Public Workshop on Competition in Television and Digital Advertising (Panel 1)


MR. WILLNER: Good
afternoon, everyone. I’m Carl Willner. I’m an attorney with
the Antitrust Division’s telecommunications
and broadband section. I’m a 29-year veteran of
DOJ and have focused in my career here on the
telecommunications and media industries. Today I’m pleased to be
moderating our first panel of speakers on the topic
of “Television Advertising – The Nuts and Bolts of
Broadcast and Cable.” We have a very impressive
array of speakers, representing broadcast
TV, cable advertising, industry analytics,
academic and advertising consumer perspectives. Our aim for this panel is
to provide a foundation for all of the panels that
you will hear during this workshop by addressing
how advertising works with broadcast TV stations
and cable TV systems, including allocation
of advertising minutes, pricing and the
effectiveness and uses of various forms
of advertising. Though the next panel
after this will be focused more specifically on
digital, we will also be addressing the
relationship of traditional linear TV
with the various forms of digital online
advertising, including mobile devices and
over-the-top OTT providers. We will also consider some
of the difficult problems of measurement and
comparison across various forms of advertising. Now, let me introduce our
five speakers in the order that you will hear from
them today, going from left to right from your
point of view in the audience. After their opening
presentations, we will have hopefully about an
hour for questions and for discussion, including a
few minutes at the end for audience questions. And I’ll start by
introducing Pat LaPlatney. Pat is the former CEO
of Raycom Media and now co-CEO and president
of Gray Television. His background is in
media sales, distribution, programming and
digital media. He has been involved in
the advertising sales business since the early
1980s and he tells me that he has the scars
to prove it. He sits on the board
of Gray Television, the National Association
of Broadcasters and the Greater Montgomery YMCA. He is also incoming
chairman of the NBC Affiliate Board. Our second speaker
will be Mark Lieberman. Mark is the president
and CEO of Viamedia, the country’s largest
independent local cross-media
advertising company. Prior to Viamedia, Mark
was the co-founder, chairman and CEO of TRA,
the first TV big data company. Before then, he was
chairman and CEO of IVT, an early video streaming
company and served as EVP of Reed Elsevier where
he oversaw a variety of broadcasting and cable
and multichannel news magazines. In the mid-1990s, he
cofounded DIVA Systems, the first commercially
viable cable VoD provider. Mark also served
in the George H.W. Bush administration as
associate deputy secretary and assistant secretary
for technology acting at the U.S. Department of Commerce. Our third speaker, next
to me, will be Rick Ducey. He’s the managing director
of BIA Advisory Services. Rick leads BIA Advisory
Services strategy consulting practice and
serves as an advisor to its affiliated investment
banking group, BIA Capital Strategies. Rick works with media,
ad tech and MarTech, a blending of marketing and
technology companies, to see and evaluate
opportunities and strategies in the market
to achieve corporate goals. He helps clients with
strategic planning, developing partnership
strategies, product planning and achieving
revenue growth and diversification. He also helps marketers,
including brands and agencies, with effective
planning and use of digital media platforms
and he also works on investment banking
projects, assisting both buy-side and
sell-side clients. Rick cofounded SpectraRep
in 2000, a technology company that specializes
in emergency alert and warning solutions. Prior to joining BIA,
Rick headed the National Association of
Broadcasters research and information group,
overseeing all research and information
technology initiatives. Rick is an adjunct
professor at George Washington University,
where he teaches classes in digital media
entrepreneurship. Our fourth speaker will
be Professor Ken Wilbur. Ken is an associate
professor of marketing at the University of
California, San Diego’s Rady School of Management,
and also an economist. His research focuses on
how people respond to ads, whether by changing the
channel, changing a brand attitude, initiating a
Web search or buying a product. He has published 15 papers
related to television advertising and media
economics in the past 15 years. And then, our final
speaker on this panel will be Marc Pritchard, who is
the chief brand officer at Procter & Gamble, who you
heard mentioned in the previous lecture. Marc is responsible for
P&G’s brand-building disciplines worldwide. He sets the company’s
multibillion dollar media, marketing and advertising
strategies and leads marketing innovations that
guide communications and brand-building for P&G’s
portfolio of trusted, quality products. As the top brand builder
and veteran of P&G for more than three decades,
Marc believes in the power of brands to serve people
with the best performing products, while also being
a force for good through ethics and responsibility,
community impact, diversity and inclusion,
gender equality and environmental
sustainability. He continually leads P&G’s
brand-building reinvention and is a leading voice in
the media, marketing and creative industry. And with that, let me turn
to our first speaker, Pat LaPlatney. MR. LAPLATNEY:
Thanks, Carl. Good afternoon. I’d just like to point
out I love that Henry Ford quote. I think I’ll use
that going forward. I’ve never heard
that one before. I’ve heard the Wanamaker
quote a lot, but not that one. MR. DELRAHIM: Right. MR. LAPLATNEY: So why
don’t we dive right in, if we can? Advertising in linear
television has been historically Gray’s
core product. There’s three types
of ad inventory. There’s network, like
primetime, local news and local programming where
the broadcaster owns and sells all of
the inventory. And then, there’s
syndicated programs, syndicated inventory where
the inventory is divided between the syndicator
and the station. It’s sold three ways —
locally, regionally and nationally. And for Gray, in 2018,
local and regional was roughly 80 percent of
our advertising revenue, excluding political. But the reality is viewing
patterns and shifts in technology are
changing the business fundamentally. Today, Gray distributes
content through mobile apps, websites, social
media and OTT platforms, in addition to linear. I’m going to walk you
through a typical media sale here in a
local market. And our example or our
illustration today is a local agency that places
the media for a key Dodge Chrysler Jeep dealer,
essentially a local car dealer, a family business. The client’s a typical
large advertiser in a small market. The agency is determined
that college football viewers are key
to selling cars. And they go out and
they produce multiple commercials featuring
different models, trucks, Jeeps, SUVs. So I’ll take you back to
2005 when reaching college football video viewers
was much, much simpler. There were essentially
four sources: the local ABC, CBS or NBC affiliate
and the local cable company who sold five
cable networks, some national, some regional,
but four sources. And according to the
agency, the buy would have looked something
like this. You’ve got 86 percent of
the buy goes to broadcast, 14 percent of it
goes to cable. A much more simple, but
more profitable time for local broadcasters. You can tell by the number
of logos on this page that times have changed. The four network
affiliates still sell college football. But the local cable
providers individually or through an interconnect
can sell national sports networks, conference
networks and regional sports networks. But wait, there’s more. In addition to the
networks, you can — in addition to the networks
you can buy from cable, you can also buy Big Four
and national cable mobile apps, essentially video
and handheld devices. You can buy video in
over-the-top outlets, which is essentially
digital video primarily on the big screen. And then, there’s other
college sites, including national college sites and
national non-live sports content sites. There’s a very fragmented
marketplace today for sports, very fragmented. So I’m going to take a
minute — I’m trying to rush through this. But I’m going to take a
minute on this one with you because it’s
rather important. This is a national media
plan for that local car dealer in a small
Southeastern market. The ABC network — I’m
going to walk you through each one of
these components. It’s a $20,000
video-only buy. So this is strictly video. The 5 percent in the upper
left is essentially a geo-targeted pre-roll on
ESPN sold through ABC. If you go to the SEC
network piece on the right, in the middle,
that’s essentially the same product. It’s a geo-targeted
digital video. The digital piece — the
piece labeled digital is video targeted college
football fans and that’s 100 percent funded by Kia. So Kia’s directing their
dealers into that medium. Then you’ve got cable
company number one, cable company number two. There’s no interconnect
in this market. Taking 25 percent of
the plan, and then local broadcast is targeted for
20 percent of this buy with an over-the-top
provider targeted at 5 percent. So if you think about that
for a second, roughly half this buy is targeted to
handheld devices, half. Very different than two
years ago, really very different than a year ago. And the reality is
that these different alternatives create a
constraining impact on pricing for broadcasters. Fragmentation
is the issue. Reduced viewership for any
one game, a lower reach, lower value to
the advertiser. The top-rated game of the
DMA may be on broadcast, it may not. And by the way, the
top-rated game in market A may not be the top-rated
game nationally. And the end of the day,
broadcasters can be bottlenecked. Rates end up getting
reduced to maintain share just simply
get on the buy. And here are the numbers
for that station from this advertiser over the
last six or eight years. It’s pretty stark
and pretty painful. This is happening with big
local clients across our universe, a lot
of car dealers. I’m going to switch gears
for a minute and go to the news market. You know, you can see from
the headline there that Nielsen has had some
methodology changes and those methodology changes
have resulted in increased ratings to cable at the
expense of local news. I don’t know if you can
see that chart very well. But I’ll try to do
my best to explain. So the four sets of bar
charts on your left would be local television
