Content uploaded by Isar Kiani
Author content
All content in this area was uploaded by Isar Kiani on Feb 11, 2017
Content may be subject to copyright.
DOI: 10.2501/JAR-53-4-431-443 December 2013 JOURNAL OF ADVERTISING RESEARCH 431
INTRODUCTION
Cross-channel advertising has grown steadily and
significantly as a means to reach consumers. Tele-
vision, the Internet, and other channels are used
together to market products. Search engines have
changed the way people look for information.
Online advertising is growing rapidly and taking
budgets away from traditional channels.
The Internet does not exist in isolation, however,
and discussion around its expansion should not
neglect the roles other channels should play. Coin-
cidental with the growth in digital ecosystem, many
marketing paradigms are shifting from passive strat-
egies in communicating with consumers to more
proactive ones with engagement in multiple commu-
nication channels (Briggs, Krishnan, and Borin, 2005).
Although many marketers are giving more seri-
ous thought to online advertising as an option, it has
yet to reach its full potential within their marketing
mix. As a result, paid listings using search engines
often are not the first element of that media mix.
Industry reports have shown that main broad-
cast channels still garner the greatest share of
advertising revenue, but many companies have
started to allocate a larger portion to search-engine
marketing (ZenithOptimedia, 2013). In September
2013, ZenithOptimedia forecast paid search to grow
at an average of 15 percent per year to 2015, “driven
by continued innovation from the search engines,
including the display of richer product information
and images within ads, better localisation of search
results, and mobile ad enhancements like click-to-
call and geo-targeting.” Meanwhile, growth in the
use of search engines by consumers provides an
even greater incentive for companies to reconsider
their advertising budgets. In March 2009, there were
•
This study tracks the effects of adver tising expenditures in different media outlets on
subsequent consumer online search behavior for advertised products.
•
The data are from a large telecommunications company compiled over 78 weeks.
•
Findings suggest that exposure to advertising on different media outlets increases the
likelihood of follow-up search by individuals. Radio is less effective than television and online
impressions in generating follow-up search.
•
For advertisers, the short-term effect in subsequent search stresses the need to synchronize
online and ofine advertising effor ts to achieve the highest impact.
Effects of Multi-Channel Marketing on
Consumers’ Online Search Behavior
The Power of Multiple Points of Connection
MICHEL LAROCHE
John Molson School of
Business, Concordia
University
laroche@
jmsb.concordia.ca
ISAR KIANI
John Molson School of
Business, Concordia
University
i_kiani@
jmsb.concordia.ca
NECTARIOS
ECONOMAKIS
Google Montreal
neconomakis@gmail.com
MARIE-ODILE RICHARD
Independent researcher
odile10@hotmail.com
Amid the plethora of research on adver tising effectiveness, the authors of the
current study believe consumers’ online search behavior, subsequent to exposure to
traditional advertising messages, has been understudied. Using data from a major
telecommunication company, this study’s ndings support the inuence of employing
multiple channels, advertising expenditures, and television and online advertising on
consumers’ tendency to follow through with their own online investigations.
432 JOURNAL OF ADVERTISING RESEARCH December 2013
EFFECTS OF MULTI-CHANNEL MARKETING ON CONSUMERS’ ONLINE SEARCH BEHAVIOR
3.2 billion searches conducted across all
search engines, an average of 131 searches
per searcher per month.1 By 2012, Google
alone accounted for 5.1 billion searches per
day worldwide.2
Even as the effectiveness of cross-channel
advertising has been a primary focus for
many researchers and marketing practition-
ers, the authors of the current paper believe
that the manner in which traditional media
channels affect new-media use largely
has been overlooked. This study aims to
determine the effects of advertising and its
impact on consumer search.
One particular area of interest is the pro-
gression from exposure to brand search.
To elicit a behavioral change, exposure
to advertising—and its impact—must be
understood. Drawing on literature, the
authors of the current paper seek to shed
new light on overlooked aspects of the
theory and practice of advertising and to
provide answers to these issues.
To find support for their research, data
from a large telecommunications company
compiled over a period of 78 weeks were
used. The analysis examines how advertis-
ing exposure and expenditure impact con-
sumer search for the company’s brand name.
LITERATURE REVIEW
The effectiveness of advertising has been a
major focus of interest (Bergkvist and Ros-
siter, 2008; Cho, 2003; Lodish et al., 1995;
Manchanda et al., 2006; Naik and Raman,
1 comScore Media Metrix, 2009
2 comScore Media Metrix, 2012
2003; Scholten, 1996; Telis, Chandy, and
Thaivanich, 2000; Yoo, Kim, and Stout,
2004). Advertising effectiveness has been
defined in terms of aspects of business
performance such as increase in sales
(Lodish et al., 1995; Tellis et al., 2000), cost
per impact (Briggs et al., 2005) and brand
awareness (Leone and Schultz, 1980; Mad-
dox, 2004; Vakratsas and Ambler, 1996).
And one study defined effectiveness as the
relationship between advertising expendi-
tures and brand sales, the affect of demand
by establishing a hierarchy of effects in its
audience (Scholten, 1996).
Media and Advertising Effectiveness
Advertising effectiveness may be
explained by factors such as brand and
category conditions, business strategies
objectives, media usage and copy related
measures (Lodish et al., 1995). Thus, dif-
ferences in media result in non-uniform
patterns of effectiveness.
This theory has led researchers to inves-
tigate responses of individuals to different
advertising media to measure their effec-
tiveness. Researchers focused on media
such as direct TV advertisements (tele-
vision commercials that include forms of
direct communication, such as 1-800 phone
numbers, that invite customer follow-up;
Tellis et al., 2000); online banners and Inter-
net advertising (Cho, 2003; Manchanda
et al., 2006); traditional and non-digital
methods (Naik and Raman, 2003); and
conventional television (Briggs et al., 2005).
