Consumer search and firm strategy with multi-attribute products

I analyze a model of directed search in which a consumer inspects a finite number of products sharing attributes with each others. The consumer discovers her valuation for the attributes of the inspected products and adapts her search strategy based on what she has learned. The consumer anticipates the optimal paths that arise after different realizations; this generates a search rule that accounts for learning systematically. In this search environment, a multiproduct seller commits to a menu of horizontally differentiated products. The seller can exploit the fact that the emerging search paths reveal the consumer’s preferences: by setting different prices for ex ante identical products, the seller can encourage specific paths to arise and exploit the information that the consumer learned through search. In some cases, the seller optimally limits the set of available products.