stations. And the bars in purple are
pre-change in methodology. The bars in green
are post-methodology. But the way, this is not
just Colorado Springs. This is the entire
country for Nielsen. You can see the drop in
ratings for the local news stations. And if you look off to
the right, that’s the aggregate cable universe
and they’re essentially the beneficiary for that. The reality is more rated
cable networks means more sellable inventory in
the local market and more competition for
broadcasters. So just a quick reminder,
digital advertising has numerous advantages. There’s strong targeting
capabilities that yield higher CPMs. It’s cheaper to buy. And increasingly, the big
guys are setting up local sales capability. Gray is investing in
automated media buying to help address this, and as
is the broadcast industry. Still at the
developmental stage. But essentially what we’re
trying to do is make our buying process
more like digital. Ad buyers are telling
us that the dollars will shift back to local
broadcast with the implementation of
automated buying. Back to college football
for a minute, which is big in our part of the world,
broadcast television had only one of the top-rated
bowls and three of the top 20 in 2018. Essentially broadcast
is losing the battle for sports viewership. And when you lose
viewers, you lose revenue. And when revenue declines,
business reality dictates that you cut costs. And as an example here,
you know, in the past there were robust, you
know, big sports teams in certain markets. Today you’re down to one
or two people doing two minutes on the local news. And that’s really
unfortunate. But what’s more
important is that further competitive dilution into
local news and weather will diminish our ability
to serve our communities, particularly in
smaller market. So we believe the division
should recognize the impact of increased
competition and allow broadcasters to achieve
the economies of scale necessary to continue to
provide our vital service. Thank you very much. MR. WILLNER:
Thank you, Pat. And now, let me turn to
our second speaker, to Mark Lieberman
of Viamedia. We need to get the second
presentation loaded. MR. LIEBERMAN: I can do
Gray again, if you want. MR. LAPLATNEY:
It’s all ready. MR. LIEBERMAN: Gray,
Gray Television? MR. WILLNER: There we go. MR. LIEBERMAN: All right. Let’s continue to drink
out of the firehose here. There’ll be a test at the
end of the conference. So this is a chart which I
like which just shows how the TV programming
industry is exploding with the different types of
content and different types of devices. And television is
actually going through a renaissance with consumers
being able to watch any kind of content they want,
whenever they want, on multiple
different devices. However, this is creating
a huge challenge for advertisers. And the advertisers, what
they want is they want data, as Professor Athey
said before, in order to drive targeted advertising
and reduce waste. And where does that data
typically come from? It comes from the last
mile or the edge, that closest to the viewer. And that data typically
comes from the MVPDs, who both have video
programming as well as ISPs. And we talk — we’re going
to be talking a lot today about TV and digital. And from my standpoint, TV
and digital are not equal and they’re not yet
considered a holistic buy for a few reasons. One is, is that, as the
professor talked about, television provides a
brand-safe environment, something I’m sure that
we’re going to hear about more. Second is that the
agencies actually have different buyers. So an agency — the same
agency may have a TV buyer, a video buyer, a
digital buyer and maybe a strategic person perhaps
even overlooking it. But typically there are
three different buyers. Fourth is that, from an ad
tech standpoint, there’s no connective tissue
between the digital world, digital ad tech
and TV ad tech. And as has been touched on
before and I think will be touched on later
is measurement. There’s no unified
cross-media measurement platform. And then lastly is that
when it come to privacy, again, something that I’m
sure will be discussed, is that there’s a privacy
paradox which is data is the oxygen that drives
the media economy. However, regulations will
inhibit the use of that data to do better
targeting, to hit those fragmented audiences. So the result of which
is at least one analyst, MoffettNathanson, said
that digital is not yet cannibalizing
TV advertising. If it’s cannibalizing
anything, it’s cannibalizing at the local
level, direct mail and newspapers. This is a slide I’m not
going to go into any detail. But it’s similar to the
slide the professor had, that video ad tech is
incredibly complex. Overlay on top of that,
when it comes to advanced television advertising,
whether it’s addressable or programmatic or the
other forms, is that agencies — each agency
has their different solution. Each MVPD has
their own solution. And the networks now
actually are embarking upon their own solutions. So again, no unified
platform, which makes it that much more challenging
for an advertiser. So let’s talk a little
bit more about television. And Pat talked about the
broadcast industry and I’ll talk a little bit
more and in more detail about local cable and
cable specifically. But TV is a brand-safe
environment. It’s foundational. Think of television
advertising — a little bit like that Henry Ford
quote — as painting your house. You’ve got to
paint your house. Why? Yeah, it’ll make it look
better, but you have to protect the wood. The wood is your brand. You don’t paint
your house? The wood is going to rot. So television
is a must buy. On broadcast, you can see
the breakdown here between national broadcast
and local broadcast. And broadcast tends to be
at the top of the funnel, of the path-to-purchase
funnel that was discussed earlier, which is that
brand-building and awareness exercise. Local cable, where I live,
is typically both at the top of the funnel
sometimes for those local merchants. But it’s also at the
bottom of the funnel where you can actually have
an outcome, a business outcome which
is a purchase. So let’s dig deeper
into spot cable. Spot cable is two to three
minutes an hour and the derivation of those two to
three minutes an hour is that when the cable
networks are negotiating a carriage deal with the
MVPDs, the MVPDs keep two to three minutes an hour. Seventy-five percent of
that is for advertising. Twenty-five percent is
typically for marketing, for the MVPD to market
their own services. It’s in aggregate just
under a $6 billion industry, growing about
$1 to $7 billion in a political year. There’s 120,000 local
merchants that advertise on television. And these local merchants,
you may have heard a lot about DTC or
direct-to-consumer brands. They have the attribute of
a direct-to-consumer brand because they have a direct
relationship with that consumer. And why? Because it’s in a
geo-targeted area. The geo-targeted area is
what’s called an ad zone. And the ad zone, there’s
210 DMAs in the United States. There’s 45,000 ad zones
in the United States that allows that MVPD to
target in their particular geographic region. How is the
sales happening? And here you can see the
biggest players in local spot cable ad sales. And some of these numbers
you’ll see are a little bit larger than what you
might expect because the large MVPDs actually run
the ad sales for many of their competitors in
a particular market. And so, this includes both
the ONO subscribers as well as the subscribers
for the MVPDs whose ad sales these MVPDs run. And interestingly enough,
they’re not just running the advertising. They’re running
the marketing. So they’re running the
marketing for their competitors in that market
to compete against their own subscriber
acquisition. So let’s drill down
a little bit further. And this pie chart is not
indicative of Viamedia. It’s indicative of our
assumption around the entire industry. There’s three types of
advertisers that access the geographic,
geo-targeted ad zone approach. There’s national, and
that’s sold through a rep firm called NCC, who
the entire industry has relationships with. There’s regional, and
regional is typically sold through what’s called an
interconnect, which Pat referred to before. An interconnect, think
of as a joint selling arrangement amongst
competing MVPDs in a market. It’s generally run by
the largest MVPD in that market. Think of it almost like a
mutual fund in which the mutual fund manager
actually has one of the stocks is stock
that they own. So what that really means
is that the mutual fund manager — the
interconnect manager — really has to be like
Caesar’s wife, has to be above suspicion and run
this fairly on behalf of the entirety of that
particular market. And how are these ads
inserted, national, regional and the
local-local ads, as we sometimes say, which are
those local merchants. Those ads are inserted. They’re trafficked. They’re remotely
monitored. They’re encoded. They’re all done by the
MVPD or a third party representative. In terms of pricing, and I
won’t go into detail here because Carl may have some
questions about it before. But suffice it to say that
the CPMs will actually be greater when there’s more
targeting; in this case, geo-targeting. So local are going to be
higher CPMs than maybe than national. One particular category
that I want to highlight — we’re in Washington,
D.C., which is the second largest category in the
space behind auto, is political. And you can see here in
the last several years how political has grown
enormously and who knows what’s going to happen
in 2020, given what’s happening in this town. It could be a total
eclipse of the sun in terms of the amount of
advertising on television. But this really brings
home the point about geo-targeting and going
after that swing voter or swing purchaser in
the case of a brand. And if you’re let’s
say a Northern Virginia congressman or
congresswoman, you may not want to advertise on
broadcast in the whole DMA because there’s going to
be wasted advertising in the District
and in Maryland. You want to advertise just
where your voters are. And that’s really the
advantage that spot cable has. It brings it home for all
the different types of advertisers. Just a quick snippet on
Viamedia and where we fit into the ecosystem is that
we do focus, like Pat, on cross-media selling,
so beyond cable. But we run the cable
advertising sales for 60 different MVPDs around the
country as a third party representative. But it’s a lot
more than that. We insert 1 million ads
every day in 140 different head-ins around the
country in 30 states. And who do we compete
