INTERNET ADVERTISING AND THE HIERARCHY
OF EFFECTS
Regina P. Schlee
School of Business and Economics, Seattle Pacific University, 3307 Third Avenue West, Seattle, U.S.A.
Anthony Schlee III
Customer Insight Group, Avenue A/Razorfish, 821 Second Avenue, Suite 1800, Seattle, U.S.A.
Keywords: Internet advertising, hierarchy of effects.
Abstract: Internet advertising has been considered by many advertisers as a medium of direct response. Though this
classification was appropriate in the early years of internet advertising when consumers were expected to
click on 468x60 banners, current technology and high bandwidths now allow much greater creative
flexibility. Nevertheless, search engine advertising continues to take the largest share of online advertising
budgets. The challenge for internet advertising is to be able to take on additional roles in the hierarchy of
effects model of communications besides the call to action. This study examines data collected in two cases
that document the role of online display ads on consumer behavior. Online advertising can
take on a variety of roles: brand building, creating consumer interest, as well as calling for a direct
response. The results of this analysis are exploratory but point to the need of considering different forms of
internet advertising as serving a variety of functions in hierarchy of effects models of communications.
1 INTRODUCTION
2005 was a banner year for internet advertising. The
Interactive Advertising Bureau (IAB) in cooperation
with PricewaterhouseCoopers estimates that internet
advertising revenues exceeded $12.5 billion in 2005
(IAB, 2006). This represents over a 30% increase
from the previous record of $9.6 billion in internet
advertising revenues in 2004. Almost three quarters
of U.S. consumers now have access to the internet,
and the proportion with broadband connections is
increasing at a rapid pace (Pew, 2006). According
to a March 2006 report by the Pew Internet &
American Life Project, within 4 years the proportion
of adult internet users with broadband connections
went from 10% to 37% (Horrigan, 2006). Internet
advertising has experienced a similar transformation.
According to Nielsen/ NetRatings AdRelevance,
traditional 468x60 banner ads have experienced a
decline of 50% from the first quarter of 2001 to the
fourth quarter of 2004, while leader boards (larger
banners that allow more creative flexibility) have
increased by 552% (Burke, 2005). However, in
spite of increased creative content, internet
advertising continues to be evaluated using some of
the standards that were developed ten years ago
when all online ads were viewed as a medium of
direct response and click through rates were seen the
primary indicator of effectiveness.
It is understandable why, as a new advertising
medium in the 1990s, internet advertising needed to
prove its effectiveness. Internet advertising agencies
prided themselves in providing more accountability
than traditional advertising, so click-through rates
(CTR) and conversions were seen as the most
appropriate indicators of effectiveness (Hollis,
2005). However, as the novelty of banner
advertising wore off, CTR declined (Drèze and
Hussherr, 2003). The decline in CTR was, in fact,
precipitous. In 1996, click through rates averaged
about 7%, but by 2002 CTR had dropped to about
0.7% (DoubleClick, 2003). As a result of a growing
disenchantment with internet advertising, online
advertising revenues declined in 2001 and 2002
(Hollis, 2005). Fortunately for internet content
providers and internet advertising agencies, this
decline was short lived. However, companies
222
P. Schlee R. and Schlee III A. (2006).
INTERNET ADVERTISING AND THE HIERARCHY OF EFFECTS.
In Proceedings of the International Conference on e-Business, pages 222-226
DOI: 10.5220/0001426402220226
Copyright
c
SciTePress
advertising on the internet continue to demand that
every dollar spent on advertising online brings the
anticipated result. This can be contrasted to
traditional offline advertising which often focuses on
brand building and is not expected to produce
immediate results. Students of advertising are
familiar with the famous John Wannamaker quote,
“I know I am wasting 50% of my marketing budget
… my trouble is that I don’t know which 50%.”
Internet advertising is expected to produce
immediate and measurable results and is still seen by
many as not being conducive to brand building.
This greater demand for accountability is best
demonstrated through the popularity of search
engine advertising. According to the Interactive
Advertising Bureau (IAB), search marketing
accounts for 40% of online advertising budgets,
compared to only 20% of online budgets spent for
display ads (Bruner, 2005). The popularity of
Google with advertisers reflects the desire to pay
only for demonstrated results. Advertisers
determine how much they want to pay when a
consumer clicks on a search keyword and bid
accordingly (Taylor, 2004). Of course, Google is
not the only player in search engine marketing as
Yahoo, AOL, and MSN have entered this lucrative
field (Acohido, 2003).
