Measuring Consumer Sensitivity to Advertising: A Field Experiment on Pandora Internet Radio
How to cite
Reiley, David. 2018. "Measuring Consumer Sensitivity to Advertising: A Field Experiment on Pandora Internet Radio." AEA RCT Registry. AEARCTR-0002932. https://doi.org/10.1257/rct.2932-1.0
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
A randomized experiment with almost 35 million Pandora listeners enables us to measure the sensitivity of consumers to advertising, an important topic of study in the era of ad-supported digital content provision. The experiment randomized listeners into nine treatment groups, each of which received a different level of audio advertising interrupting their music listening, with the highest treatment group receiving more than twice as many ads as the lowest treatment group. By keeping consistent treatment assignment for 21 months, we are able to measure long-run demand effects and compare them to the short-run effects we might obtain with only a month-long experiment. We estimate
a demand curve: number of hours listened as a function of the number of ads per hour received. We also measure the impact on the number of days listened and on the probability of listening at all in the final month of the experiment. We separately vary the number of commercial interruptions per hour and the number of ads per commercial interruption, to measure the extent to which these influence consumer listening behavior. Lastly, we measure the impact on demand for a substitute good: the number of paid subscriptions to the ad-free version of Pandora.
Intervention
We vary the audio ad capacity for each individual listener.
Experimental design
3x3 experimental design. We separately vary the number of commercial interruptions per hour (3, 4, or 6) and the number of ads per commercial interruption (1, 1.5, or 2, where 1.5 means alternating between 1 and 2). We also have a control group 10x as large as each treatment group, and this control group matches the values at the center of the matrix (4x1.5), matching the previous default values listeners received at Pandora before the experiment.
Design details
Cited by
No citation data available for this registration yet.