Measuring the Effect of Brand Advertising on Search Behavior

• Activity Number: 313
• Type: Invited
• Date/Time: Tuesday, August 5, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Marketing
• Author(s): Colin McCulloch and Madeleine Cule*+ and Wai Liu and Sheng Ma and Pezhman Firoozfam and Yonathan Schwarzkopf and Tim Hesterberg
• Companies: Google and Google and Google and Google and Google and Google and Google
• Keywords: zero inflation ; sparse data analysis ; computational advertising

Abstract

Brand advertisers are mainly interested in spreading awareness about their brand. Traditional internet advertising performance metrics like click-through rates, conversion rates, and cost per conversion are not very relevant for brand advertisers. We are developing new metrics for measuring brand awareness based on searches on google.com for keywords relevant to the advertiser. In doing so, we face a number of statistical challenges: (1) although overall search volume is high, searches for the relevant keywords are very rare, (2) achieving proper randomization is difficult, and (3) we are measuring causal effect under imperfect randomization.