But, are the indicators of advertising
effectiveness that were developed for internet
advertising ten years ago still the most appropriate
measures of effectiveness, or can internet advertising
be viewed as providing some of the same influences
on consumer attitudes as traditional, offline
advertising? In other words, is all internet
advertising still a direct response medium, or can
different forms of internet advertising provide
different effects on the behavior of online
consumers? This study examines two case studies
that provide preliminary information on how online
advertising functions in the context of search
engines, as well as in the context of traditional
offline media. The first case analyzes the
contribution of online display ads for an advertiser
on consumers’ likelihood to use a search engine to
get additional information about the advertiser’s
service. The second case examines the contribution
of online display advertising in an environment
where consumers were also exposed to offline
media. Results from these two cases will be
interpreted using the hierarchy of effects model
examining the effect of different media on
consumers’ purchasing behavior.
2 LITERATURE REVIEW
Advertisers have long known that people generally
do not make spontaneous decisions when buying
products, but need to be taken through as series of
steps that have been called hierarchy of effects. The
logic of hierarchy of effects models is simple.
Consumers must first become aware of a product or
service before they buy. Then they have to get some
information about the product or service to develop
interest and a desire to buy. Barry (1987) reports
that the first complete model of hierarchy of effects
was developed by St. Elmo Lewis in the very early
years of the 20th century and contained four stages:
attention, interest, desire, action (AIDA). Barry
cites 38 elaborations of the original hierarchy of
effects model developed by Lewis, including the
widely cited DAGMAR model (Defining
Advertising Goals for Measured Advertising
Results) developed by Colley (1961) that aims to
measure the effects of advertising as consumers
move from awareness, to comprehension,
conviction, and action. Hierarchy of effects models
are discussed in almost all marketing and advertising
textbooks (Belch and Belch, 2006; Clow and Baack,
2004; Kotler and Keller, 2006). Nevertheless, there
is no discussion in those textbooks as to how internet
advertising fits within such models. And, given the
variety of internet advertising options available, do
all internet ads function in the same fashion, or do
they have different effects in the consumer’s
decision process?
Most advertising and marketing textbook also
discuss the concept of integrated marketing
communications (IMC). Integrated marketing
communications models work by having different
media working simultaneously on consumers to
produce the desired effect. For example, television
and magazine ads can create an interest in the
category, while sales promotions such as coupons
can create the call to action by giving consumers an
opportunity to “buy now.” Interestingly, at that time
when IMC models were first generated, consumer
search was conducted through telephone directories
such as the yellow pages which were viewed as
directional media, the kind of media one went to
after the decision to purchase a product had been
made. For marketers and advertisers in the early part
of the 21st century, there is little existing research on
how internet advertising complements other
advertising media.
It should be noted that hierarch of effects
models are not without their critics. Most criticisms
of hierarchy of effects models of advertising focus
INTERNET ADVERTISING AND THE HIERARCHY OF EFFECTS
223
on the possibility that consumers can take action
without going through a temporal sequence as
outlined in the AIDA and DAGMAR models
(Vakratsas and Ambler, 1999). A consumer’s
involvement level with the product category can, in
fact, influence how he or she processes advertising
information. There are situations where consumers
act with only a small amount of information about a
product or service and then develop attitudes about
their experiences. Behavioral influence and
experiential perspectives in consumer behavior
articulate different ways that consumers engage in
various actions and develop attitudes about the
products and services they have consumed
(Solomon, 2004). Critics of the model also claim
that the concept of integrated marketing
communications is antithetical to hierarchy of
effects models because all company communications
have to be directed to selling one idea (Weilbacher,
2001). However, hierarchy models do not imply
that the advertising be used to communicate a
different message at different stages, but rather than
the message be adapted to a consumer’s stage in the
decision making process. After the consumer has
obtained all the necessary information about the
product or service, it may be both unnecessary and
counterproductive to continue repeating the same
message in the same manner. Thus, realizing the
consumers use different modes of information at
different stages in the decision process helps
advertisers communicate more effectively and
efficiently.
3 RESEARCH FINDINGS
The object of this research is to examine how online
display ads can work together with other forms of
advertising to move consumer along the decision
process. Does display advertising providing
information about an advertiser’s product or service
result in a higher proportion of consumers who
request additional information through search
engines? In other words, does display advertising
lift the effectiveness of search engine marketing?
