The moderating effects of keyword competition on determinants of ad position in sponsored search advertising (2023)

Advertisers face difficult decisions when bidding on keywords in their search engine campaigns. This document provides guidance by empirically examining the decisions of sponsored search advertisers, a search engine, and consumers across multiple keywords with different characteristics in an empirical and comprehensive manner. Specifically, we study online merchants' bidding patterns, search engine ranking decisions, and consumer click patterns for various related keywords using Tobit's simultaneous equations fixed effects models. Our results show how advertiser bids and consumer clicks are related across multiple keywords. Furthermore, our optimization results suggest that advertisers can allocate advertising budgets more efficiently when keyword interrelationships are taken into account.

In paid search advertising, companies need to make decisions about a matching strategy between broad match and exact match to reach their current and potential customers. Broad match expands the targeted area, while exact match allows for more precise targeting. This thesis examines the relative effectiveness of matching decisions using secondary data and data from a field experiment. Both studies show that broad match keywords result in lower ad performance compared to exact match keywords. Additionally, the negative effect of broad match on CTR is more pronounced for more specific or lower (better) ranking keywords, and the impact of broad match on CR, sales, and revenue is less negative for keywords with higher (worst) rating. These results suggest that paid search advertising campaign managers should consider keyword specificity and position when making optimal match strategy decisions for their keyword portfolios.

We examine a variant of the display ad attribution problem, where an online publisher must decide which subset of ad spaces to use to fulfill guaranteed contracts and which subset to sell on delivery platforms (SSPs) to meet expectations to maximize revenue. . Our modeling approach also considers the uncertainty associated with selling an impression through an SSP. The way information is revealed over time allows us to model the display ad attribution problem as a two-stage stochastic program. We refer to our model as the stochastic programming (SP) model. Numerical experiments show that the SP model works well in most cases. We compare the SP model solutions with the solutions of an allocation policy used in practice (priority assignment heuristic (PA)). We found that the performance gap between the PA heuristic and the SP model depends on the proportion of total impressions that should be allocated to guaranteed contracts. The results indicate that the benefit of using the SP model is greater during periods of high website traffic compared to guaranteed contract targets.

Many online retailers provide stock availability information on their websites and use sponsored search advertising to drive traffic to their retail websites. Increased user engagement over the Internet is encouraging retailers to move to Sponsored Search Advertising. In this article, we designed a decision support model to enable a retailer selling a short-lived, perishable product to make strategic pricing and supply decisions using Sponsored Search Advertising to drive customers to the site. The retailer complements Sponsored Search offerings with dynamic pricing in a multi-period stochastic dynamic planning framework. Our analysis shows that it is ideal for the retailer to invest heavily in bidding when inventory is low, while using price action to improve profits when inventory is high. We also found that the optimal bid and price increase as the average increases and are subject to customer price variability.

Information Processing and Management, Volume 50, Number 4, 2014, pp. 508-523

With sponsored search, many advertisers have not performed as expected, while the search engine also has a lot of room to improve its earnings. In particular, due to weak keyword bidding, many advertisers cannot survive competing ad auctions to get the ad impressions they want. Meanwhile, a significant fraction of searches do not display ads on their search results pages, even though many of them have commercial value. We suggest recommending a set of relevant but less competitive keywords to an advertiser. Thus, the advertiser may have the opportunity to earn some advertising space (originally empty) and collect several impressions. At the same time, search engine revenue can also increase as many empty ads are filled. Mathematically, we model the problem as a mixed integer programming problem that maximizes advertiser revenue and recommended keyword relevance while minimizing keyword competition, subject to bid and budget constraints. By resolving the issue, we can offer an advertiser an optimal set of keywords and their optimal bid prices. Simulation results showed that the proposed method is very effective in increasing ad impressions, expected clicks, ad revenue and search engine revenue.