— and we have 6,000 advertisers. And who do we
compete with? We compete with other
MVPDs to run third party representative services
for other MVPDs in that market and we compete
against those MVPDs as well for ad sales. Here’s just a map of where
we sit and logos of the different operators whose
ad sales that we run and around the country, all
run out of one center in Kentucky. And this is a snapshot of
the different advertisers that we have, local,
regional and national. Thank you. MR. WILLNER: Thank you. Now, let’s turn to our
third speaker, Rick Ducey of BIA, to address some
of the perspectives on industry analytics. DR. DUCEY: Great. Thank you, Carl. So I’ll get started here. We are an advisory
services firm. And I’ll — let me
get the deck up here. We provide information
services, analytic services and advisory
services to companies in the local media business
particularly and also to government agencies. Let me see if I
can get this up. So this is sort of a
snapshot of some of the companies we work with,
media companies, some digital companies and
we also provide our data services and analytics to
some government agencies, the FCC and to the
Department of Justice as well, all trying to
understand what’s going on in a market. So Pat and Mark gave deep
views of what’s going on from the local broadcast
perspective and the local cable perspective. We track all of the ad
movement and the dynamics going on in local
advertising, activation in local markets. So we do a forecast
looking at 16 different media platforms providing
local advertising, from direct mail to broadcast
TV and a lot of digital entities in that
space as well. We’ll look ahead
five years. People need to see
what’s coming up next. Pat would like to know
what his market will look like and what the
opportunities are for adjustable markets and
certainly the local cable wants to see that too. So we try to inform
leaders in the industry, trying to see where
things are headed. So for 2019, the total
sum across the U.S. of the local kind of
spending that Pat and Mark have been talking about
sums to about $146 billion or so. And we heard how that is
a fragmented market just within the video slice
between for broadcast and for cable. But it’s actually
even more fragmented. We look at, as I
mentioned, 16 different media platforms. So that $146 billion that
gets spent targeting local consumers at a TV market
level or in some of the zones that Mark was
talking about gets split up among a number of
different kinds of advertising categories. Direct mail was one of
the ones we were talking about, radio, outdoor. There’s a number of
different ways that advertisers want to
reach out to consumers to influence, as Professor
Athey was saying, that purchase funnel or the
journey to purchase. What we’re seeing when
we’re looking at how the advertising gets invested
in different media activations in the local
markets, just as Pat showed in detail what Gray
Media’s experience has been as the competition
gets a little bit more fragmented within video,
it’s even more fragmented across the board. But generally we’re
seeing a lot more digital competitors coming in as
media platforms into local marketing and advertising. So the ad spend going to
these different digital platforms was maybe a
third or so as recently as last year. We see that by 2020 going
up to — well, approaching half. I mean, so almost one
out of two ad dollars targeting local audiences
will come from some sort of a digital platform,
whether that’s Hulu or Google AdWords or email
has become a particularly viable advertising medium
and has new technology capabilities. If we look just at
video, I think, Mark, you mentioned that video
is foundational. And we certainly see that. If we look at just the
video world, again, Pat gave us that more detailed
view of what’s happening from an operator’s
perspective. We see the same thing. So the blue bar in the
bottom, that’s the amount of advertising dollars
invested in local broadcast TV stations. The bar sitting on top
of that, that’s the world that Mark lives in and he
was talking about that was sort of $6 billion-ish. That’s local cable, that
two or three minutes that local cable networks
make available. And you can see the blue
bar is fairly stable across our forecast period
going out to three — going out to 2023;
roughly the same for the orange/reddish bar,
which is cable. There are some ups and
downs a little bit in the even years, as we were
talking about, eclipsing the sun, political spend
in the even years and Olympics spend gives a lot
of lift opportunity for cable and TV and
local markets. But together the cable
and the TV ad forecast spending, from Marc
Pritchard’s perspective, or revenue, from the
perspective of Pat and Mark, relatively stable. So it’s kind of
foundational. It does its job. But the ability to target
by geo and by demographics and get some additional
data about return on investment, attribution,
things like that, those kind of tools and datasets
and practices are much more prevalent in
a digital world. That’s starting to attract
a lot of spend, eventually perhaps as much as 50
percent of the ad spend targeting local audiences. So in the video world,
those top bars are relatively small compared
to broadcast and cable. But that’s where
the growth is. So there’s a lot of spend
being attracted into over-the-top, mobile
and desktop video. And that’s significantly
reshaping the dynamics of the local marketplace. If we look at Washington,
where we all are today, about $3.6 billion is
being spent if you’re from Washington to target you. And if you’re here today
from out of Washington, you’re still being
targeted anyway since you’re here. So the details are a
little bit too small. But you can see that pie,
kind of we call it the share of wallet. There’s a certain amount
of money that advertisers are investing to reach
people in Washington. The pie has a lot
of different pieces. And Professor Athey
was talking about that purchase funnel. Somehow that all gets
sorted out into how advertisers and agencies
are trying to reach their target consumers in a
market like Washington through local kinds
of activation. So in this chart, what
we’re showing is how print and traditional media and
other media kind of shape up. And you can see the
blue is the digital. That’s going to
be increasing. That’s digital. And if you look over
there, $413 million or so goes to Google on the
digital side out of the Washington, D.C. market. Let me just back
this up a little bit. Local TV, the red bar over
there, takes away less than Google. So all the TV stations
in the Washington, D.C. market earn from
their advertising revenue less than just one of the
digital providers, Google. Just sort of a fact
that’s interesting to contemplate. I’ll spend a couple of
minutes before I close to talk more about that
purchase funnel. It is complicated. And this is a study that
was sponsored by Google, Millward Brown and Polk
looking at automotive. We’ve been talking
about automotive, a very important
advertising category. So what is that
actually like? Pat did a great job
walking us through what that’s like from the TV
operator’s perspective. From the advertiser and
consumer perspective, according to this study,
from the moment we say, geez, I think I’ll buy a
car to the time you turn the key and drive off the
lot, according to this study, there’s about 24
steps along that consumer journey. And each of those steps
involves a touchpoint for an advertiser, a marketing
touchpoint to influence whatever that person
is thinking about then. And again, it’s not
necessarily a linear process. But at each of those 24
touchpoints, there’s an opportunity for an
advertiser to give a little nudge saying
remember me, remember Kia, remember our
local dealer lot. And the industry, both the
advertising side and the media side, is trying to
come up with a nice set of solutions in that
multisided market. The media’s job is to
aggregate the audience as potential consumers and
the audience’s job is to say I don’t want to look
for every single show individually. I’d like to go into some
sort of official platform to find those shows. And then, the advertiser
says I want to get to all those audiences
you’re getting to. And I have that media
split choices of how I can influence each of the
steps along this multipath decision tree basically
for consumers as they do their journey. So 19 of those
decision-making points, touchpoints, are digital. And so, that has become
very influential. That’s why so much spend
is moving into the digital domain. And media allocations
for auto dealers and manufacturers, here
we break it out. And you can see that
digital gets about 40 percent, which is typical
for a number of the different business
categories. TV comes in heavily
in this market. This happens to be a very
strong newspaper market. So newspaper gets a little
bit more ad spend than TV overall. But just looking a little
bit more from the Kia perspective in a
Southeastern market that Pat was looking at, and
more generally across the country, as he said, it’s
not just that one market. And that’s what we’re
seeing too in our forecasts and our
analytics looking at a category like auto and
then more broadly across all of the local
categories. So a few quick takeaways. I mean, I think the data
shows and our forecast supports the notion based
on our model and drivers that we’re seeing. TV and cable, the media
that Pat and Mark are talking about,
are foundational. They have an effect. They’re effective. But the data, the
targetability and the accountability that comes
from digital media are pretty persuasive
to advertisers. So they’re allocating
more spend there. And that’s eliciting
various competitive responses among all the
players in the 16 media platforms that
we’re looking at. Growth in video is
concentrated in digital, a lot of fragmentation,
like Pat said. And all of these local
media companies, there’s 24 touchpoints to buying
an ad — sorry, along the way to buying a car, all
of those media competing to be part of each of
those touchpoint spends. So there’s a lot of kind
of fungibility among the different categories. One example I was talking
about earlier is there was one advertiser that was a
hundred percent in direct mail, which typically
takes about 25 percent of the total ad spend. Direct mail is huge. And a TV seller said
just give me some of your budget for TV. We’ll draw out some
business metrics that were looking at search
life and website lift. And those business
outcomes were achieved. And so, that buyer went
from a hundred percent direct mail, convinced
of the power of local TV stations and flipped and
they became a hundred percent local TV. So you don’t normally
