Identifying locations for targeted advertising on the Internet

Due to steady erosion in the effectiveness of on-line advertising (e.g., banners and buttons placed at frequently visited sites), on-line businesses need to target their ad campaigns more precisely to reach the segments they are interested in. This paper examines the issue of how to identify idea I paid advertising, banner exchange, or affiliate partner locations, and proposes a model based on consumer search behavior. Calibrated with data obtained From searches for information in 18 different categories, the model allows for heterogeneity by permitting consumers to belong to different segments that have idiosyncratic search points and thresholds. It also includes a segment-membership function, specified in terms of consumer demographics, that con be used to identify the demographics associated with different focal groups.