These findings supported the effect of
advertising on firm performance indica-
tors. The support, however, was not uni-
form. For example, one study found higher
effectiveness (lower cost per impact) for
online compared to television advertis-
ing (Briggs et al., 2005). Furthermore,
the distinction between traditional and
new media was the focus of research-
ers who suggested novel media possess
stronger effectiveness through allowing
individuals to demonstrate higher degrees
of responsiveness to advertising (Fortin
and Dholakia , 2005).
Much attention has been directed
toward understanding key factors in
advertising effectiveness through tradi-
tional media (e.g., traditional and direct TV
advertisements, direct mail). The authors
of the current study, however, see the need
for further research on the effectiveness
of online advertising when accompanied
by other means of advertising communi-
cations, and to capture the mutual effects
of marketing communication channels in
broader time horizons.
Along those lines, the authors identified
past research showing that advertising
through one media outlet could influence
advertising effectiveness through another
media outlet (Assael, 2011). Exposure to
banner advertising significantly influ-
ences Internet purchasing (Manchanda
et al., 2006), and similarly, online adver-
tising has been found to have significant
effects on offline sales (Lewis and Reiley,
2013). In addition, the Internet’s appeal as
a low-cost advertising medium (Kim and
Balachander, 2010; Briggs et al., 2005) and
constraints on offline advertising (Gold-
farb and Tucker, 2011) have influenced the
use of online media as a desirable advertis-
ing communication channel.
Consumer responsiveness to advertising
and interactivity between individuals and
advertising media also have been influ-
ential in advertising effectiveness. High
program involvement, however, was a
potential blocker of mental processing of
advertising (Levy and Nebenzahl, 2006).
Research found lower involvement of
individuals to negatively affect subsequent
communication by advertisers (Cho, 2003).
Although user involvement may have a
significant effect on advertising effective-
ness, changes in interaction between media
and individuals could affect involvement,
which in turn could positively influence
Cross-channel advertising
has grown steadily and
significantly as a means
to reach consumers.
December 2013 JOURNAL OF ADVERTISING RESEARCH 433
EFFECTS OF MULTI-CHANNEL MARKETING ON CONSUMERS’ ONLINE SEARCH BEHAVIOR
advertising effectiveness. Besides varia-
tion of advertising effectiveness due to dif-
ferences in media, differences in product
categories (Lodish et al., 1995), the timing
of advertising, and vividness were pro-
posed to affect advertising effectiveness.
Furthermore, market and product charac-
teristics also influence advertising effec-
tiveness (Tellis et al., 2000). These instances
hint at the multi-dimensional nature of
advertising effectiveness.
Multichannel Advertising and Synergy
Over the last decade, advertisers increas-
ingly and successfully have used multi-
platform communications to achieve
synergistic results in getting messages
across to consumers within a single mar-
keting campaign. Recent reports released
by ESPN on viewers’ simultaneous usage
of TV and the Internet confirm the exist-
ence of such synergies (Enoch and John-
son, 2010). And NBC’s sponsorship of the
2010 Winter Olympics has been used as a
platform to further measure mutual impact
of television (in-home and out-of-home),
mobile, and Internet advertisements to
better understand cross-platform market-
ing communications (Assael, 2011).
Novelty of a stimulus often has a pro-
found effect on its effectiveness. Research
has shown that a second exposure to a
novel stimulus with similar information
attracts more attention than exposure to the
same stimulus (Putrevu and Lord, 2003).
Another study found that employing both
digital and traditional advertising chan-
nels improved overall advertising effec-
tiveness (Naik and Raman, 2003) through
enhancement of processing and improve-
ment of memory performance than when
a single medium serves as the advertising
channel (Edell and Keller, 1999). Further-
more, subsequent research found a syner-
gistic effect of a second medium to result
from higher impact on cognitions and
increased processing, and that repetition
of an advertisement through the same
medium (television) is less effective than
when a second medium (Internet) is used
(Chang and Thorson, 2004).
Moreover, interactivity influences its
synergistic effects with other complimen-
tary media, as it defines a proactive role for
audiences, increasing their involvement in
the communication process (Allen, Kania,
and Yaeckel, 1998). Such characteristics—
combined with television’s attention-
getting nature, which stems from its sound
and imagery effects (Blackwell, Miniard,
and Engel, 2001; Chang and Thorson, 2000;
Rossiter and Bellman, 1999)—lead to high
levels of advertising effectiveness.
The characteristics of the Internet and
television as advertising media led some
researchers to focus on the synergistic
effects of their combined use in marketing
campaigns. Some pointed to similarities
between the Internet and traditional print
media (DeFleur, Davenport, Conin, and
DeFleur, 1992; Sunders and Nass, 1996;
Wakolbinger, Denk, and Oberecker, 2009);
others emphasized the differences (Eve-
land and Dunwoody, 2002; Karson and
Koraonkar, 2001) to explain and predict its
effects on audiences.
The visual and informative characteris-
tics of the Internet and television, in fact,
can result in different comprehensions
of the message, when the sequence of
being exposed to them could be differ-
ent (Chang and Thorson, 2000). Findings
of brand-memory extension when chan-
nels similar to the Web and television
are combined (Edell and Keller, 1999)
tend to support the elaboration likeli-
hood model (ELM), where motivations
to scrutinize arguments are linked to the
likelihood of message elaboration. Indi-
viduals exposed to campaigns form their
attitudes through central processing, in
contrast to those who move along the
peripheral route, when they are exposed
to campaigns with repeated messages
(Chang and Thorson, 2000; Petty and
Cacioppo, 1986, 1996a).