think of direct mail competing with TV. But it happens. You know, we see that. And that is my
presentation. Thank you, Carl. MR. WILLNER: Thank you. And now, we’ll turn to our
fourth speaker, Professor Ken Wilbur, who will
provide some academic perspectives on
advertising and its effectiveness. And take it away, Ken. DR. WILBUR: Thank you. So — MR. WILLNER: And
let’s be sure we have your presentation loaded. DR. DUCEY: It’s kind of
a scary picture up there now. MR. WILLNER: But you could
do the BIA presentation too, I’m sure. DR. WILBUR: So I’m going
to try to provide maybe four relevant facts that I
think might be relevant to the department’s
regulatory goals, based on high quality evidence. If these slides cited
research papers that might be useful, everything is
freely downloadable on my personal website, which
is under my own name. So the first fact relates
to the possibility for brand advertisers to
discern the economic value of those expenditures. And much of the public
discussion in this area is done by advertising
agencies who talk a very good game. However, if you scratch
the surface, you find out that many of the people
speaking don’t seem to know what constitutes
high quality evidence. And they may have economic
incentives to distort the evidence, such as the
endemic principle agent issues in this
marketplace. So what I tell my MBA
students is that brand advertising is much
like human nutrition. We would like very much to
have a formula that would tell us what combinations
of foods to eat to achieve particular health
objectives. However, we have nothing
even close to that. The reason is threefold. One, measurement
is a problem. Two, statistical power
is often a problem. We’re talking about very
fairly small effects in a very noisy environment. And three is attribution,
which is just distinguishing causality
from correlation. And you can go deeper
on each one of those. What I’m showing you
here, the points are brand attitude surveys. And the lines are
moving averages. And the bars are
the amount spent on advertising by each brand
each week over a five-year period. It’s true that many
television — many television ads provide
new information. But many do not. They simply remind you
about brands you already know about. And when you look at that
data, you see enormous variation in the
advertising spend and very little variation in the
favorable attitudes held by the consumers
of those brands. And so, this just gives
you a sense of the challenge in the
statistical power aspect of estimating the
economic value of brand advertising. Now, there’s a smaller
proportion of advertisers where the situation
is very different. We often call those direct
response or performance advertisers. And as Susan said, this is
more at the campaign level than the brand level. Many large brands
are doing both types. So what I’m showing you
here is the Google Search volume for “DraftKings”
and “FanDuel” minute by minute during the first
full hour of the opening NFL football game in 2015. The light gray bars, which
might not be showing up very well on the monitor,
are the commercial minutes. And I don’t have to tell
you which commercial minutes contained ads for
“DraftKings” and “FanDuel” because the search for
those brands jumps by a factor between 15 and 25
each time one of their ads appeared on TV. You also see positive
spillovers to the competing brand. You see the effect die
out within five minutes. You see no apparent
hangover of the search volume after the ad. And you see no
corresponding increase during the commercial
minutes that did not have the ads, suggesting that
it was the presence of the ads and not simply the
absence of the game. Now, similar effects have
been found by quite a few researchers in quite
a few contexts now. You might think that the
industry would be jumping all over this to use
this as a measure of advertising effectiveness. I’ve personally been a bit
disappointed by how slowly the take-up has been. As an example, I showed
this graph to executives at one of the
brands involved. And if you sell ads to
them, you might be able to guess which one. And it was very clear that
they had no idea this was going on. And as the discussion laid
bare, they had an external team that handled
TV advertising. They had an external team
that handled search engine marketing. And they had an internal
team that handled search engine optimization
and website analytics. And they had no structures
in place for these teams to talk to each other. In fact, the search
advertising and TV advertising teams
essentially were in competition with each
other for budget. I hope we’ll
catch up here. But there’s still
a long way to go. Now, third, the way
consumers are consuming television has changed. There may be a little
bit left to go. But it’s mostly over. Television has mostly gone
from a focal medium to a peripheral medium. There are many
indicators of this. And I don’t need to do
much to convince you of this because you’ve
observed it in your own households. There are some very
interesting new data sources. One example is a startup
out of Cambridge called Tvision Insights. And what they do is
they pay a few thousand households to put their
cameras in front of the TV. The cameras run algorithms
which are very good at detecting household
members that are in the room, identifying which
household members each one belongs to based on facial
recognition, identifying whose eyes are open and
whose faces are pointed at the television screen. They do this about
six times per second. And here are some broad
descriptives of their data. You might not be able
to see the axis at the bottom. So let me summarize. Across a very large
sample of TV advertising exposures, approximately
95 percent of households remain tuned for the
entire ad duration. However, only about 53
percent of the people remain in the room and
only about 7 percent of viewers point their faces
at the screen during the TV ad. There is some variation
across day parts and networks and
program genres. But that variation
seem to be limited. However, this
is promising. For the first time
advertisers may be able to get measures of how
many people are actually consuming the ads as
opposed to simply having the opportunity to see. The final relevant fact
goes to market definition for advertising. So what I did was I pulled
some data from a source called Kantar Stradegy,
which is a pretty good database of expenditures
on five different categories of
TV advertising. It’s not perfect and it’s
not fully comprehensive. Unfortunately it excludes
local spot cable. However, what I did to
get a sense — (Laughter.) MR. WILLNER: That’s
what it’s all about. DR. WILBUR: I wish
it was comprehensive. This was the best I could
do in the couple of weeks I had. What I did was first I
ranked all the advertisers by their total spend in
2018 and I considered the top thousand, which
accounted for about 85 percent of the
measured TV revenues. And it’s over $60 billion
in spending last year. So it’s appreciable. And then I counted the
proportion of those top thousand who spent at
least 5 percent of their ad budget on each pair
of TV advertising media. So on the diagonal, you
have the proportion of those top thousand
advertisers who spend at least 5 percent on each
individual medium and then you see sort of the
intersection of those on the off-diagonal elements. And to summarize, there is
a very strong correlation in the advertisers served
by all five categories of TV advertising, which I
think could suggest that these markets are rival in
expenditure within these advertising organizations. So partial evidence and
a bit ad hoc, but perhaps relevant and perhaps
useful for defining markets in future
enforcement actions. I’m not going to go
through the full slide. I may have
exhausted my time. I just wanted to explore
one point that Susan touched on, which was one
very different element of linear TV markets from
digital advertising markets is that
there are referees. Traditionally Nielsen
had a monopoly on this. Now they have some good
competition from ComScore and other firms in telling
advertisers how many people may have seen
their advertising. By contrast, most digital
industry estimates are that between 10 and 30
percent of advertising expenditures are paid
for machines to view advertisements. Now, I don’t know how
they come up with those numbers. There are different
estimates. This is an inherently
unmeasurable phenomenon. The important point I
think is that the way the industry is currently
structured is that Facebook tells its
advertisers how many humans were exposed to the
ads they buy on Facebook. And Facebook assesses the
probability that any given exposure is to a
human or a machine. And Facebook does not
reveal to the advertiser how they assess that
probability and how they trade off the risk of a
false positive versus a true negative. So there are — sometimes
my students ask me is TV going away. I think that the only way
that might occur is if the digital platforms find
a way to appropriate the high quality programming
that’s currently served by TV providers. But the marketplace
characteristics for advertisers are I think
much better in linear TV than they are in
digital video. So I think those were
the main points I thought might be relevant and
happy to pass the clicker. MR. WILLNER: Great. Thank you very much, Ken. And can we put up the
slide for the last presentation? And Marc, Marc Pritchard
of P&G, of course, will be doing our last
presentation today. However, you’re only going
to see one slide during this presentation. There’s nothing wrong
with your monitor. It’s just that he’s
delivering written remarks rather. But we wanted to have at
least a slide up during the course of the
presentation so you wouldn’t see a
blank screen. You’re not supposed
to see that on TV. MR. PRITCHARD: Yeah. Thank you, Carl. And good afternoon,
everyone. You’re probably familiar
with most of our brands at P&G — Tide, Pampers,
Crest, Charmin, Olay, Vicks, Head &
Shoulders, Gillette. I could go on, another 65. And every day, P&G brands
reach about 5 billion people on the
planet every day. We spent about $7 billion
last year on advertising and related merchandising,
which is about 10 percent of our sales. And we do so because it
creates awareness of the benefits that these brands
offer to consumers and hopefully convert that
awareness into purchase. Advertising has been
crucial to the success of our business since P&G’s
been around for over 181 years. Back in the late 1800s,
when the printing press was invented, Ivory soap
ran print ads in local newspapers with the
message “99 44/100 percent pure, so pure it floats.”
