Bidding for Multiple Keywords in Sponsored Search Advertising: Keyword Categories and Match Types

• 成果类型: Article
• 署名作者: Du, Xiaomeng; Su, Meng; Zhang, Xiaoquan (Michael); Zheng, Xiaona
• 署名单位: Peking University; Chinese University of Hong Kong
• 刊物名称: INFORMATION SYSTEMS RESEARCH
• ISSN/ISSBN: 1047-7047
• DOI: 10.1287/isre.2017.0724
• 发表日期: 2017
• 页码: 711-722
• 关键词: position auctions; PAID; internet; strategy; MODEL; reviews; product; MARKETS; IMPACT; sales

摘要

Although keyword auctions are often studied in the context of a single keyword in the literature, firms generally have to participate in multiple keyword auctions at the same time. Advertisers purchase a variety of keywords that can be categorized as generic-relevant, focal-brand, and competing-brand keywords. At the same time, firms also have to choose how the keywords can be matched to search queries: exact, phrase, or broad. This study empirically examines how keyword categories and match types influence the performance of advertising campaigns. We build a hierarchical Bayesian model to address the endogeneity problem contained in the simultaneous equations of the click-through rate, the conversion rate, cost per click, and rank, and we use the Markov Chain Monte Carlo method to identify the parameters. Our results suggest that it is important to differentiate among the various bidding strategies for various keyword categories and match types. We also report results related to financial performance such as number of sales, profit, and return on investment for different keywords. These findings shed light on the practice of sponsored search advertising by offering insights into how to manage ad campaigns when advertisers have to bid on multiple keywords.