Applying the ELM to the Internet/tel-
evision advertising synergy resulted in
researchers speculating higher attention
(Allen et al., 1998; Blackwell et al., 2001;
Brock, Albett, and Becker, 1970; Grass and
Wallace, 1969; Putrevu and Lord, 2003;
Rossiter and Bellman, 1999) and more will-
ingness to scrutinize arguments, compared
to those subjected to repetition (Edell and
Keller, 1999; Harkins and Petty, 1981a,
1981b, 1987; McCullough and Ostrom,
1974) among consumers exposed to such
campaigns (Chang and Thorson, 2000).
These results are based on the effects that
multiple sources have on increasing mes-
sage credibility (Petty and Cacioppo, 1986,
1996b; Zimbardo and Leippe, 1991) because
of higher diversity in sources (Harkis and
Petty, 1987; McLuhan, 1964). Findings also
supported the synergy created by Internet
and television on advertising effectiveness.
For instance, although Chang and Thorson
(2004) did not find synergy to influence
credibility or attitude toward either brand
or advertisement, or purchase intention,
they obtained other findings.
These findings supported higher atten-
tion and positive thoughts and perceived
message credibility. And they further
reinforced the speculations on the impact of
Internet-television synergy on individuals’
cognitions, rather than their affective state.
DEVELOPMENT OF HYPOTHESES
The authors believe that applying the
ELM—to synergistic effects of multiple
media in communicating an advertising
message—results in the expectation that
attitude change is a function of likelihood
to elaborate the message.
When elaboration likelihood is low,
the individual more likely changes atti-
tudes along the peripheral route, where
initial attitudes and biases are more likely
to play an important role. The synergy
434 JOURNAL OF ADVERTISING RESEARCH December 2013
EFFECTS OF MULTI-CHANNEL MARKETING ON CONSUMERS’ ONLINE SEARCH BEHAVIOR
among multiple channels is expected to be
more effective when individuals who are
exposed to the message are more attentive
to it, allowing for changes in their attitudes
to happen along the central route (Petty
and Cacioppo, 1986).
Consumers with high levels of elabo-
ration likelihood also are expected to be
more motivated in scrutinizing argu-
ments to understand the true merits of a
brand (Petty and Cacioppo, 1986; Chang
and Thorson, 2000). Thus, consumers at
high levels of elaboration likelihood are
expected to make attempts to scrutinize
arguments presented via one communica-
tion medium through other media.
The similarities between the Internet
and human-mind processing information
(Eveland and Dunwoody, 2002) make it a
prime candidate for this purpose. There-
fore, it is expected that a higher number
of individuals with high levels of elabora-
tion likelihood refer to the Internet when
exposed to advertisements communicated
through other media.
Hence, the following relationships are
hypothesized:
H1a: An increase in television adver-
tising leads to increased searches
for the company’s brand online.
H1b: An increase in radio advertising
leads to increased searches for
the company’s brand online.
H1c: An increase in online advertis-
ing leads to increased searches
for the company’s brand online.
Although the ELM focuses on attitude
change and motivation to scrutinize argu-
ments in advertising media, there is research
on the influence that simultaneous exposure
to multiple channels to communicate a mes-
sage can have on the effectiveness of that
message. Multiple messages influence cred-
ibility of the message (Chang and Thorson,
2000) because of the diversity created for
audiences (Harkins and Petty, 1987).
Study of these influences, however,
barely has gone past conventional meas-
ures of advertising effectiveness. In par-
ticular, the likelihood of consumers’
becoming motivated to scrutinize the
information communicated via multiple
channels has been under-studied. Based on
the ELM, multiple channels of communica-
tion are more effective if they direct con-
sumers to take the central route by raising
their elaboration likelihood.
Similarly to previous arguments, con-
sumers more likely employ accessible and
convenient media at their disposal for this
purpose. Thus, the authors expected con-
sumers to be more likely influenced by the
additional credibility presented by multi-
ple advertising channels to conduct their
own investigation in understanding the
true merits of a brand. Hence:
H2: The number of search-engine
queries for a company’s branded
keywords is a function of the
exposure to advertising placed
in more than one medium (tele-
vision, radio, online).
The ability of companies to use multiple
advertising channels is related to their
expenditures. Expenditures are an accept-
able measure for the amount of advertis-
ing directed to consumers. Therefore, the
authors expected to find an increase in
advertising expenditures to result in expo-
sure to advertisements communicated
through multiple channels. Hence:
H3: An increase in total advertising
expenditure leads to increased
searches for the company’s
brand online.
Also, similar to previous arguments,
the authors also expected exposure to
advertising to result in drawing more
attention to the brand online. Therefore:
H4: An increase in advertising
exposure will lead to increased
organic clicks.
METHODOLOGY
Sample and Data
To test the hypotheses in the current study,
data from a major telecommunications
company were gathered. The time frame
of the data collection was extended over
78 weeks, from January 1, 2007 to June 30,
2008. This company regularly used televi-
sion, radio, and the digital media to adver-
tise its products to consumers. To collect
the appropriate data, several variables
were taken into account, among them:
• Costs
Costs from each of the campaigns of the
company during these 78 weeks were
collected. Data for weekly costs were
collected separately for radio, television,
and online advertising campaigns.
• Impressions
As a part of its online advertising cam-
paign, the company measured perfor-
mance through a pay-per-click metric.
The costs were incurred on the basis
of only number of impressions, which
referred to the number of times the paid
listing appears alongside the users’
queries. The total number of impres-
sions was extrapolated from the search
The ability of companies
to use multiple advertising
channels is related to
their expenditures.
