
take another example. Let's go to this website called Times of
website called Times of India. Now before the site loads, you
India. Now before the site loads, you see this ad here. It says
see this ad here. It says timesofindia.com will load in a few
timesofindia.com will load in a few seconds. This ad that you see before a
seconds. This ad that you see before a site loads is something called a
site loads is something called a roadblock. It is high impact again
roadblock. It is high impact again because it's something a user would
because it's something a user would definitely see before they move on to
definitely see before they move on to site. And this ad that you see on the
site. And this ad that you see on the screen now is what you call a homepage
screen now is what you call a homepage takeover. You'd see the entire homepage
takeover. You'd see the entire homepage taken over by the same brand. So those
taken over by the same brand. So those were some examples of roadblocks, mast
were some examples of roadblocks, mast heads, and homepage takeovers. So now
heads, and homepage takeovers. So now you know what these are and why they're
you know what these are and why they're called high impact ad units because they
called high impact ad units because they give advertisers massive reach and
give advertisers massive reach and visibility. Now that you know what
visibility. Now that you know what premium inventory is, let's move on to
premium inventory is, let's move on to understand what we called earlier the
understand what we called earlier the publisher waterfall. It's also sometimes
publisher waterfall. It's also sometimes called the daisy chain. It's nothing but
called the daisy chain. It's nothing but the way in which publishers prioritize
the way in which publishers prioritize their ad inventory. An ad server, for
their ad inventory. An ad server, for those of you who are not familiar, is
those of you who are not familiar, is just a software that helps publishers
just a software that helps publishers prioritize their inventory, manage ads
prioritize their inventory, manage ads on their websites, as well as track the
on their websites, as well as track the ad
ad performance. So when you load a
performance. So when you load a publisher website, the ad server would
publisher website, the ad server would first look for any direct deals that
first look for any direct deals that have been done between the publisher and
have been done between the publisher and any advertiser. A direct deal are deals
any advertiser. A direct deal are deals that are done offline without going
that are done offline without going through any programmatic advertising
through any programmatic advertising platforms. This is the old way of
platforms. This is the old way of digital advertising that we talked about
digital advertising that we talked about in an earlier lesson where advertisers
in an earlier lesson where advertisers and publishers close the deal offline
and publishers close the deal offline and everything is done without really
and everything is done without really using any programmatic platforms. The
using any programmatic platforms. The reason publishers continue to do this is
reason publishers continue to do this is because they can charge a lot more on
because they can charge a lot more on inventory sold directly which is why
inventory sold directly which is why this is given the first priority. Always
this is given the first priority. Always remember this process tries to ensure
remember this process tries to ensure that publishers make the most money from
that publishers make the most money from their digital assets.
their digital assets. If there are direct deals available,
If there are direct deals available, then the inventory that was promised to
then the inventory that was promised to the direct deal is given away. Next, the
the direct deal is given away. Next, the ad server checks if there were any
ad server checks if there were any programmatic guaranteed deals. We talked
programmatic guaranteed deals. We talked about programmatic guaranteed in the
about programmatic guaranteed in the previous lesson. This is basically a
previous lesson. This is basically a deal that is struck offline, but the ads
deal that is struck offline, but the ads are delivered through the programmatic
are delivered through the programmatic ecosystem. The inventory here is
ecosystem. The inventory here is guaranteed and the price is fixed. Since
guaranteed and the price is fixed. Since the publisher can sell a large volume of
the publisher can sell a large volume of inventory at a predefined price, this is
inventory at a predefined price, this is a second priority for them. Water
a second priority for them. Water inventory has been committed to a PG
inventory has been committed to a PG deal is given away here. Next, the ad
deal is given away here. Next, the ad server checks if there are any preferred
server checks if there are any preferred deals that have been done. Preferred
deals that have been done. Preferred deals, like we discussed in the previous
deals, like we discussed in the previous lesson, are deals where the advertiser
lesson, are deals where the advertiser and the publisher negotiate a fixed
and the publisher negotiate a fixed price but do not guarantee each other on
price but do not guarantee each other on the volume that would be bought or sold.
