Estimating Multichannel Advertising Effectiveness using a Dynamic Integrated Marketing Model
With the continuing rise of new communication channels, multichannel marketeers are increasingly challenged to optimally allocate their media budget across channels to maximize profits over the long term (Danaher and Dagger, 2013; Leeflang et al., 2009; Neslin and Shankar, 2009; Wiesel et al., 2011). The combination of online and offline media channels as well as sales in web stores and physical stores hinders marketeers to thoroughly understand to which extent a particular channel contributes to sales. As increasingly used data-driven attribution models fail to incorporate offline advertising and offline sales (Kannan et al., 2016), it is necessary to combine attribution with marketing-mix modelling in order to examine the effectiveness of offline media advertising as well. Using data of online and offline media advertising and sales provided by retailer Coolblue, the current study proposes a dynamic integrated marketing model to investigate the dynamic effectiveness of online and offline channels on sales and to examine spillover effects across and interactions among these channels. We find substantial evidence for strong long-term effects of offline advertising as well as time-varying effects of upper funnel channels on lower funnel channels. Further, we find positive spillover effects from offline upper funnel channels on the traffic through and conversion rates of online channels. Finally, we conclude that cannibalization as well as synergies exists among several online lower and upper funnel channels. Translating these results into a pragmatic overview of channel contributions, we conclude that paid search advertising accounts for approximately three fifth of total sales, followed by contributions of approximately one tenth of direct traffic, TV and organic traffic and smaller contributions of radio, display and email advertising. By means of our proposed model, the current study contributes to science and business by providing more accurate channel contributions that can help multichannel marketeers improve their marketing strategies and optimize profits (Kannan et al., 2016; Leeflang et al., 2009; Naik and Peters, 2009).
|Keywords||Multichannel, attribution, marketing-mix, dynamic|
|Thesis Advisor||Fok, D.|
Verkroost, F.C.J. (Florianne). (2017, September 11). Estimating Multichannel Advertising Effectiveness using a Dynamic Integrated Marketing Model. Econometrie. Retrieved from http://hdl.handle.net/2105/39166