Summary advertising – important scientific articles
Danaher, P. J., Danaher, T. S., Smith, M. S., & Loaiza-Maya, R. (2020).
Advertising effectiveness for multiple retailer-brands in a multimedia and
multichannel environment. Journal of Marketing Research, 57(3), 445-467.
Kim, H., & Hanssens, D. M. (2017). Advertising and word-of-mouth effects on
pre-launch consumer interest and initial sales of experience products. Journal
of Interactive Marketing, 37, 57-74.
Tirunillai, S., & Tellis, G. J. (2017). Does offline TV advertising affect online
chatter? Quasi-experimental analysis using synthetic control. Marketing
Science, 36(6), 862-878.
Guitart, I. A., & Stremersch, S. (2020). The Impact of Informational and
Emotional Television Ad Content on Online Search and Sales. Journal of
Marketing Research, 58(2), 299-320.
,Danaher, P. J., Danaher, T. S., Smith, M. S., & Loaiza-Maya, R.
(2020). Advertising effectiveness for multiple retailer-brands in
a multimedia and multichannel environment. Journal of
Marketing Research, 57(3), 445-467
In this research, the authors monitor 4,000 customers over two years, linking their exposure
to three media (email, catalogs, and paid search) to their in-store and online purchases for
three retailer-brands in the clothing category.
They find that email and sometimes catalogs from a focal retailer-brand have a negative
influence on other retailer-brands in the category, whereas paid search influences only the
focal retailer-brand. However, competitor catalogs often positively influence focal retailer-
brand sales, but only among omnichannel customers.
This study has three previously unexplored objectives:
(1) to evaluate the effect of own- brand and cross-brand advertising on sales for each
advertising medium separately and for two purchase channels;
(2) to gauge the extent to which customers differentially purchase across multiple brands
and multiple channels, thereby creating brand/channel customer segments; and
(3) to examine the response to focal firm and competitor multimedia advertising for these
brand/channel segments.
They find, for example, that own-brand advertising has the largest influence on sales for a
focal brand, but advertising from other brands owned by the parent company can have
either a positive or negative influence on the focal brand. For instance, catalogs from a focal
firm sometimes positively spill over to increase sales for other brands, while email from a
focal brand always has a negative influence on sales for other brands within the portfolio.
The model for multiple brands, purchase channels, and media channels enables firms to
gain previously elusive insights into how to manage advertising allocation at a parent
company level rather than just the brand level. For instance, one of the brands has lower ad
elasticities than the others for each medium, which suggests that the parent firm could
reduce ad expenditures for this brand and redeploy that spend to its other brands. The
researchers also find that catalogs could be more effective for the parent firm if their
delivery dates were spread out over time, compared with the current practice of
simultaneously mailing each brand’s catalog within the first week of each month. Finally, the
segmentation of the market by brand and channel usage reveals four customer segments
comprising mul- tibrand/single-channel (20.5%), single-brand/omnichannel (20.8%), single-
brand/single-channel (37.5%), and multi- brand/omnichannel (21.2%) buyers. Moreover,
each of these segments differ in their responsiveness to advertising media, with catalogs
being very effective for the single-brand/omni- channel segment, while email and paid
search are very effective for the highly prized multi-brand/omnichannel shoppers.
Generally, the heaviest buyers of each brand are the most responsive to focal-brand
advertising, but they are often also the most responsive to advertising from other brands,
especially the multi-brand/omnichannel segment.
,Conceptual framework
The researchers develop a framework showing how different forms of advertising
expenditure by firms (on email, catalogs, and paid search), along with consumer-initiated
touchpoints (online search and website visits), can influence in-store and online sales in a
multi-brand setting.
First, an email or catalog for B1 might prompt a customer to undertake a search for the
brand name, product category, or a specific product mentioned in the email or catalog
(Dinner, Van Heerde, and Neslin 2014). The resulting search query will produce organic
search links, possibly along with paid search links depending on the search terms chosen
and the amount retailer B1 has bid for paid search keywords (Narayanan and Kalyanam
2015). Second, exposure to B1’s email or catalog might encourage a customer to visit B1’s
website by typing the brand name into their browser or by selecting the brand from their
favorites list. That is, a website visit occurs, but not as the result of an online search query.
Our clickstream data capture enables us to distinguish between organic search, paid search,
and direct website visits as the route for how a customer arrives at a brand’s website.
Search queries and direct website visits occur when customers are at a more advanced
stage of their purchase decision making process offset these boxes to the right in the
figure.
The article also note that it is not a requirement for email and catalog exposures to precede
a search query or website visits. Instead, a customer can voluntarily initiate online and
purchase activity in the absence of recent advertising. Figure 1 shows that email and catalog
advertising have two paths to purchase, generating sales either directly or through an
indirect route by prompting online activity, which then leads to purchase. In contrast, there
is only a direct link from paid search advertising to sales.
The influence of email, catalogs, online search, and website visits on sales occurs
analogously for brands B2 and B3, with solid lines in Figure 1 connecting focal brand
advertising to sales. Figure 1 also illustrates that advertising and online activity for one
brand can also influence sales for the other two brands, as shown by the dotted lines
indicating that they expect cross-brand effects to be weaker in magnitude than own-brand
effects (Clarke 1973).
The researchers expect the own-brand effects of email, catalogs, and paid search to be
positive. Although cross-brand ad elasticities are always expected to be negative (Clarke
1973), there is mounting empirical evidence that positive values are also possible, usually
among brands with similar features (Janakiraman, Sismeiro, and Dutta 2009; Sahni 2016),
when there are barriers to switching (Anderson and Simester 2013) or when a larger brand
gains from a smaller brand’s advertising (Danaher, Bonfrer, and Dhar 2008). Therefore, the
sign of cross-brand ad elasticities is determined empirically.
Data
, Table 1 shows that B1 customers are the least likely to buy online, while B2 customers are
the most likely. Part of the reason for this is that B2 has less than half the number of stores
as both B1 and B3, which reduces access to physical stores thereby making online purchases
necessary for many customers.
Panel A gives the conditional probability of purchasing one of the brand/channel
combinations given a purchase of a different brand/channel. For example, there is an 83.9%
chance that a person who makes an online purchase of brand B1 will also make an in-store
purchase of the same brand. While an overlap of this magnitude might be expected given
that it covers the same brand, the researchers also see high overlap for a different brand,
with online B1 buyers having a 55.1% chance of buying brand B3 in-store.
Handling endogeneity
Table 3 summarizes the potential sources of endogeneity and how the research proposes to
handle them. The first possible source of endogeneity is due to targeting, and the two media
most likely to suffer from this are email and catalogs because they are directly controlled by
managers responsible for sending messages to consumers (Zantedeschi, Feit, and Bradlow
2017). A second potential source of endogeneity arises from paid search advertising,
whereby customers signal their interest in a brand or product category by initiating online
search, a form of self-selection.
The type of endogeneity associated with email and catalogs arises from brand managers
deliberately targeting consumers they perceive as more likely to purchase. Digital media,
such as email, can be used to micro-target customers at a much more granular level than is
possible when using traditional media. Thus, without knowing the criteria an advertiser uses
to target ads to customers, an outside observer can easily over-estimate the impact of
advertising on sales because heavier buyers or those exhibiting recent interest in a product
have a higher likelihood of receiving more advertising.
The second major source of endogeneity arises from self- selection, whereby consumers
who are “in the market” for clothing, for example, might elect to use an online search
facility (e.g., Google) to seek out product categories or particular brands as part of the
process for eventual purchase. Consequently, a consumer who signals their possible intent
to purchase through an organic search query is different from a consumer who has not