December 2013 JOURNAL OF ADVERTISING RESEARCH 435
EFFECTS OF MULTI-CHANNEL MARKETING ON CONSUMERS’ ONLINE SEARCH BEHAVIOR
engines. Thus, accurate data of the total
number of impressions were collected.
“Clicks” referred to the number of
times users’ clicked on the advertis-
er’s sponsored results. Sales data were
obtained from each person’s click as well.
Each time someone clicked on an image
or link, a cookie was installed on the
browser, and the sale can be tracked. The
data for impressions were collected from
three service providers: Google, Yahoo,
and MSN. In 2013, these three providers
account for close to 96% of the market.3
• Clicks
The data for the number of clicks on the
advertising messages—and the num-
ber of times the products were visited
online through natural searches on
search engines—were collected. The
paid advertising clicks included those
from paid search, and from paid online-
display campaigns. The data for organic
searches were gathered using the com-
pany’s Web-analytics tool that gathered
visits made from non-paid searches. The
other data were collected from providers
of the paid online advertising services:
Online site sales: Total online sales
from the Web analytics tool were
collected. A “sale” referred to a com-
pletion and validation of a person
making a purchase of any of the prod-
ucts on the Web site.
Press-release data: Using Google
News, the number of online articles
that mentioned the company’s name
was collected. This information was
useful because it indicated the pop-
ularity of the company during any
given week.
Control data: Several additional data
sources were collected to serve as
control variables. Wages, salaries, and
3 August 2013 U.S. Search Engine Rankings, comScore.
com.
supplementary labor income as well
as personal disposable income and
personal expenditure on consumer
goods and services were collected.
FINDINGS
To test the study’s hypotheses, the authors
implemented a three-step approach:
• Unit root tests were conducted to deter-
mine whether the data were stationary
over time.
• Data analysis using a vector time series
model in the form of an autoregressive
model with exogenous variables (VARX)
was conducted. Vector auto regressive
models allow for the evolution and
interdependencies between multiple
time series to be captured (Sims, 1980).
• Corresponding impulse response
functions were generated to examine
the effect of a one-standard deviation
shock on one of the endogenous vari-
ables, in this case marketing spend-
ing or media impressions (Dekimpe
and Hanssens, 1999). T-statistics were
then conducted.
The first hypotheses focus on the influence
of advertising in each medium on subse-
quent searches of individuals who were
exposed to them. H1a, H1b, and H1c pre-
dicted that increased advertising through
television, radio, and Internet would
increase the number of online searches
conducted by individuals, following their
exposure to advertising.
Brand Impressions in Relation to
Television Impressions
To test H1a, total brand impressions result-
ing from customer searches subsequent to
the exposure to impressions from televi-
sion are analyzed using a VARX model.
Television impressions affected brand-
search behavior for 10 weeks, and the
authors witnessed an increased effect in
customer reactions on search engines as
well (See Figure 1). Interestingly, there was
a peak followed by a valley on customers’
2018161412108642
–80,000
–40,000
0
40,000
80,000
120,000
Figure 1 Total Impressions on Branded Keywords in
Response to Television Impressions
436 JOURNAL OF ADVERTISING RESEARCH December 2013
EFFECTS OF MULTI-CHANNEL MARKETING ON CONSUMERS’ ONLINE SEARCH BEHAVIOR
brand search behavior that lasted for four
weeks. The permanent effect was minimal,
a result that may have been attributable to
the fleeting nature of advertising messages.
Following the practices of an earlier
study, the adjustment period referred to
the period between the end of duration
of immediate response (fourth week) and
the time where the effect returns to normal
(tenth week; Pauwels, Hanssens, and Sid-
darth, 2002). This type of analysis origi-
nally was conducted for price promotions,
but it also can apply to customer search
behavior. In the case of television impres-
sions in the current study, the adjustment
period was six weeks.
The significance was calculated using
a t-statistic. A cut-off point of 1.0 was
selected. This cut-off point allowed for
relatively wide confidence intervals. Con-
sequently, instead of obtaining precise esti-
mates, the results of the current study were
indicative of cumulative effects (Dekimpe,
Hanssens, and Silva-Risso 1999). For H1a,
the t-test was 1.81; it was significant, sup-
porting H1a.
Elasticity also was calculated (See
Figure 2). The net effect during period
two was 0.0035, which illustrates the
amount of change in brand impressions
as the result of every impression made by
television advertising.
Television exposure resulted in high
levels of elaboration likelihood and, as a
result, the message delivered had more
impact than other types of advertising
messages. A 30-second spot enabled peo-
ple with interest in a brand to obtain their
product information through both sight
and sound. This enhanced elaboration
resulted in a longer-term effect on brand
online-searching behavior.
Brand Impressions in Relation to
Radio Impressions
Analysis of the VARX model for H1b
indicated that radio impressions had an
impact on brand online search behavior.
In fact, the effect of radio exposure on
brand keyword searches was noted dur-
ing the first four weeks followed by a
subsequent drop (See Figure 3). The adjust-
ment period, however, was found to be
less significant than for television
impressions.
The effects from radio impressions had
the same elasticity as the effects from tel-
evision during the first period, but it rap-
idly dropped afterward (See Figure 4).
–0.00400
–0.00350
–0.00300
–0.00250
–0.00200
–0.00150
–0.00100
–0.00050
0
0.00050
0.00100
135791113151719
Figure 2 Elasticity of Total Brand Impressions versus Total
Television Impressions
2018161412108642
–80,000
–40,000
0
40,000
80,000
160,000
120,000
Figure 3 Total Impressions on Branded Keywords in
Response to Radio Audience
December 2013 JOURNAL OF ADVERTISING RESEARCH 437
EFFECTS OF MULTI-CHANNEL MARKETING ON CONSUMERS’ ONLINE SEARCH BEHAVIOR
However, the t-test was 0.51. As the over-
all impact of radio was not significant, H1b
was not supported.