Our analysis of data in Case 1 seeks to answer this
question. Case 2 focuses on how display advertising
interacts with traditional offline media to facilitate
conversions (taking action) at the advertiser’s site.
Both sets of analysis seek to explore the manner in
which online display advertising acts as part of an
integrated marketing communications plan. Display
advertising can act both as an antecedent to
information gathering, as well as the call to action.
3.1 Case 1 – Effect of Display Media
on Search
One of the most difficult areas in advertising is the
ability to attribute buyer behavior on a specific form
of advertising. Earlier, we discussed the popularity
of search engine advertising through keywords. But,
how does a consumer get the idea to search for a
specific keyword? We can assume that consumers
decide to search for a specific product or service, or
for a specific company or brand because of another
stimulus. Hierarchy of effects models suggest that
another, earlier, stimulus created the awareness and
stimulated the desire for action. But, would display
advertising on the internet result in greater interest in
a specific company or brand? Our analysis of Case
1 seeks to examine how exposure to a retailer’s
display ad affected consumers’ usage of specific
search keywords. All data for this analysis were
provided by a global online advertising agency.
To test the effect of display ads on consumer’s
search behavior, we utilized an experimental design
whereby a segment of visitors to certain web sites
during 2005 were exposed to a display ad for an
online retailer (experimental/test group). Another
set of visitors where shown a public service ad at a
matched set of online websites. Cookie data allowed
us to track the behavior of internet users who were
exposed to the retailer’s ad and to compare their
search activities with the behavior of internet users
who were exposed to the public service ad (our
control group). Only individuals who allowed
cookies on their computer were included in our
experiment. Once an individual deleted or
“crunched” their cookies, he or she was taken out of
the study as we could no longer assign him or her to
the control or test (experimental) group. The data
for this study are presented in Table 1.
In total, 12% of individuals who were exposed to
the web media followed up with a search to obtain
more information about this retailer’s product. This
compares to 10% of the individuals in the control
group who were not exposed to the online media.
Though the difference between 12% and 10% in
search activity may appear small, it represents a 23%
higher response for those exposed to web media.
This analysis demonstrates that it is possible to
measure the independent effect of display
advertising on consumer behavior. Furthermore, it
demonstrates that display advertising on the web had
a substantial added effect to all the other media used
by that advertiser and that it preceded in the
hierarchy of effects the use search engine
advertising.
ICE-B 2006 - INTERNATIONAL CONFERENCE ON E-BUSINESS
224
Table 1: Effect of web media on search.
3.2 Case 2 – Effect of TV
Advertising and Display Media
on Conversions
Hierarchy of effects models assume that some
advertising generates awareness, while other types
of advertising generate interest and comprehension,
or a call for action. In situations where an advertiser
uses broadcast advertising to generate an emotional
response, can display ads on the internet increase
conversions at the client’s website through a call to
action?
To test the effectiveness of online media
compared to offline media (TV and radio), we
examined the effect of exposure to display ads under
three conditions: simultaneous exposure to offline
media in large metropolitan areas (primary DMAs),
smaller metropolitan areas (secondary DMAs), and
areas of the country where the client was not using
any offline media. The same methodology described
in Case 1 was used. A percentage of site visitors
were assigned to the experimental group and thus
were exposed to the advertiser’s display ad. The
control group was composed of visitors to the same
web sites who were shown a public interest ad. To
examine the effect of the online display ads on
consumer purchasing behavior, we tracked cookies
of those who were exposed to the display ad who
ordered the client’s product/service at the client’s
web site. The results of this research analysis are
presented in Table 2.
The findings of this study demonstrate a
substantial positive effect for online display media.
It should be noted that the effect of online media is
larger when combined with offline media. In
primary DMAs where the client used the heaviest
weight for offline media, the effect of online media
was largest. The online display media resulted in an
increase in conversions at the client’s website by
79.9%. In secondary markets where the weight of
the offline campaign was lower, the online display
media contributed to an increase in sales at the
client’s website by 48.1%. Where no offline media
were used, the lift for the online campaign compared
to the test group was 39.3%. These findings support
the integrated marketing communications model
discussed earlier whereby different elements of the
promotion mix are able to move consumers through
the hierarchy of effects leading to a sale.
Table 2: Effect of TV advertising and online display
media on conversions.
