Optimal Keywords Grouping in Sponsored Search Advertising Under Uncertain Environments

ABSTRACT

In sponsored search advertising, advertisers need to make a series of keyword decisions. Grouping these keywords to form several adgroups within a campaign is a challenging task because of the highly uncertain environment of search advertising. This paper proposes a stochastic programming model for keywords grouping, taking click-through rate and conversion rate as random variables, with consideration of budget constraints and advertisers’ risk-tolerance. A branch-and-bound algorithm is developed to solve our model. Furthermore, we conduct computational experiments to evaluate the effectiveness of our model and solution, with two real-world data sets collected from reports and logs of search advertising campaigns. Experimental results illustrated that our keywords grouping approach outperforms five baselines, and it can approximately and steadily approach the optimal solution. This research generates several interesting findings that illuminate critical managerial insights for advertisers in sponsored search advertising. First, keywords grouping does matter for advertisers, especially with a large number of keywords. Second, in keywords grouping decisions, the marginal profit does not necessarily show the marginal diminishing phenomenon as the budget increases. Therefore, advertisers should try to increase their budget in keywords grouping decisions to garner additional profit. Third, the optimal keywords grouping solution is the result of a multifaceted trade-off among various advertising factors. In particular, assigning more keywords into adgroups or having a larger budget will not definitely lead to higher profits. This study suggests a warning for advertisers: It is not wise to use the number of keywords as a single criterion for keywords grouping decisions.

KEY WORDS AND PHRASES:

• adgroups
• chance constrained programming
• keyword decisions
• keywords grouping
• online advertising
• sponsored search advertising

Acknowledgments

We are thankful to anonymous reviewers who provided valuable suggestions that led to a considerable improvement in the organization and presentation of this paper. This work was partially supported by National Natural Science Foundation of China grants (71672067, 71810107003).

Notes

1. Note that CTR and CVR might be, more or less, improved through keywords grouping strategies, however, which is not the ultimate goal—but intermediary performance indexes—for advertisers. Moreover, neither CTR nor CVR provides a comprehensive clue for the ultimate advertising goal [27]. In this work, we distinguish the CTR (CVR) inherited in keywords themselves and the CTR (CVR) raised by adgroups, and use the product of them to represent the CTR (CVR) of a keyword assigned in an adgroup.

2. We implicitly assume that the parameter $d_i$ is reasonably large so that the two conditions just given are satisfied.

3. $ROI=E\left[{\sum }_{j=1}^m{\sum }_{i=1}^n{x}_{ij}{d}_i{c}_{ij}\left(r_{ij}{v}_i-p_{ij}\right)\right]/E\left[{\sum }_{j=1}^m{\sum }_{i=1}^n{x}_{ij}{d}_i{c}_{ij}{p}_{ij}\right]$. The ROI is defined as the expected profit divided by the expected total cost.

Additional information

Notes on contributors

Huiran Li

Huiran Li ([email protected]) is a Ph.D. candidate with the School of Management, Huazhong University of Science and Technology. She is also a researcher in the Internet Sciences and Economic Computing research center at the university. Her current research interests include computational advertising and online advertising.

Yanwu Yang

Yanwu Yang ([email protected]) is a full professor and the head of the Internet Sciences and Economic Computing research center at the School of Management, Huazhong University of Science and Technology, China. He received his Ph.D. degree in computer science from the graduate school of École Nationale Supérieure des Arts et Métiers, France. His research interests include e-commerce, recommender systems, web personalization, and computational advertising.