Brand Impressions in Relation to Online
Advertising Impressions
To test for the effects of online advertis-
ing on subsequent online search, brand
impressions in relation to online advertis-
ing impressions were analyzed using the
VARX method.
Display of online advertising had a con-
siderable impact on search behavior. In
the impulse-response function (See Fig-
ure 5)—as opposed to the finding for both
television and radio exposure—the effects
from online exposure had a steady (and
not sudden) decline. The t-test was 1.52—a
significant result—supporting H1c.
The finding implies that online adver-
tising engages people more than does tra-
ditional media. The fact that someone is
online surfing the Web is a plausible expla-
nation. A person exposed to display adver-
tising and engaged with the message can
search more easily for related brand infor-
mation. The immediacy of the message is
facilitated by the channel’s interactivity.
In addition, the long-term effect resem-
bles that of television, which implies that
online advertising can stimulate a change
in search behavior over a longer period of
time. Furthermore, the 0.012 elasticity (See
Figure 6) reflected a stronger response for
online advertising on brand search.
Brand Impressions in Relation to Total
Media Impressions
To test H2 and to see whether elaboration
developed through different advertis-
ing channels affected the search behavior
online, total brand impressions from cus-
tomer search behavior subsequent to all
of the impressions in different media (i.e.,
total media impressions) were analyzed
using the VARX model.
–0.00450
–0.00400
–0.00350
–0.00300
–0.00250
–0.00200
–0.00150
–0.00100
–0.00050
0
0.00050
135791113151719
Figure 4 Elasticity of Total Brand Impressions vs.
Radio Audience
2018161412108642
–50,000
0
50,000
100,000
150,000
Figure 5 Total Impressions on Branded Keywords in
Response to Online Display Impressions
The finding implies
that online advertising
engages people more than
does traditional media.
438 JOURNAL OF ADVERTISING RESEARCH December 2013
EFFECTS OF MULTI-CHANNEL MARKETING ON CONSUMERS’ ONLINE SEARCH BEHAVIOR
In an analysis of the impulse-response
function (See Figure 7), there was an imme-
diate effect of total marketing impressions
on branded keyword impressions that
lasted two weeks, and the adjustment
period lasted nine weeks. The t-test was
1.2—well above the cut-off point of 1.0—
and the result, therefore, was significant,
supporting H2.
In elasticity calculated for total brand
impressions, the net effect during week
two was 0.008 (See Figure 8). Thus, paid
media impressions from all different media
channels combined influenced brand que-
ries to a large extent.
Brand Impressions in Relation to Paid
Media Expenditures
To test the effect of advertising expen-
ditures on subsequent search, brand
impressions in relation to paid media
expenditures were analyzed using the
VARX model (See Figure 9). Total adver-
tising expenditures had a similar impact as
total media impressions on branded key-
word searches, though to a lesser extent.
The t-test result was 1.09, a significant find-
ing that supported H3.
Total Marketing Impressions in Relation
to Organic Searches
The effects of advertising exposure on
subsequent search through organic clicks
(H4) were tested using the VARX model.
The current study measured the effect of
total marketing impressions on organic
search traffic, which could be observed
in the impulse-response function (See
Figure 10) and the elasticity chart (See
Figure 11). Organic traffic was not sig-
nificantly affected by total marketing
–0.0140
–0.0120
–0.0100
–0.0080
–0.0060
–0.0040
–0.0020
0
0.0020
0.0040
135791113151719
Figure 6 Elasticity of Total Brand Impressions vs. Online
Display Impressions
2018161412108642
–80,000
–40,000
0
40,000
80,000
120,000
Figure 7 Total Impressions on Branded Keywords in
Response to Total Media Impressions
Consumer responsiveness
to advertising and
interactivity between
individuals and
advertising media have
been influential in
advertising effectiveness.
December 2013 JOURNAL OF ADVERTISING RESEARCH 439
EFFECTS OF MULTI-CHANNEL MARKETING ON CONSUMERS’ ONLINE SEARCH BEHAVIOR
impressions. The t-test was 0.59, a level
that failed to support H4. Moreover, the
elasticity was low and, in week two, it
was 0.0001.
Organic clicks measure total visits from
search engines regardless of keyword cat-
egory and are not the result of paid adver-
tising. A more in-depth analysis could be
conducted on organic traffic by examining
the keyword categories (i.e., branded ver-
sus generic). Such analysis would deter-
mine whether only frequency of search
for branded keywords saw an increase or
whether the same effect also was found for
online searches of general keywords (e.g.,
the words “cell phone” instead of a specific
brand of cell phone).
The authors’ findings support the effects
of total marketing expenditures, total mar-
keting impressions, television impres-
sions, and online display impressions
on consumers’ subsequent online search
behavior. However, the results of the
current study do not lend support to the
effect of radio impressions on subsequent
online search. They also fail to support the
hypothesized relationship for the effect of
media impressions on organic clicks (See
Table 1).
DISCUSSION
The effects of cross-channel online
advertising remain a relatively under-
–0.01
–0.00
–0.00
–0.00
–0.00
0
0.00
0.00
13579111315171921
Figure 8 Elasticity of Total Brand Impressions vs. Total
Media Impressions
2018161412108642
–40,000
0
40,000
80,000
120,000
Figure 9 Total Impressions on Branded Keywords in
Response to Total Media Expenditures
The authors’ findings
support the effects
of total marketing
expenditures, total
marketing impressions,
television impressions,
and online display
impressions on
consumers’ subsequent
online search behavior.