the volume that would be bought or sold. Since the publisher does not have any
Since the publisher does not have any guarantee on the volume here, this is
guarantee on the volume here, this is given priority only after direct deals
given priority only after direct deals and programmatic guaranteed deals. So
and programmatic guaranteed deals. So advertisers who have a preferred deal
advertisers who have a preferred deal would now have access to inventory that
would now have access to inventory that has not already been consumed by the
has not already been consumed by the direct deal or with the programmatic
direct deal or with the programmatic guaranteed deal. Once the preferred
guaranteed deal. Once the preferred deals in the system, if there were any
deals in the system, if there were any are executed, the ad server would then
are executed, the ad server would then check for private auctions. So the
check for private auctions. So the inventory that has not been sold yet is
inventory that has not been sold yet is then given out to the private auction
then given out to the private auction where a certain set of advertisers who
where a certain set of advertisers who already struck a deal with the publisher
already struck a deal with the publisher now bid and the winning bidder then gets
now bid and the winning bidder then gets their ad to show. After all this is
their ad to show. After all this is done, whatever inventory is left is then
done, whatever inventory is left is then passed on to the open auction. So the
passed on to the open auction. So the open auctions you must remember almost
open auctions you must remember almost always get what we call remnant
always get what we call remnant inventory or what most publishers prefer
inventory or what most publishers prefer calling nonpremium because they don't
calling nonpremium because they don't like using the word remnant. So you'd
like using the word remnant. So you'd see both these terms being used in the
see both these terms being used in the industry. So either non-premium
industry. So either non-premium inventory or remnant inventory is what
inventory or remnant inventory is what is available to the open auction. So
is available to the open auction. So this process is called the publisher
this process is called the publisher waterfall. And it's important for you to
waterfall. And it's important for you to keep these in mind because in the next
keep these in mind because in the next lessons when we talk about the pros and
lessons when we talk about the pros and cons of the different types of
cons of the different types of programmatic advertising, what you've
programmatic advertising, what you've learned in this lesson will help you.
learned in this lesson will help you. Now, this entire process takes place in
Now, this entire process takes place in just about 200 milliseconds. It takes
just about 200 milliseconds. It takes you 300 milliseconds to blink your eye
you 300 milliseconds to blink your eye through a time period faster than the
through a time period faster than the time it takes you to blink your eye. The
time it takes you to blink your eye. The publisher ad server completes this
publisher ad server completes this entire process. It first looks for
entire process. It first looks for direct deals, gives the committed
direct deals, gives the committed inventory there, then checks for PG
inventory there, then checks for PG deals, then preferred deals, then
deals, then preferred deals, then private auctions, and finally the open
private auctions, and finally the open auction. Today, some publishers also use
auction. Today, some publishers also use another process to have their ad
another process to have their ad inventory sold, which is something
inventory sold, which is something called header bidding. In header
called header bidding. In header bidding, instead of the inventory being
bidding, instead of the inventory being given away step by step like you see
given away step by step like you see here, it is done simultaneously through
here, it is done simultaneously through a tech heavy process. The concept of
a tech heavy process. The concept of header bidding is beyond the scope of
header bidding is beyond the scope of this course and as an advertiser or a
this course and as an advertiser or a marketer, you wouldn't really need to
marketer, you wouldn't really need to know too much about it. But if you're
know too much about it. But if you're really interested, I'd suggest looking
really interested, I'd suggest looking up Google or YouTube for it. Trust the
up Google or YouTube for it. Trust the concept of the publisher waterfall. Keep
concept of the publisher waterfall. Keep this in mind for the next lesson and
this in mind for the next lesson and I'll see you there. Welcome to this
I'll see you there. Welcome to this lesson on the pros and cons of the open
lesson on the pros and cons of the open auction. Before we move further into the
auction. Before we move further into the lesson, a quick recap of what an open
lesson, a quick recap of what an open auction is. Like we've discussed
auction is. Like we've discussed earlier, an open auction is that
earlier, an open auction is that realtime bidding process where you have
realtime bidding process where you have an auction with multiple publishers,
an auction with multiple publishers, multiple advertisers bidding on the ad
multiple advertisers bidding on the ad inventory. It's a second price auction
inventory. It's a second price auction and the winning bidder would have their
and the winning bidder would have their ad displayed. We also saw in the
ad displayed. We also saw in the previous lesson on the publisher
previous lesson on the publisher waterfall that open auctions are the
waterfall that open auctions are the last priority when it comes to publisher
last priority when it comes to publisher ad servers. So let's look at the
ad servers. So let's look at the advantages of the open auction. The
advantages of the open auction. The first one is setting up an open auction
first one is setting up an open auction is much easier compared to the other
is much easier compared to the other types of programmatic advertising. You
types of programmatic advertising. You don't really need to get in touch with a
don't really need to get in touch with a publisher. You don't have to strike a
publisher. You don't have to strike a deal. You can just head to your DSP,
deal. You can just head to your DSP, create a new campaign just like you
create a new campaign just like you would do on, let's say, the Google Ads
would do on, let's say, the Google Ads platform, set your targeting, upload
platform, set your targeting, upload your creatives, and you could be up and
your creatives, and you could be up and running in no time. Unlike other types
running in no time. Unlike other types of programmatic advertising which have
of programmatic advertising which have some level of human negotiations
some level of human negotiations involved, open auctions are completely
involved, open auctions are completely driven by the system. Next is with open
driven by the system. Next is with open auctions, you've got multiple powerful
auctions, you've got multiple powerful targeting options. So for those of you
targeting options. So for those of you who are familiar with Google ads, you
who are familiar with Google ads, you have multiple targeting options there.
have multiple targeting options there. You can target by geography, by age, by
You can target by geography, by age, by gender. You have something called
gender. You have something called inmarket audiences. You've got affinity