In the 1920s, P&G pioneered radio soap
operas, sponsored by brands such as Oxydol
laundry detergent, hence the name soap operas. We pivoted to visual
television ads in the late 1940s and used the mass
medium to reach nearly every household in
the United States. And then, in the 1990s, as
the Internet brought new ad formats, we got into
search, social media and online video. Now today, advertising
is being disrupted. Traditional TV
continues to decline. Even networks are starting
to cut back on the number of ads that they run. Digital advertising is
the leading form of media. But ad-blocking
is increasing. Over-the-top streaming
services are emerging with no ads at all. E-commerce is growing,
which is driven by algorithms as
opposed to ads. The agency world
is in flux. And data analytics and
technology are pervasive in how we build our brands
and how we do business every day. So in the roughly
$700 billion worldwide advertising market,
there’s never been more choices for how to
reach consumers. But the famous quote by
advertising pioneer John Wanamaker is as true today
as it’s ever been, which is, “Half my
advertising is wasted. The trouble is I don’t
know which half.” While choices abound, the ad
industry has too much waste, driven by
lack of transparency. Buying practices favor
providers and too many middlemen are between
advertisers and consumers. This waste negatively
affects consumers because seven out of 10 people
say that ads are annoying. And that’s because too
often people see the same ad over and over again in
the same program, on the same channel and across
multiple platforms. Now, what P&G is doing is
trying to constructively disrupt the industry by
reinventing brand-building from what is currently
wasteful mass marketing to mass reach with
one-to-one precision. We want to do that to
drive out waste and we want to improve, most
importantly, the consumer experience. But it’s quite a
challenge, given how the ad industry operates. Television media does
have some advantages. I agree with everything
everybody’s says. Content quality is
known and controlled. So advertisers can be sure
that the brands, that our ads show up on high
quality programs. TV media buying follows
common measurement standards such as ratings
and cost per thousand metrics or CPMs,
as you know. And there’s really no
better way to reach large mass audiences at once,
especially with major events like the Super Bowl
that could reach up to 100 million highly
engaged viewers. However, although TV
ratings and viewership continue to decline,
pricing continues to rise year after year. I always thought that
was a real mystery. Why does that happen? You look at how the bulk
of TV media is bought and it’s kind of
understandable. There’s a time-honored
ritual called the upfronts which is happening this
week where media providers present to buyers. They showcase the
programming for the new season. They talk about the
inventory available and they provide what
their offers are. And the name of this
ritual belies the intent, which is to sell as
much media inventory as possible upfront. Once it’s committed, it’s
difficult for buyers to course correct. With the upfront
commitment, we have limited flexibility
to make changes. Media is largely locked in
for as much as a year and for at least the
next two quarters. And although there are
opportunities to give back extra inventory bought as
an option, it’s not really much of an option because
the only choice you really have is to return it
without a penalty. Available ad inventory and
pricing is determined by the media providers
and adjusted for actual viewership based on data
from household panels after the fact. So there’s little
flexibility to make choices or changes along
the way in real time like there is in other areas. And media agencies largely
conduct negotiations on behalf of marketers,
bringing a lot of spending together from multiple
companies so they can achieve economies
of scale. This service then comes
with a fee and the media agency often receives
rebates from media providers, sometimes
disclosed, sometimes not, which raises the ultimate
price to consumers and to marketers. Now, this timeworn TV
media buying system continues to favor
providers in our view with too many middlemen
in between. And what it results in
is too many ads that are placed on the same program
to fulfill inventory commitments. And that annoys consumers. Digital media seemed like
a welcomed respite from the TV problems, with
the promise of unlimited inventory, data for
precision placement, options for supply and
demand competition and creativity beyond the
constraints of a 30-second ad. But there’s been a dark
side, most notably lack of transparency which is
leading to substantial waste, outright fraud
through lots and issues of brand safety. In 2017, P&G called for
the industry to clean up and we insisted on one
viewability measurement so we could know when we’re
reaching an audience and they’re actually
seeing the ad. We asked for transparency
— print transparent agency contracts so we
knew how agencies were using our money. We wanted to know — we
wanted third party, MRC, which is the Media Ratings
Council, verified audience measurement so we know we
were getting the ads that we paid for. And we wanted to have
certified fraud prevention so we know that humans
are seeing our ads, not robots. And finally, brand safety
so that we know our ads show up in the
right place, not in objectionable content like
ISIS terrorist training videos, which
there were some. Progress has been made. But problems persist. Ads are still found on
unacceptable, graphic and horrible content. We’re still wasting money
in annoying consumers with too many ads because
marketers are flying with little visibility and have
really very little idea of whether our ads are
reaching the same person over and over again
across platforms. And pricing is largely
determined now by algorithmic auctions, with
no visibility on how the algorithm works and little
room for negotiation, if any. Advertising is essential
to the success of P&G brands. So we’re taking action
to lead constructive disruption of
the industry. In addition for calls
of full measurement transparency within
providers, we’re now calling for common
measurement standards across all media platforms
to eliminate, if nothing else, just excess ads. We’re calling for controls
over content quality so our brands show up
in the right places. We’re calling for common
national privacy data collection standards so we
have one way in which we measure data, not the
27 different that happen today from a
privacy standpoint. We’re actually moving
to more programmatic solutions and to more
performance media and marketing solutions
and bringing more media planning in-house and
buying in-house to get direct access to providers
and to make more informed choices to
drive out waste. And most important, not
just eliminate waste, but make the ad experience
better for the consumers we serve. Thank you. MR. WILLNER: Thank you, Marc. for that very
thoughtful perspective from a very major and
committed advertising consumer. And now, I’m going to
turn to questions to the panelists. I think we may have a
little less time for questions that I
originally envisioned. But I will try to cover a
number of important topics and also still leave a
little time for audience questions at the end
before we move to the next panel. So my first question, and
I’d like to really throw this one open to the
entire panel because several of you have
touched on this, is whether you — can you
directly compare prices of digital and online
advertising with broadcast TV or cable. How are each of these
normally measured? And if you can’t compare
them now, what is it going to take to get to a
unified metric, taking into account price
and effectiveness? And I’m happy to let
anyone volunteer to be the first respondent to that. MR. LAPLATNEY: So we would
love to see that, you know, speaking for Gray
and I think for the local television industry. I know that both comScore
and Nielsen have said they’re moving in
that direction. But frequently they’re
moving in that direction for long periods of time
before that actually occurs. So, you know, over time, I
think the best outcome for our industry is to have
basically a single pool of inventory across all of
our platforms that we sell and we have, you know, a
reasonable amount of data associated with that. And ultimately I think
that’s best for the advertising community
and for local TV. MR. LIEBERMAN: I’ll just
jump in to say I think that it’s going to be
a long time before it happens, to have a really
unified cross-media measurement platform. Having said that, one
of the things that we’ve done, and the biggest
challenge that I’ve found is not necessarily
technology. There’s challenges there. It’s not necessarily
business models. There’s challenges there. It’s training. We have a couple hundred
sellers in the local markets that have been
selling television for 20, 30 years. So how do you retrain them
on how to sell other forms of media and how do you
retrain the local agencies and the local advertisers
on how to buy different forms of media. So we’ve actually
launched what we call a cross-metrics dashboard
in which the advertiser or the agency or our account
executives can log in and see the different forms of
media tied to whatever the KPIs are. But it’s still siloes. It’s three or four or
five different siloes — social, email,
display, OTT, cable TV. But at least the
advertiser can kind of start to see how each one
of them are performing, not as to the other, but
how they’re performing in a silo. MR. DUCEY: Part of it
is specifying the goal. From a video perspective,
what is the audience experience? What’s the consumer
experience? So is it the full-screen
commercial experience like you would expect on linear
TV, like broadcast or cable, or a full episode
player on Hulu or some OTT platform. So if it’s a full-screen,
so the viewability is not a question. It’s there. And it’s some duration. It’s 30 seconds. It’s six seconds. It’s seven seconds. So there the ad unit
is full-screen of some duration and then it gets
to all the challenges of the technology, of
measuring and reporting that. But at least when we’re
talking about the video ad unit, there’s a lot of
different kinds of video ad units, out-stream,
in-stream, pre-roll, mid-roll, can standardize
in some of that part of the inventory so that from
a media perspective, they know what they’re selling
and then, from a buyer perspective, there’s
some consistency and some expectations that, okay,
this is what we’re going to get when we buy. DR. WILBUR: I think
it’s a great question. The advertising industry
has tried on multiple occasions through some of
its collective research bodies to define terms
like engagement and has largely failed to do so. I think it’s sort of like
asking to compare the price of a general
contractor versus a carpenter or versus a
plumber, the latter two of which might be hired by
a general contractor. Everyone has a
different home. Everyone’s home has
different needs and everyone demands different
services for different reasons. And so, I am not
optimistic that we will find widely applicable
and widely useful common metrics due to the
heterogeneity in advertiser needs and the
way that they use these various advertising
formats. MR. PRITCHARD: Yeah, Carl. We can compare prices. But what we need to
compare is price per some other unit, such as
how many people are we reaching. And what we really are
trying to get is how many unique people
are we reaching. The biggest in advertising
today across every platform is the fact that
there’s too much waste because there’s too much
frequency and that is — why it’s a problem is not
just for the businesses. But it’s for the consumers