440 JOURNAL OF ADVERTISING RESEARCH December 2013
EFFECTS OF MULTI-CHANNEL MARKETING ON CONSUMERS’ ONLINE SEARCH BEHAVIOR
studied field. This paper examined the long-
and short-term effects of offline and
online advertising on search—more
specifically, searches for the advertiser’s
brand on search engines, using the
adjustment period and total effects of
advertising on brand search queries (See
Table 2).
The current study found that total mar-
keting impressions, total marketing expen-
ditures, and television and online display
exposure were found to have significant
impacts on brand searches. Radio exposure
did not have a significant effect on brand
searches. The impact of total marketing
impressions on organic clicks also was
found to be not significant.
The findings revealed a significant short-
term increase in queries for the target
brand. More important, they showed that
the increase in subsequent searches did not
sustain over time. The authors also found
that total marketing spending and total
advertising impressions had the greatest
short-term impact on brand queries. The
adjustment time period lasted only three
weeks and then leveled off, and the elas-
ticity was greatest during the first three
time periods.
The findings for television and online
advertising effects on brand searches were
similar (but to a lesser extent) to the effects
of marketing spending and total adver-
tising impressions. The total effects of all
the advertiser’s efforts sustained a lift in
brand queries. Albeit for a short time, the
–0.00002
0
0.00002
0.00004
0.00006
0.00008
0.00010
0.00012
0.00014
135791113151719
Figure 11 Elasticity of Total Organic Clicks versus Media
Impressions
2018161412108642
–40,000
0
40,000
80,000
120,000
Figure 10 Total Organic Clicks in Response to Total Media
Impressions
TABLE 1
T-Test Results for Each
Hypothesis
Variable T-Statistic Significant?
Total Marketing
Spend 1.21 Yes
Total Marketing
Impressions 1.09 Yes
Television
Impressions 1.89 Yes
Radio
Impressions 0.51 No
Online Display
Impressions 1.52 Yes
Organic Clicks 0.59 No
December 2013 JOURNAL OF ADVERTISING RESEARCH 441
EFFECTS OF MULTI-CHANNEL MARKETING ON CONSUMERS’ ONLINE SEARCH BEHAVIOR
effects of online display advertising were
more pronounced. The immediacy of the
medium allowed searches for the brand to
occur much more readily than through tra-
ditional channels.
The findings emphasize the importance
of consistency of messages across differ-
ent advertising media, particularly for
integrated marketing campaigns. The out-
come of our analysis suggests that large
advertising campaigns will drive more
traffic to their Web sites, and this addi-
tional traffic must be sustained through the
company’s Web site. Marketers can seize
such opportunities to present additional
purchase incentives.
One key implication of the findings is
that firms can maximize brand-related
queries with their marketing endeavors.
As brand queries are known to occur dur-
ing the short term, firms may be able to
design their media plans to maximize cus-
tomer immediate searches through offer-
ing additional online information.
LIMITATIONS AND FUTURE RESEARCH
The study is not without limitations,
specifically:
• User involvement with the advertising
message was not captured because the
relevance of the advertising message to
audiences is not captured by the data.
Using only secondary data does not
allow examining if the actual message
resonates with audiences.
• Existing brand awareness might affect
how audiences perceive the brand and,
thus, their search behavior. A less-well-
known brand might not display the
same effects as those in this study; rep-
licating the study with a different brand
from a different industry would help
validate the results.
• A third limitation concerns the lack of
precise measurement of existing con-
sumers looking for the company’s
brand online. A number of brand
queries might have originated from
existing clientele, thus eliminating the
need to advertise directly to them. This
is a reality of any advertising campaign.
The current study suggests areas for
further research, in that it specifically
examines only the telecommunications
industry specifically. Further research
can replicate it by focusing on a differ-
ent industry. Customer search might
vary across different industries, and
it would be important to test whether
these findings hold across different
product categories.
Future studies may also involve focusing
on Web analytics data more in depth to
understand the quality of each visit on the
company’s site. Looking at this Web data
may help researchers better understand
the behavior of visitors who did a Web
search as compared to the rest of the visit-
ors exposed to different channels. Finally,
an area of future research is to examine
the impact of offline advertising on
mobile search behavior (smartphones
and tablets).
MiChel laroChe is the Royal Bank Distinguished
Professor of Marketing, John Molson School of
Business, Concordia University, Montreal. His main
research interests are in marketing communications,
Internet and services marketing, and retailing, with
an additional interest in the role of culture and brand
decision processes in consumer behavior. He has
published more than 280 articles in proceedings and
journals, including the Journal of Consumer Research
and the Journal of Adver tising Research. Laroche
is the managing editor of the Journal of Business
Research and a member of the Academy of Marketing
Science board of governors.
TABLE 2
Summary of Findings
Hypothesis Supported?
H1a: An increase in television advertising leads to increased searches for
the company’s brand online
Yes
H1b: An increase in radio advertising leads to increased searches for the
company’s brand online
No
H1c: An increase in online advertising leads to increased searches for the
company’s brand online
Yes
H2: The number of search engine queries for a company’s branded
keywords is a function of the exposure to adver tising placed in more than
one medium
Yes
H3: An increase in total advertising expenditures leads to increased
searches for the company’s brand online
Yes
H4: An increase in advertising exposure leads to increased organic clicks No
One key implication of
the findings is that firms
can maximize brand-
related queries with their
marketing endeavors.
