Decomposing the Impact of Advertising: Augmenting Sales with Online Search Data

Abstract

Unlike sales data, data on intermediate stages of the purchase funnel (e.g., how many consumers have searched for information about a product before purchase) are much more difficult to acquire. Consequently, most advertising response models have focused directly on sales and ignored other purchase funnel activities. The authors demonstrate, in the context of the U.S. automotive market, how consumer online search volume data from Google Trends can be combined with sales data to decompose advertising's overall impact into two underlying components: its impacts on (1) generating consumer interest in prepurchase information search and (2) converting that interest into sales. The authors show that this decompositional approach, implemented through a novel state-space model that simultaneously examines sales and search volumes, offers important advantages over a benchmark model that considers sales data alone. First, the approach improves goodness-of-fit, both in and out of sample. Second, it improves diagnosticity by distinguishing advertising effectiveness in interest generation from its effectiveness in interest conversion. Third, the authors find that overall advertising elasticity can be biased if researchers consider only sales data.

Does advertising spending improve sales performance?

“…First by "making consumers interested enough in the focal product that they would seek information about it and, second, by converting information-seeking consumers into buyers" (Hu et al, 2014;p.300). By improving market awareness, advertising also improves the firm's competitive position, increases customer preferences, and strengthens the brand image (Koslow et al, 2006;Kulkarni et al, 2003;Tellis, 2010).…”

Does Online Search Predict Sales? Evidence from Big Data for Car Markets in Germany and the UK

“…Choi and Varian (2012) suggest using specific keywords to isolate s a i,t . Hu et al (2014), who focus on pre-purchase searches, attempt to further exclude post-purchase and fandom searches by excluding searches that involve specific keywords indicating an interest in repairs, spare parts and the like. Reliance on category filters and attempts to exclude searches that do not precede a purchase can be problematic if Google alters the underlying algorithms or if consumer behavior changes so that new keywords need to be added to the exclusion list.…”

“…The modeling approach we propose does not require collection of product characteristics, such as prices. With one exception, Hu et al (2014), the importance of structural assumptions for identification has gone unnoticed in the literature on use of aggregate search data. We apply our model to data on online search for products and product-level sales in the German and UK car markets.…”

“…Data capturing consumers' internet searches have been related to demand in a number of recent papers. Hu et al (2014) study the interaction between search and advertising using aggregate search engine data to analyze a long panel (T > 100) of product level sales and advertising data. They also model search, but do not consider the empirical challenges presented by the tall nature of their data.…”

Efeitos das Mídias Digitais nas Novas Vendas B2B: Um ensaio sobre inbound marketing, mídias pagas e ganhadas on-line

“…No âmbito de marketing digital, percebe-se que os efeitos das mídias pagas e ganhadas on-line nas vendas é um assunto intrínseco às organizações e, assim, amplamente pesquisado na área de marketing. A literatura mostra que mídias pagas e ganhadas impactam no resultado de vendas das organizações (Trusov, Bucklin, & Pauwels, 2009;Norris, Kay, & Prasad, 2012;Onishi & Manchanda, 2012;McQuarrie, Miller, & Phillips, 2013;Dewan & Ramaprasad, 2014;Dinner, Van Heerde, & Neslin, 2014;Draganska, Hartmann, & Stanglein, 2014;Hu, Du, & Damangir, 2014;Fang, Li, Huang, & Palmatier, 2015;Kumar, Bezawada, Rishika, Janakiraman, & Kannan, 2016;Kumar, Choi, & Greene, 2017). Contudo, ainda faltam iniciativas teóricas de modelos que possam englobar os efeitos do inbound marketing, das mídias pagas e das mídias ganhadas on-line nas novas vendas B2B, justificando um ensaio teórico como este.…”