more than anything else. So we just want to know. We’ve asked many of our
providers just solve the problem of excess
ad frequency. Just solve it. That’s all we want, okay? Just give us a signal that
says if we reach Susan, we don’t reach her again
and again and again on multiple platforms. I think we’re closer than
we’ve ever been because I think that now some of
the digital providers are coming together and
recognizing that. And if we could just solve
that problem, then we’ll start worrying about some
of the other ones later. MR. WILLNER: I think my
next question is primarily for Pat, though I will ask
the other panelists also to react to it as well if
they — I’ve kind of — as I’ve been listening to
everyone speaking today, I’ve been puzzling about
how is it that at the same time we hear that
broadcast TV is facing a lot of these new
challenges from growing media, digital and other
competitive challenges, and yet at the same time
we’ve heard from Rick, and I think this was also
in Susan’s initial presentation, that
broadcast revenues are staying relatively stable. And we’ve even heard from
Marc that prices — from Marc Pritchard that prices
were going up in TV. And so, when I looked
at Gray’s press release announcing its results
for the first — the last three months of the first
year of 2018, I saw that they referred to record
results of operations, including strong net
income and record revenue in broadcast cash flow. And the annual report
indicated that overall revenues had risen
steadily overall for the past three years, while
even the broadcast advertising revenues
appeared to have been fairly stable, apart from
the cyclical political advertising fluctuations. So I wanted to ask
just Pat is that a fair description with what’s
happened with Gray and is it reflective of the
broadcast industry overall. And what accounts for
this phenomenon of greater competitive challenges
that you describe, while at the same time we’re
dealing with stable revenues and even, as Marc
referred to, some price increases as well. What is going on? What accounts for all of
these things happening at the same time? Are the CPMs, the cost
per thousand going up for broadcast TV and why do
you have these different phenomena operating? MR. LAPLATNEY: So I guess
I don’t know if I can make my answer as long
as your question. (Laughter.) MR. WILLNER:
I’ll grab — I’ll grab that. MR. LAPLATNEY: There was
quite — there was quite a bit there. I guess, so I would — I
think, and I don’t want to speak for Marc, but I
think Marc was referencing the national advertising
market when he was talking about the upfronts. We deal locally. So those are really
two different markets. I think that’s
the first point. As it relates to Gray and
our revenues, so on the advertising sales side
of our business, over the last few years, our core,
which is local — well, our core local is
essentially flat during odd years and down a few
points during even years and that’s due
to political. National business has been
down every year for the last few years. So it’s not — you know,
it’s down single digits. But it is down. Now again, as it relates
to — if you want to talk about Gray’s topline
revenue, you know, our revenue is also impacted
by retransmission revenues. And so, you know, that
has an impact on topline. And that offsets some
of the decrease in advertising revenue. And then, again, speaking
strictly for Gray, you know, we have cut costs
aggressively over the last few years. We’ve very publicly, you
know, publicly announced we dropped a rep firm. We no longer work with
Nielsen in diary markets. So we’ve a
aggressively cut costs. I think that accounts for
the numbers that you’re seeing. But the reality is
our core business is declining, you know,
unlike network television. Now, I can’t speak for
the entire industry. But that’s our experience. MR. WILLNER: Reactions
from others on the panel? MR. PRITCHARD: I will say
that, according to Media Dynamics, TV media for the
’18-’19 season went up 5.2 percent, 10 percent
increase in CPMs and with an 8 to 9 percent
viewership decline. And I think that is — and
largely the networks, the national networks, 75
percent of the media is bought upfront. Now, I believe there’s
— that’s a factor. I also think there’s a
factor in quality as well. I think people are willing
to pay a little bit more because of the quality
that comes from TV. And I think that’s one
of the advantages of TV. But I think that’s
still the upfront buying approach is one that
is a timeworn one that facilities the ability to
continue to have ratings declines and continue
to have prices go up. It’s kind of what we call
in the advertising world the funnel effect. I’d better get my money in
now or I won’t be able to get a decent price or I’ll
be subject to the scatter market, which, in my
experience, is always higher, so — which seems
odd, but it always is. And I think those are —
that’s just an effect that needs to change. I’m looking forward to the
future of programmatic. And frankly, I wish we
could spend more in local. I wish that we
had an ability. But local is so fragmented
to us right now, it’s just very difficult to be
able to make a good buy. The data is not there. The measurement
systems aren’t there. And the visibility for at
least marketers like us isn’t there. If we got that, I think
that would create some competition with the
national players, and that would be quite welcomed. MR. DUCEY: We’re
working on it. MR. LIEBERMAN: We
can talk after. (Laughter.) DR. WILBUR:
And I think one factor you can’t ignore is
the macro trends. So advertising spend in
the aggregate tends to be strongly pro-cyclical. And we’ve had a very
sustained economic expansion. So that’s a partial answer
to your question as well. MR. DUCEY: To comment on
what Pat was saying about the core revenue or the
broadcast revenue local, I mean, we’re calling that
relatively flat, a little bit of even year/odd
year bounce. But I mean flat is —
we’re doing nominal. That’s not adjusted. So it’s actually
declined for one thing. The other thing, we call
it foundational because it’s a part of a
lot of campaigns. We do surveys of
advertisers and advertisers that are on
cable and/or TV use an average of 13 media
platforms in their media mix. And they can reallocate
that spend across those different platforms,
unless there’s a competitive response from
other elements of that media mix, that pie split,
the share of the wallet I was talking about. So I think Pat was saying
that one response is we’re going to have to drop
rates to keep, you know, a piece of that buy that we
want to get from that Kia dealer or whatever the
advertiser is, so to kind of protect nominally
flat, but in real terms a declining share of
the ad spend market. There’s a couple of
different responses that media companies have
available for them to pursue. MR. WILLNER: I noticed
that on one of the Viamedia slides, there was
a presentation about the CPMs, the cost per
thousand, for regional and local and national cable. And the national was
showing as lower. And I understand from
Mark that was not just for Viamedia, but for the
industry as a whole, that national is around $15 to
$25 whereas the CPMs for the regional and local
cable were on the order of $20 to $30 each. Could you address, Mark,
what explains that and what is the — you know,
why are the regional and local higher than
the national? And also, are the regional
— are the CPMs for cable — how do they compare to
the CPMs for broadcast? And I’d be happy to hear
Pat address the question of the comparability
of the CPMs for the two industries as well. MR. LIEBERMAN: Well,
a couple of things. At the local cable level,
we typically don’t sell on a CPM basis to start. That’s how we back in to,
because your question was in preparation, what
are relative CPMs. But as I said in the
presentation, is that the more targeting —
principle of advertising, whether geo-targeting,
psychographic targeting, demographic targeting,
more targeting you can have. If I can reach that one
right person, I’m willing to pay a lot of money for
that right one person, that swing voter, the
mythical swing voter. So it just goes to say
that if you’re a national advertiser, because
there’s going to be inherent waste when you’re
a national advertiser, even though you’re
going to be able to do geo-targeting, because you
might buy a whole DMA, but you’re really just
interested in certain pockets, number one. So that’s why national
CPMs are going to be lower. Local and regional, even
though it said on the slide that they were
equivalent, it tends to skew. Logic would dictate that
the regional advertiser is going to be willing to
pay a higher CPM than national. But local is going
to pay a higher. So even though they were
both $20 to $30 CPMs, local ends up skewing
higher than regional in terms of CPMs. As to how it compares
to broadcast, it really depends. I think that Pat’s
example, which was a great one, of college football,
tends to be in much smaller markets where
there may not even really be an interconnect. It may be that an
advertiser, an auto dealer wants to reach
everybody in a DMA. And so, you’re not really
an ad zone advertiser there. So the ad zone advertiser
might be equivalent to the regional advertiser
because the market is that small. However, in bigger
markets, where there is an interconnect — and
interconnects were designed in effect
to compete against broadcasters by putting
all the different cable companies or MVPDs
together, is that in those markets is that you’re
selling at the zone level because you really want to
advertise only where the trading area is for that
particular merchant. And there again, you’re
willing to pay a higher CPM because you’re just
advertising in that particular zone. Smaller campaign
size, but higher CPM. MR. DUCEY: Just
a clarification. Higher CPM and we’re
talking about waste and targeting, but if you look
at the effective CPM for who you’re actually
delivering, the national CPM may look lower. But if you look at the
actual target audience that gets delivered,
that CPM may well be more expensive than what the
CPM — MR. LIEBERMAN: Yeah. In fact, there’s sometimes
a fake KPI called eCPM — MR. DUCEY: Right. MR. LIEBERMAN: — which
is the effective CM — MR. DUCEY: Right. MR. LIEBERMAN: — which is
a win-win on both sides. You’re willing to pay a
higher CPM, but for less viewers because it’s more
effective and you’re able to reach the
right audience. MR. DUCEY: Yeah. MR. WILLNER: Pat, do you
want to address the CPM issue too? MR. LAPLATNEY: No. No. In general, I
agree with Mark. I mean, you know, where
they can zone, the CPMs are going to be higher
because you can target better. It’s the same
thing in digital. If you’ve got, you know,
a tighter target you can hit, the CPMs are
going to go up. MR. WILLNER: And how do
the broadcast CPMs compare to what you’re
seeing in cable? MR. LAPLATNEY: So again,
I think it varies. So if — and I think you
have to look at different day parts, sports
versus sports. Probably pretty similar. You know, news, I would
guess — and I can’t say for certain, but, you
know, if it’s the highest rated station in the
market, probably a little higher than cable. If it’s a lower rated
station, it’s probably similar. You know, I think — I
think overall though the primary difference is that
cable has the targeting capabilities that
broadcast doesn’t. So that generally pushes
the CPMs up a little. MR. WILLNER: And that’s I
think a great segue into my next question which is
also initially for you, Pat, but also I think it
definitely impacts Mark. And I’ll be happy to hear
from others on it as well. A large part of the
Viamedia business related to what Mark had referred
to as the local-local business, this local zone
business which is sub-DMA, the ads that don’t reach
the whole DMA but are only targeted to geographic