442 JOURNAL OF ADVERTISING RESEARCH December 2013
EFFECTS OF MULTI-CHANNEL MARKETING ON CONSUMERS’ ONLINE SEARCH BEHAVIOR
iSar kiani is a PhD candidate in marketing at the John
Molson School of Business, Concordia University
(Montreal, Canada). Her research interests include
diffusion of market information, adver tising, online
and digital marketing, cognitive and neurological
antecedents of consumer decision making, and
cultural inuences in consumer perception and
decision making. Kiani holds an MBA degree
from Sharif University of Technology (Tehran) and
previously served as the marketing manager at Fara
Management Organization in Iran.
neCtarioS eConoMakiS, a Concordia University alumnus
(MS, marketing), is an account executive for Google,
Montréal, where he leads the development of
quebec clients. Nectar uses his academic and years
of advertising agency experience to help promote
Google’s products. He is also co-chair of the IAB
Canada Search Committee.
Marie-odile riChard (PhD, marketing, HEC-University
of Montreal) has research interests in marketing
communications (including Internet marketing),
neuromarketing, services marketing, and cultural
effects on individual responses. She has published
more than 30 articles in jour nals and proceedings,
including the Journal of the Academy of Marketing
Science, Journal of Adver tising Research, Journal of
Business Research, and Journal of Social Psychology.
REFERENCES
allen, C., D. kania, and B. yaeCkel. Internet
World: Guide to One-on-One Web Marketing. New
York: Wiley, 1998.
aSSael, C. “From Silos to Synergy: A Fifty-Year
Review of Cross-Media Research Shows Syn-
ergy Has Yet to Achieve Its Full Potential.” Jour-
nal of Advertising Research 51, 1 (2011): 42–48.
Bergk viSt , l., and J. r. roSSi ter. “The Role of
Ad Likability in Predicting an Ad’s Campaign
Performance.” Journal of Advertising 37, 2 (2008):
85–98.
BlaCkwell, r. D., P. w. miniarD, and J. F.
engel. Consumer Behavior. Mason, OH, South-
Western, 2001.
BriggS, r., r. kriShnan, and n. Borin.
“Integrated Multichannel Communication
Strategies: Evaluating the Return on Market-
ing Objectives—the Case of the 2004 Ford F-150
Launch.” Journal of Interactive Marketing 19, 3
(2005): 81–90.
BroCk, t. C., S. m. alBert, and l. a. BeCker.
“Familiarity, Utility, and Supportiveness as
Determinants of Information Receptivity.”
Journal of Personality and Social Psychology 14, 4
(1970), 292–301.
CaywooD, C. l., D. e. SChu ltZ, and P. wang.
A Survey of Consumer Goods Manufacturers. New
York: American Association of Advertising
Agencies, 1991.
Chang, y., and e. thorSon. “Television and
Web Advertising Synergies.” Journal of Advertis-
ing 33, 2 (2004): 75–84.
Cho, C. h. “The Effectiveness of Banner Adver-
tisements: Involvement and Click-through.”
Journalism & Mass Communication Quarterly 80,
3 (2003): 623–645.
DeFleur, m. l., l. DavenPo rt, m. Cronin, and
m. DeFleur. “Audience Recall of News Stories
Presented by Newspaper, Computer, Television,
and Radio.” Journalism Quarterly 69, 4 (1992):
1010–1022.
DekimPe, m. g., and D. m. hanSSenS. “Sus-
tained Spending and Persistent Response: A
New Look at Long-Term Marketing Profitabil-
ity.” Journal of Marketing Research 36, 4 (1999):
397–412.
DekimPe, m. g., D. m. hanSSenS, and J. m.
Silva-riSSo . “Long-Run Effects of Price Promo-
tions in Scanner Markets.” Journal of Economet-
rics 89, 1-2 (1999): 269–291.
eDell, J., and k. l. keller. “Analyzing Media
Interactions: The Effects of Coordinated TV-
Print Advertising.” Marketing Science Institute
(1999): 99–120.
enoC h, g., and k. JohnSon. “Cracking the
Cross-Media Code: How to Use Single-Source
Measures to Examine Media Cannibalization
and Convergence.” Journal of Advertising
Research 50, 2 (2010): 15.
evelanD Jr., w. P., and S. h. DunwooDy. “An
Investigation of Elaboration and Selective Scan-
ning as Mediators of Learning from the Web
versus Print.” Journal of Broadcasting & Electronic
Media 46, 1 (2002): 34–53.
Fortin, D. r., and r. r. Dholakia. “Interactiv-
ity And Vividness Effects on Social Presence
and Involvement with a Web-Based Advertise-
ment.” Journal of Business Research 58, 3 (2005):
387–396.
golDFarB, a., and C. e. tuCker. “Search Engine
Advertising: Channel Distribution When Pric-
ing Ads to Context.” Management Science 57, 3
(2011): 458–470.
graSS, r. C., and w. h. wallaCe. “Satiation
Effects of TV Commercials.” Journal of Advertis-
ing Research 9, 3 (1969): 3–8.
harkinS, S. g., and r. h. Petty. “Effects of
Source Magnification of Cognitive Effort on
Attitudes: An Information-Processing View.”
Journal of Personality and Social Psychology 4, 3
(1981a): 401–413.
harkinS, S. g., and r. h. Petty. “The Multi-
ple Source Effect in Persuasion: The Effects of
Distraction.” Personality and Social Psychology
Bulletin 7, 4 (1981b): 627–635.
harkinS, S. g., and r. e. Petty. “Information
Utility and the Multiple Source Effect.” Journal
of Personality and Social Psychology 52, 2 (1987):
260–268.
December 2013 JOURNAL OF ADVERTISING RESEARCH 443
EFFECTS OF MULTI-CHANNEL MARKETING ON CONSUMERS’ ONLINE SEARCH BEHAVIOR
karSon, B. J., and P. k. korgaonkar. “An
Experimental Investigation of Internet Adver-
tising and the Elaboration Likelihood Model.”