4 SUMMARY AND DISCUSSION
The two cases discussed in this study indicate that
online display ads can have a significant effect on
consumers’ online search and purchase activities.
They also indicate that display advertising online
can function differently than what is expected from a
simple direct response model. In Case 1, display ads
provided a substantial lift to the direct response
model of internet search marketing. The display ads
function in a manner consistent with the brand
building activities of providing interest and
comprehension in hierarchy of effects models of
consumer communications. In Case 2, the display
ads interacted with traditional offline advertising to
increase conversions at the retailer’s website.
Though the specific mechanism whereby
display ads affected consumers’ decision processes
has not been articulated yet, it is apparent that more
research needs to be conducted in this area. As the
penetration of broadband access to the internet
increases, so will the creative flexibility of internet
advertising. Future research needs to examine how
different forms of internet advertising can facilitate
the movement of consumers from the first stage of
the hierarchy of effects to the stage of placing an
order. Attention needs to be devoted to research that
measures the interactions between exposure to
different types of online media and offline media.
As advertisers become increasingly concerned with
the escalating cost and audience fragmentation of
offline media (especially TV advertising), internet
advertising will provide the means for maximizing
the effectiveness of integrated marketing
communications.
Saw web media Did not see web media Total
Search and no conversion 228,750 2,385,800 2,614,550
Search and converted 32,500 267,800 300,300
Total searches 261,250 2,653,600 2,914,850
Search conversion rate 12% 10% 10%
Primary DMA
Offline and online
Secondary DMA
Off line and
online
No Offline Ads Total
Test 14.9% 15.9 13.6 14.8
Control 8.3 10.8 9.7 9.1
Lift 79.9 48.1 39.3 61.6
INTERNET ADVERTISING AND THE HIERARCHY OF EFFECTS
225
REFERENCES
Acohido, B., 2003. AOL reels in search engine
Singingfish. USA Today, Nov. 11, 2003.
Barry, T. E., 1987. The development of the hierarchy of
effects: An historical perspective. Current Issues and
Research in Advertising, 10 (2): 251-295.
Belch, G. E., and M. A. Belch, 2006. Advertising and
Promotion: An Integrated Marketing Communications
Perspective, 6th ed. New York: McGraw-Hill/Irwin.
Bruner, R. E., 2005. The Decade in Online Advertising,
1994-2004. DoubleClick, April 2005, accessed at:
http://www.doubleclick.com/us/knowledge_central/do
cuments/RESEARCH/dc_decaderinonline_0504.pdf
Clow, K.E. and D. Baack, 2004. Integrated Advertising,
Promotion, and Marketing Communications, 2nd ed.
New Jersey: Pearson/Prentice Hall.
Colley, R. H., 1961. Defining Advertising Goals for
Measured Advertising Results. New York:
Association of National Advertisers.
DoubleClick, 2003. DoubleClick 2002 Full-Year Ad
Serving Trends.
Drèze, X. and F.X. Hussherr, 2003. Internet advertising: Is
anybody watching? Journal of Interactive Marketing,
17 (4): 8-23.
Hollis, N., 2005. Ten years of learning on how online
advertising builds brands. Journal of Advertising
Research, 45 (2): 255-268.
Horrigan, J. B., 2006. For many home broadband users,
the internet is the primary news source. Pew Internet
& American Life Project (22 March 2006), accessed
at:
http://www.pewinternet.org/pdfs/PIP_News.and.Broad
band.pdf
Interactive Advertising Bureau, 2006. Internet advertising
revenues estimated to exceed $12.5 billion for full
year 2005. IAB Press Release (March 1, 2006),
http://www.iab.net/news/pr_2006_03_01.asp
Kotler, P., and K.L. Keller, 2006. Marketing
Management, 12th ed. New Jersey: Pearson/Prentice
Hall.
Pew Internet & American Life Project, 2006. 73% of
Americans go online. Accessed at
http://207.21.232.103/press_release.asp?r=127.
Solomon, M. R., 2004. Consumer Behavior, 6th ed. New
Jersey: Pearson/Prentice Hall.
Taylor, C. P., 2004. Engine of change. Adweek, March 15,
2005: 18-20.
Weilbacher, W. M., 2001. Point of view: Does advertising
cause a “hierarchy of effects?” Journal of Advertising
Research, 41 (6): 19-26.
ICE-B 2006 - INTERNATIONAL CONFERENCE ON E-BUSINESS
226