portions of the DMA. Does Gray TV have any
plans at some point to sell advertising at
levels below the whole DMA throughout its markets,
going down to these kind of local zone levels or
something else below the DMA level? How far is that away
in terms of technology? What needs to be done to
get there and what are the obstacles? And is this something
— I know this is a long question too, but have you
— have you tried to do this — have you tried to
do this — have you tried to get into this area
before and run into obstacles in doing that? MR. LAPLATNEY: Yeah. So we actually do
sell some sub-DMAs. We probably take some
sub-DMA approaches. It’s generally markets
where we have multiple stations or multiple
transmitters where we can feed a different ad in. So, for instance, in Waco,
Texas, we have one in Waco, out of Bryan, a
College Station station. So we can do two
different ad feeds there. And we can compete better
with cable in those instances. But it’s only — I think
it’s Albany, Georgia, Waco, maybe Bismarck,
North Dakota and one other. As to the question — by
the way, we’d love to do it — we’d love to be able
to sell zone advertising. We’d love to, you know,
sell better targeted advertising
any way we can. Have we tried
to do it before? We’ve not — we’ve not
tried to — well, you know, is that a question
about have we approached cable operators about it? Because I think that’s
a legitimate question. And if there was a way
to put broadcast together with — you know, put
broadcast audience, combine that with cable’s
data, I think that’s a really interesting
combination. We haven’t had any really
serious conversation around that. But I think at some point,
maybe that conversation could be had. MR. LIEBERMAN: I would
just say, and you can comment, Pat, but there
are two big initiatives or two initiatives, if you
will, that would allow the broadcast industry to
start to more narrow cast and one is ATSC 3.0 —
MR. LAPLATNEY: Right. MR. LIEBERMAN: — where
you can start to do more addressable advertising
and addressable advertising is the
ultimate local, which is to be able to do it at the
household level, target the household level. And then, there was a
recent announcement of a consortium company called
OAR, I believe, or not — a consortium called OAR
— MR. LAPLATNEY: Yeah. MR. LIEBERMAN: — which
is open, addressable, whatever the R stands for. And there, they’re
using VIZIO-connected television, same way ATSC
3.0 works, with being able to actually send an ad to
an individual television. It sounds like there was
a project a few years ago called Project Canoe. It sounds like Canoe 2.0. MR. LAPLATNEY: Yes. Yeah. Yeah. MR. LIEBERMAN: And so,
it’s great that the industry is
trying to do this. One of the advantages of
course is that cable is there today. MR. DUCEY: The connected
TV, there are some broadcasters doing
targeting with that, connected TV data
like Mark is saying. So that’s not native only
to the broadcast platform. But they are doing
some targeting there. And then, the ATSC 3.0 is
the new platform that’s kind of Internet-friendly
for broadcasting that’s starting to be deployed at
scale over the next year or two. MR. LAPLATNEY: Yeah. That got lost in
the long question. My bad. So ATSC 3.0 does have
tons of promise for the broadcast industry. But the reality is
it’s a few years off. And it’s a pretty
significant investment as well. MR. PRITCHARD: If the
local broadcasters can do what you’re describing,
it would be quite welcome. We have data now
from retailers at the neighborhood level. And if we could
synchronize activities that we have going on
with retailers at the neighborhood level,
combined with media, that’s where it’s
not just price. It’s effectiveness
and that’s what’s most important is that we want
to ensure — we’re willing to pay a higher CPM if the
effectiveness is better and if we haven’t wasted a
bunch of money at the same time. So the reason why I think
sometimes CPMs are low is because it’s an attempt to
be able to get the price low enough to be able to
deal with the waste that you have. So greater precision with
technology will I think really make the local
industry very competitive. DR. WILBUR: I think that’s
a great point and it’s a broader trend where many
advertising organizations have — they now have
customer-level data that allow them to estimate
advertising returns on a lifetime value basis. So they have more clarity
on what the value to them could be if they’re able
to measure what the effect of the ads are. MR. PRITCHARD: Yeah. A lot of our digital is
shifting to that as well. It’s performance-based. It’s not just the ads. I mean, an ad on a social
media — the average view time for a social media
ad is 1.7 seconds. It’s kind of hard to get
a message across in 1.7 seconds that’s going
to deal with anything. So what we do is target
people in a way to try to get performance
coming out of there. If we do the same thing
with TV, it will be far more lucrative. MR. WILLNER: Marc, you had
alluded before to some of the experiences you had
with the efficacy of digital advertising
against TV. And, you know, looking
back at some of the things you’ve said before, I came
across some interesting quotes. You had been quoted as
saying, “We moved $100 million out of digital and
sales went up 2 percent.” I also came across
another quote. You had observed at one
point, “We are returning to traditional television
for the bulk of our spending. Television has a unique
relationship with its viewers, providing
engagement and trust between the viewer and the
brand.” I’d like to just ask you if you could add
some context to those remarks, tell about what
happened that led to those experiences. MR. PRITCHARD: Yeah. About 10 years ago, when
the digital advertising industry was really
starting to get going, we helped fuel that. We helped monetize many
of these platforms through digital. We spent, you know, maybe
5 or 6 percent in digital and then ramped up to well
over more than a third. And as we started looking
at the amount of spending in digital media, we were
starting to wonder why we weren’t getting
effectiveness. So we went to the
different platforms and said we’d like some more
data to help us figure out, you know,
what’s going on. Are people
viewing these ads? How many people
are we reaching? Are we reaching the right
people, not bots and whatnot? And they didn’t
have the data. And so, we pretty much
put a demand out for it. And it took nine or 10
months to make little progress. So then, that’s when I
went public and I asked for the different
standards that we called for, which is the one
viewability standard, the third party audience
verification, antifraud and brand safety. Then when we got the data,
we realized it wasn’t really all that it
was cracked up to be. That’s when we realized
that the average view time was 1.7 seconds. Then when we viewed that,
we were reaching people over and over again when
we started getting into the verified data, that we
had a high degree of bots that were seeing
our ads, not people. And then, we recognized
that some of our ads were showing up really in very
bad content because it was across the entire
video platform. So we actually moved off
of YouTube for a full year because we couldn’t feel
— we didn’t feel that the advertising — our
brands were safe. They changed. I give them a
lot of credit. They changed. They reduced the number
of channels that they monetized on. So we think it’s
a lot better now. Now we actually have what
we call safe lists, only advertise on those
channels that are absolutely safe, similar
to what we have on TV. And then we’re pivoting to
performance marketing on other platforms. But the bottom line was is
that we just found once we got transparency, it
helped us make better decisions. And then, we fell back
in love with TV, started looking at the TV
opportunities and started finding different ways
to be able to assess the degree to which
people are engaged. And we’re measuring
engagement and measuring sales lists. One of the things you guys
might want to understand is we test DMA to
DMA many times. We’re doing A/B testing
all the time to figure out what the benefit is
of different mixes. And what that provides us
is more confidence that we can invest. MR. WILLNER: Any reaction
to that from the rest of the panel? In particular, I know Pat
and Mark Lieberman, your companies have both gone
to some extent into the digital businesses or OTT
or other activities as well. I mean, you haven’t
limited yourselves to linear TV. And so, I wanted to ask
you, you know, how you see those two businesses as
fitting together for your companies, the traditional
linear business and whatever you’re doing
in the digital areas and whatever — how you’ve
dealt with the kind of problems that Marc
Pritchard was talking about. MR. LIEBERMAN: Yeah, and
our philosophy in going after digital is — showed
the growth on digital, is the Willy Loman line, “Go
where the money is.” And so, we, as I said, trained
our salesforce on how to sell other media types,
display, OTT, pre-roll video, mid-roll video,
social, email, et cetera. And so, we really see
that as reach extension. And so, on the TV side —
and our TV has been flat to slightly up. But digital is growing
rapidly — is when we advertise on television,
we can only advertise in those households where our
MVPDs have subscribers. On digital, we’re
not so limited. So we can extend the reach
to a whole DMA, to a whole state, to other markets. We can do the same thing
demographically where we can use digital as
reach extension. So we really see it
as a complement, not a substitute. And the same
thing with OTT. OTT was a shiny object
certainly during March Madness where it was huge
and the CPMs were out of control. They were $50
to $200 CPMs. And so, we really see it
as a complement with TV as the anchor and foundation. MR. LAPLATNEY: So digital
is critically — growth in digital advertising and
advertising revenues is critical to our company. You know, we — like
Mark’s company — set up a training center in
Charlotte, North Carolina and trained people
on selling digital. So we take it
quite seriously. We have created different
approaches for different categories. So we have a digital
approach for the health category, a digital
approach for the auto category. And essentially, I mean,
you know, the Willy Loman thing, that’s where the
growth in audience and that’s where the growth in
revenue are going to come from going forward. MR. DUCEY: Just to put
a number on that, in the research we’re doing
with local stations as an example, they sell linear
TV, like Pat was showing in his example. But they sell dozens and
dozens of other products. That could be print. It could be digital. It could be
experiential-based, like events. So there’s a lot that goes
into that mix from local cable operators
and local stations. MR. LIEBERMAN: Yeah. I wanted to add one other
thing to Pat’s point as well is that with our 200
sellers in the field and the 6,000 advertisers,
that’s a personal relationship. So it may not necessarily
be high touch, but it’s touch. It’s touch. And so, for 10, 20 years,
that account executive has a relationship, a trusted
relationship with that advertiser. So when that advertiser
says how can I re-extend the reach here, what
should I do, what are some of the other tactics,
they’re going to go to that trusted
account executive. And the bigger players,
the Facebooks and the Googles of the world, when
they’re going after local, they’re doing it on
an automated basis. So there’s no trusted
relationship there. Great platforms, but
there’s not a trusted relationship, which
is always going to be important. It’s going to be that
people, process and technology. You’re not going to get
rid of the people side at the local level. MR. PRITCHARD: We found
some interesting things about digital, that it has