Journal of Current Issues and Research in Advertis-
ing 23, 2 (2001): 53–72.
kim, a., and S . BalaChanDer. “Coordinat-
ing Traditional Media Advertising and Search
Advertising.” Unpublished manuscript,
Purdue (2010), Purdue University, West
Lafayette, IN.
leone, r. P., and r. l. SChultZ. “A Study of
Marketing Generalizations.” Journal of Marketing
44, 1 (1980): 10–18.
levy, S., and i. D. neBe nZa hl. “Programme
Involvement and Interactive in Interactive Tel-
evision.” International Journal of Advertising 25,
3 (2006): 309–332.
lewi S, r. a., and D. h. reiley. “Does Retail
Advertising Work? Measuring the Effects of
Advertising on Sales via a Controlled Experi-
ment at Yahoo! Unpublished manuscript, Yahoo!
Retrieved from http://www.davidreiley.com/
papers/DoesRetailAdvertisingWork.pdf
loDiSh, l. m., m. aBraham, S. kalmenSon, J.
livelSBerger, B. luBetkin, B. riCharDSon, and
m. e. StevenS. “How T.V. Advertising Works: A
Meta-Analysis of 389 Real World Split Cable T.V.
Advertising Experiments.” Journal of Marketing
Research 32, 2 (1995): 125–139.
maDDox, k. “Search Ads Lift Brand Aware-
ness.” B-to-B 89, 9 (2004): 12.
manChanDa, P., J-P. DuBé, k. yong goh, and P.
k. Chintagunta. “The Effect of Banner Adver-
tising on Internet Purchasing.” Journal of Market-
ing Research 43, 1 (2006): 98–108.
mCCullough, J. l., and t. m. oStrom. “Rep-
etition of Highly Similar Messages and Atti-
tude Change.” Journal of Applied Psychology 59,
3 (1974): 395–397.
mCluhan, m. Understanding Media: The Exten-
sions of Man. Cambridge: MIT Press, 1964.
naik, P. a., and k. raman. “Understanding the
Impact of Synergy in Multimedia Communica-
tions.” Journal of Marketing Research 40, 4 (2003):
375–388.
PauwelS, k., D. m. hanSS enS, and S. SiDDarth.
“The Long-Term Effects of Price Promotions on
Category Incidence, Brand Choice, and Pur-
chase Quantity.” Journal of Marketing Research
39, 4 (2002): 421–439.
Petty, r. e., and J. t. CaCioPPo. Communica-
tion and Persuasion: Central and Peripheral Routes
to Attitude Change. New York: Springer-Verlag,
1986.
Petty, r. e., and J. t. CaCioPPo. Attitude and
Persuasion: Classic and Contemporary Approaches.
Boulder, CO: Westview, 1996a.
Petty, r. e., and J. t. CaCioPPo. Attitude and
Persuasion: Classic and Contemporary Approaches.
Boulder, CO: Westview, 1996b.
Putrevu, S., and k. r. lor D. “Processing Inter-
net Communications: A Motivation, Opportun-
ity and Ability Framework.” Journal of Current
Issues and Research in Advertising 25, 1 (2003),
45–59.
roSSiter, J. r., and S. Bellman. “A Proposed
Model for Explaining and Measuring Web Ad
Effectiveness.” Journal of Current Issues and
Research in Advertising 21, 1 (1999): 13–31.
SCholten, m. “Lost and Found: The
Information-Processing of Advertising Effective-
ness.” Journal of Business Research 37, 2 (1996):
97–104.
SimS, C. a. “Macroeconomics and Reality.”
Econometrica 48, 1 (1980): 1–48.
SunDe rS, S. S., and C. naSS. “Source Effects in
Users’ Perception of Online News.” Presented
at the 46th Conference of the International Com-
munication Association, 1996.
telliS, g. J., r. k. ChanDy, and P. thaivaniCh.
“Which Ad Works, When, Where, and How
Often? Modeling the Effects of Direct Television
Advertising.” Journal of Marketing Research 37, 1
(2000): 32–46.
vakratSaS, D., and t. amBler. “Advertising
Effects: A Taxonomy and Review of Concepts,
Methods, and Results from the Academic Lit-
erature.” Working paper. Cambridge: Marketing
Science Institute 1996: 96–120.
wakolBinger, l. m., m. Denk, and k.
oBereCker. “The Effectiveness of Combining
Online and Print Advertisements.” Journal of
Advertising Research 49, 3 (2009): 360–372.
yoo, C. y., k. kim , and P. a. Stou t. “Assess-
ing the Effects of Animation in Online Ban-
ner Advertising: Hierarchy of Effects Model.”
Journal of Interactive Advertising 4, 2 (2004): 49–60.
ZenithoPtimeDia. “ZenithOptimedia Fore-
casts Stable Ad Growth in 2013 Will Pave
Way for Recovery in 2014 and 2015,” Press
release, September 30, 2013. Retrieved www.
zenithoptimedia.com/zenithoptimedia-
forecasts-stable-ad-growth-in-2013-will-pave-
way-for-recovery-in-2014-and-2015/
ZimBarDo, P. g., and m. r. lei PPe. The Psychol-
ogy of Attitude Change and Social Influence. New
York: McGraw–Hill, 1991.
CopyrightofJournalofAdvertisingResearchisthepropertyofWarcLTDanditscontent
maynotbecopiedoremailedtomultiplesitesorpostedtoalistservwithoutthecopyright
holder'sexpresswrittenpermission.However,usersmayprint,download,oremailarticlesfor
individualuse.
