Measuring the Effects of Advertising: The Digital Frontier
Online advertising offers unprecedented opportunities for measurement. A host of new metrics, clicks being the leading example, have become widespread in advertising science. New data and experimentation platforms open the door for firms and researchers to measure true causal effects of advertising on a variety of consumer behaviors, such as purchases. We dissect how the new metrics and methods can be used to provide insight on the returns to advertising by setting out the questions that we think can be answered by current data and methods, those that we believe will be in play within five years, and those that we believe could not be answered with arbitrarily large and detailed data. We pay close attention to the advances in computational advertising that are not only increasing the impact of advertising with algorithmic targeting, but also by shifting the focus from "who to hit" to "what do I get."
Much of this work was done when all the authors were at Yahoo! Research. We thank Garrett Johnson, Dan Nguyen, Sergiy Matusevych, Iwan Sakran, Taylor Schreiner, Valter Sciarillo, Christine Turner, Michael Schwarz, Preston McAfee, and numerous other colleagues for their assistance and support in carrying out the research.
Justin M. Rao
I have no funding or financial relationships to disclose.David H. Reiley
I currently work for Google, Inc., a company that earns substantial revenue from online advertising. I previously worked for Yahoo!, another online-advertising company, during the time that most of the work was done.
It is a matter of personal integrity for me to do unbiased scientific research to the best of my ability. I am grateful to Yahoo! and Google for affording me the opportunity to do this; I would not have left academia otherwise, because I care too much about learning the truth.