logo

ASSOCIATION FOR
INFORMATION SYSTEMS

AIS SAAISSAC

Abstract:

The aim of this study is to investigate the forces and mechanisms driving the search engine during query generation. The analysis further aims to improve understanding and design a conceptual model of intervention which will improve the search business and provide strategic indicators which can boost the performance of online businesses.

Search engines are commercial entities that require revenue to survive. The most prevalent revenue stream for search engines is sponsored search, where content providers have search engines service their links to users in response to queries or in a contextual manner on relevant websites. In exchange for providing this service, content providers pay search engines based on the number of clicks generated by the relevant advertisements. Web sponsored search remains one of the most profitable business models in the world of search. It accounts for the main income of the highly successful Google, and generates revenues of at least 25 billion dollars per year.

A key technical challenge in sponsored search is to select advertisements that are relevant to the user's query. Identifying relevant ads is challenging since queries are usually very short, and because users, consciously or not, choose terms intended to lead to optimal Web search results and not to optimal advertisements. Furthermore, the ads themselves are short and usually formulated to capture the reader's attention rather than to facilitate query matching.

The objective of this research is to investigate and identify the search engines users' preference between generic and branded keywords during query generation. It has been established that less than half of search engine users with a university qualification select search engine advertising results as being more relevant than organic results. This research sets out to establish a division between branded and generic keywords as generated by search engine users.

The research methodology includes a survey of relevant literature and a questionnaire. The questionnaire was designed, and a pilot study was done to test its accuracy. The final version of the questionnaire was launched on a busy South African website. Statistical analysis was done, based on the categories; gender, age, educational qualification, race, Internet experience, working field and others.

The results provide a clearly different pattern of preference between generic and branded keywords. Further investigation is in progress to finalize these findings. In this paper, the preference of keywords (branded or generic) in sponsored search is outlined. Clear patterns were discovered, closely linked to education, age, and other factors. Conclusions include that online advertisers should consider these results, as they reflect the views of a proportion of search engine users, and they may be generalized. These results may be used by business modelers, content providers, and researchers. It is considered to be a unique form of information retrieval - not just a mode of advertising for web owners.

Reference:

  • Sabwa, J.M. & Weideman, M. 2012.
  • Understanding choices of generic and branded keywords during query generation.
  • Working Paper.
  • Presented as a M&D Conference Paper at SAICSIT 2012, Pretoria, South Africa.
  • 1 October 2012.
  • http://cput.academia.edu/MeliusWeideman/Papers

 

Associations

Digital Libraries