completely changed how we advertise. But what’s interesting
is we find it’s most effective to create
content that goes on digital that is sometimes
long-form content, like four-and-a-half, five
minutes or two to three minutes that then people
see and then they go search for our brands. And in many cases,
“#LikeAGirl” is a classic example of a
three-and-a-half minute video, “Always
#LikeAGirl,” that went on the Internet, ended up
getting 100 million views. Then we put it on the
Super Bowl and then it added even more to that
business and the continued to get more
and more viral. So that’s what’s been good
about the digital world. What’s been bad about the
digital world is some of the content quality
issues, some of the editorial comments and
the lack of transparency. So once again, looking for
that with every provider. MR. WILLNER: Well Ken, I
wanted to address the next question to you since
it’s really coming from an academic perspective. I wanted to know if you
could address what type of academic or scholarly
research has been done on this question of
substitutability of various forms of media
advertising between broadcast TV, cable
and digital online. And what studies are
considered to be the most robust? And to the extent that
there aren’t many reliable studies, why is that? DR. WILBUR: So, you know,
when you think about the introduction of digital
advertising media, sometimes people think
Google created the search engine marketing
platform from nothing. Actually in the ’90s,
Yellow Pages executives used to compare the
profitability of their industry to
narco-trafficking, saying it was number
two in the world. They don’t make
those claims anymore. That’s — we’re doing
much more with our search engines than we ever
did with our phone directories. But, you know, I think
it’s important to recognize that people
have always used different media for
different purposes. And there are many markets
where finding information on demand is the
determinative factor of who someone goes to
for their business. There was one study
published comparing television advertising
effectiveness to YouTube video advertising
effectiveness. And what they found was
that the populations of viewers reached by each
of those two media were different in some
important ways. However, once you adjusted
for those differences, they no longer found
variation in how effective the two advertising
media were. As to why there is not
more academic research on this topic, I don’t know
that anyone has frequently expressed that interest. But if it seems like it
would be useful, I would encourage you to tell that
to academics and maybe even — you know, if you
really want them to do it, put just a little bit of
money behind it and you’ll probably get those
questions answered in more detail than you
would probably like. MR. WILLNER:
Who would guess? Also Ken, I think in
your 2008 article on a two-sided empirical model
of television advertising and viewing markets, you
had referred to television as the single most
important advertising medium. Does that remain
true today? We’ve heard from a number
of the other panelists about TV continuing
to be foundational. I’d like to ask you to
address your reactions to what’s changed over
the past 10 years. And is it still
the king today? And to what — and what
does the concept of foundational mean, as the
various panelists have addressed it? MR. LIEBERMAN: Before you
answer the question, do you want me to offer to
hire you and then you can try to answer it? DR. WILBUR: Yes. Quo pro quid. So I think it’s much less
clear what is the single most important
medium these days. I think more and more
people are consuming multiple media
simultaneously. And, you know, it always
depended on the nature of the business and the
nature of consumer demand. I think in many markets,
search is the bottleneck. And if you are not at the
top of the search results, especially in the top two
or three, you just won’t be considered. But for building brands,
I think it’s hard to substitute for — at least
for video advertising. I may be in the least good
position to speculate on this on the table. So I’d rather hear others. MR. PRITCHARD: Well, I
would say that TV is still critical. It is foundational. I think your point about
video is absolutely correct. I think that more and more
now, what we’re seeing is that people, consumers
want — are willing to and want to engage in really
high quality video content from a brand. We’ve got a brand called
SK2, which is a premium skin care brand right
now, that has put out four-and-a-half minute
epic stories that then people go to search. They put up — right now,
they have a two-minute video series called “Bare
Skin Chat” with James Corden and Chloe Grace
Moretz, which is just — it’s hilarious. And it’s all —
it’s all video. It’s all online. Had an 86 percent
increase in search. And that’s their model, is
to be able to get people engaged, get people to
search and then kind of move down
towards purchase. And so, I don’t — I
think your point is right. And I don’t think it’s any
one medium is going to be, you know, the
king or the queen. And it really depends by
brand and it depends by country. We spend, you know,
a third of our media spending in digital right
now and 60 percent in TV in the United States. We spent 80 percent of our
media in digital in China. We spend, you know, more
like 80 percent of our TV — or sorry, our ads on TV
on Tide and we spent, you know, more like 80 percent
of our spending on Olay is in digital. So it just depends. And it depends on
reach, effectiveness and efficiency. MR. LIEBERMAN: Carl, I did
at one point, I wanted to mentioned when I joined as
CEO of the company, I did a ride-along with my
account executives in different markets. And what’s really
refreshing about these local merchants is they
kind of know what the ROI is. Ad runs, more people
come into their auto dealership. And as the industry moves
to more data-driven and programmatic, as an
example, television, the reality is that local
merchant may want to do what’s called — still
wants to do what’s called fixed position. I want to be on Monday
night football because what I’m going to do is
I’m going to bring my family and friends over
to see my ad on the big screen with me with
the balloons and so on. I’m talking about
an auto dealer ad. That’s not going
to go away. And that’s
really important. And they’re not going to
gather around a computer to see the video ad
that’s on a computer. They’re always going
to gather around the television. And that’s just the
way local works. And that’s those
120,000 local merchants. MR. WILLNER: I think,
Mark, you had also — in your presentation, you had
also made the point that digital is not
cannibalizing TV ad dollars. And I think your source
for that, you referred to a MoffettNathanson report. It’s the April 5,
2019 report on “U.S. Advertising: A
False Dichotomy?” Could you explain why that
was the case, why that cannibalizing was
not taking place? And then, I’ll invite
others to react. MR. LIEBERMAN: Well, I
think in their report, they talked about it as
showing macroeconomic trends on television and
the trends on digital. And I’ve forgotten
who it was. It may have been Susan
or Ken who showed the reduction in direct mail,
as Rick mentioned before, the reduction in
newspaper and print. So as a macroeconomic
trend, I think that was the basis for
their thesis. But we’re also
seeing that. We monitor account level
spend by actual advertiser and our KPIs are always
to try and maintain television, but
upsell them digital. And I’ll give you
a great example. In the auto dealer
category, we might be — it might be $100,000 ad
campaign and we might be 10, 20 percent
of the market. We want to be able
to maintain that. And in the past, our
account executives would see digital as just an
add-on, let’s say an extra 10 percent. But if they’re spending
half their spend now on digital, what we’ll do is
we’ll say we have a much bigger total
addressable market. So let’s keep
television flat. But let’s go after their
entire digital budget, in competition with others
in the local markets. And so, anecdotally, some
of it may cannibalize. But as a whole, I don’t
think it’s cannibalizing yet at least. MR. WILLNER:
Reactions from others? MR. LAPLATNEY: Yeah, I
would just maybe — so I would essentially agree
with what you said. I think there could be
modest cannibalization. But I don’t think it’s
material at this point. MR. WILLNER: Yes. I think you were signaling
me, Erica, and I should be rightly — I was keeping
my mind on it also, that there is a — we did
promise a few moments for audience questions. And so, though there are
many more things I could ask all of these
distinguished gentlemen, I mean, there’s a vast
amount of knowledge they have to share, let me give
all of you a chance to — if there are questions
from the audience for this panel. If there is anyone,
just please stand up. Ask your question and I’ll
try to repeat it so that the — everyone
can get that. Is there any question
from the audience? MR. LIEBERMAN: Well, you
can fill in the blanks with one of your
other questions. MR. WILLNER: Well,
I’m not seeing any. And I realize —
MR. LAPLATNEY: Oh, there’s one. MR. WILLNER: Oh, I do
see one in the back. There we are. Please step forward. Yes? AUDIENCE MEMBER: Sure. For digital advertising,
are platforms like Google and Facebook dominant in
the market or do you see them as being on a more
competitive playing field? MR. WILLNER: Did
everyone hear that? So for digital
advertising, are platforms like Google and Amazon
dominant in the market or is it more of a
level playing field? MR. DUCEY: So actually, I
shared some data on that for Washington. Oh, sorry. Go ahead, Pat. MR. LAPLATNEY: No. I was going to say,
Bill, over to you. MR. DUCEY: Okay. MR. LAPLATNEY: Yeah. You’ve got the data. Yeah. MR. DUCEY: Okay. Well yeah, in Washington,
D.C., one of the data points that I showed was
that in fact Google is one brand I called out. They make more in
advertising revenue in this market and typically
in most markets than all of the local TV
stations combined. So they’re clearly
a dominant ad brand throughout all markets. DR. WILBUR: I’ve seen some
estimates that in recent years they have captured
more than 100 percent of the total growth in
digital advertising expenditure. And if that sound
impossible, the way to understand it is that they
are not only grabbing all of the new dollars,
but they are taking a substantial amount of
business from other digital platforms such
as Yahoo, which are on a substantial
declining path. MR. WILLNER: Thank you. Are there any other
audience questions for the panel? One in the back? AUDIENCE MEMBER: So I come
from a digital advertising market, and I’m just
curious, since we’re in the DOJ Antitrust
Division, are there any thoughts of taking the
findings of today and the competition you guys have
been talking about, which is actually, as discussed,
a very small part of the market and looking at
Google and Facebook and considering how that
dominance is affecting the rest of the industry. MR. WILLNER: Well, I think
the broader questions for what DOJ will do, I will
have to defer to the assistant attorney general
and others who will make those decisions
rather than me. But I think — AUDIENCE
MEMBER: Well, but they’re here listening. MR. WILLNER: Yes, but I
— but we are — we are certainly listening. And that’s of course part
of the purpose of what we’re doing here is to
gather this information and to be in a position to
address these issues going forward as they come up. If there are no other
questions for the panel, I will ask you to join me
all in thanking these panelists. (Applause.)

About the Author: Garret Beatty

Leave a Reply

Your email address will not be published. Required fields are marked *