Digital media advertising attribution in multi-channel and multi-media contexts

Mr Zhao Yunkun
Dr Goh Khim Yong, Associate Professor, School of Computing

  15 Mar 2019 Friday, 02:00 PM to 03:30 PM

 Executive Classroom, COM2-04-02


Advertising media attribution has long been viewed as a difficult topic in terms of investigating customers' conversion behavior throughout their purchase funnels. Due to the digital technology advancements, companies are now able to track customers' footprints in digital platforms. They can attribute customer conversions to various advertising media exposures more accurately and precisely, even within multi-channel and multi-media environments. These have raised many interesting but unresolved media attribution questions.

In this dissertation, I propose two studies to evaluate and quantify multi-channel multi-media attributions in different contexts. The first study relates to the omni-channel context. Omni-channel retailers face the challenging tasks on how to effectively manage various customer touchpoints to influence customer demand. This study empirically examines the relative effectiveness of different touchpoints in terms of owned, paid, and earned media, the spillovers of touchpoint exposures from/to focal and rival brands, and the interdependencies among different touchpoints. Using data from an international omni-channel retailer, I conduct empirical analyses on customers' purchase expenditure, purchase quantity and purchase choice. Results show that owned (paid) media touchpoint exposure is more effective than that of paid (earned) media in influencing customers' purchase expenditure, quantity, and choice. In addition, owned media touchpoint exposures promoting rival brands' advertising campaigns may dissuade customers from purchasing the focal products. This study also unravels a substitution relationship between owned and paid media touchpoints, and a complementary relationship between paid and earned media touchpoints, in affecting customers' demand for the focal products. The second study is relevant to online-to-offline (O2O) context. Multi-brand multi-product O2O companies are concerned about how to make their online advertising exposures of different strategies more influential in boosting demand/sales, whereas in fact, the effectiveness of online advertising strategies are subject to offline sales interventions. This study seeks to empirically analyze the effectiveness of online advertising of different strategies (e.g., brand- and product-oriented advertising), the cross-channel interactions with salesman skill attributes, as well as the cross-channel interactions with product line management strategies. The results show that a salesperson's completion of brand sales training can further increase the effectiveness of online product-oriented advertising. Moreover, product line length across different brands can decrease the impacts of online brand-oriented advertising on customer store-visits. Likewise, product line length within the same brand can increase the impacts of brand-oriented advertising on store transactions, but can decrease the impacts of product-oriented advertising on customer store-visits.

Overall, these findings offer novel and important implications for the theory of digital media attribution and the practice of information channel marketing within different multi-channel multi-media environments