International Journal of Information Management, Banda 38, Número 1, 2018, S. 107-116

The study shows how harmful digital marketing is when done by unskilled service providers. It shows how sponsored search engine marketing (SEM) services are not as successful as they seem and sometimes negatively impact business. This study uses social network analysis to derive information from Twitter using descriptive, content, and network analysis. Methods like hashtag analysis, polarity and emotion analysis, word analysis, topic modeling, and other relevant approaches were used to extract the user-generated content. A qualitative case study in an electronic marketplace serves to validate the results. SEM services provided by small organizations and freelancers are not as beneficial as those provided by established players. The services provided by these companies turned out to be harmful for the customers based on the user experiences of these services on social media and discussions on specific forums. This study shows how SEM often not only fails to deliver, but also destroys value if not done correctly. Transaction costs such as agency issues, coordination costs, non-contractual depreciation and adjustment costs are also identified with possible consequences that affect long-term benefits. The entries will be useful for SEM planning and outsourcing practice.

Journal of Interactive Marketing, Band 28, Número 4, 2014, S. 285-301

Despite its importance to both e-commerce and traditional offline providers, managing paid search campaigns is often based on trial and error. In particular, neither research nor practice has dealt extensively with identifying relevant keywords and setting up appropriate matching options. The authors develop a model based on the intrinsic and extrinsic information content of a keyword to clarify how keyword properties affect campaign performance. Intrinsic information includes linguistic aspects and functions related to the user and content that are indirectly determined by extrinsic information, such as B. Matching options can be changed. Using an advertiser-level dataset across multiple industries, this study assesses the impact of these criteria on conversion and click-through rates. The authors found that the query variance index, which measures whether a keyword contains enough information to correctly identify a user's information needs, is an effective indicator of keyword performance. Furthermore, they show that the relationship between two of the most important predictors, query variation and advertiser names, and keyword performance is moderated by the advertiser's choice of matching options.

Decision Support Systems, Bd. 96-106

Sponsored search advertisers face a number of complex decisions when planning and executing a new sponsored search advertising campaign. These decisions include choosing keywords, defining landing pages, and formulating bidding strategies. Assisting with keyword selection has received relatively little attention in recent research, and most studies on sponsored search advertising focus on formulating bidding strategies and budgeting strategies. We present a new approach to automatically generate sponsored search keywords based on the theory of consumer search behavior. Our approach uses an online store's internal search history to extract the keywords used by consumers in their search process, as recent research has shown that consumers who are particularly likely to convert and who have search patterns are targeted and the non-exploratory buy internal search engine. . We empirically tested our approach using a store's internal search engine and identified the impact of this approach compared to a state-of-the-art approach. Our analysis shows that our approach significantly increased the number of profitable keywords, improved the in-store conversion rate by approximately 41%, and reduced the average cost per click by over 70%.

Information Processing and Management, Volume 54, Number 2, 2018, pp. 175-183

Sponsored Search is an online advertising channel that has grown in importance around the world. The biggest challenge is determining the types of keywords to bid on and the appropriate options to use for the keywords. In this white paper, we address this issue by providing a comprehensive analysis of how various search traffic metrics (length, CTR, average cost per click (CPC), average position, and quality score) affect your bidding results when you bid. wide, i.e. exact H., too much sentence, too broad. Building on purchase objective theory, we also profiled the metrics associated with more focused search intent on matching options. Using a random sample of keywords selected from 9,640 keyword bids by online advertisers across multiple markets, the results show that overall keyword traffic metrics increase as the match option decreases, i.e. , H. from broad to precise. , except for cost, which does not differ significantly between match options. Longer keywords, typically associated with more focused search intent, generate more clicks and have a higher quality measure on average across all matching options. Longer keywords are cheaper for the exact match option, but more expensive for the other match options. The results are not meaningful in terms of the position of the longest keywords on a results page in the match options. Therefore, narrower match options and longer match keywords that a customer would typically use to search for a company's products and services should be targeted to ensure more visits to the company's website.

Information and Management, Volume 54, Edition 4, 2017, pp. 531-543

This study proposed a topic-based competitive keyword suggestion method called TCK to improve search engine advertising. Based on query logs, the method examines indirect associations between keywords and extracts hidden topic information to identify competitive keywords. It can help advertisers not only expand keyword options, but also implement a competitive search advertising strategy. Extensive experiments were performed to demonstrate the effectiveness of the proposed method. The results show that the proposed method performs better than existing keyword suggestion methods and therefore contributes a lot to the keyword suggestion advertising market.