Deguchi Lab. LogoDeguchi Lab. Sub Logo

Home > Research > Analysis on Website Advertising Strategy in Rental Housing Market Using Agent-Based Simulation (Yagisawa 2012)

Analysis on Website Advertising Strategy in Rental Housing Market Using Agent-Based Simulation (Yagisawa 2012)

In recent years, under the growing of Internet users and the development of electronic devices, number of people who search housing information on Internet is dramatically increasing. Keyword targeted advertisement is one of the most effective website marketing strategies of housing information websites since it is directly responsive to users’ searching keywords. We construct an agent-based model of Internet housing service, in which Internet user, search website, housing information website agents interact with each other’s. We analyze dynamical change of keyword targeted advertisement by the housing information website agent in response to user searching behaviors and its rivals’ reaction. The research contributes a method of choosing potential website marketing strategy to administrators of Internet housing information websites. SOARS was used to construct this model.

Details

In recent years, under the growing of Internet users and the development of electronic devices, number of people who search housing information on Internet is dramatically increasing. Keyword targeted advertisement is one of the most effective website marketing strategies of housing information website since it is directly responsive to users’ searching keywords.

In a previous research related to keyword targeted advertisement, based on an analysis on bidding price history of certain keywords, an equation of bidding price, expected conversion number and advertising cost per day has been established. However, in the market of keyword targeted advertisement, competition between bidders in acquiring powerful keywords plays an important role in deciding the distribution of advertising cost of the market. In order to analyze effect on profit of one bidder from changes in CPC (Cost Per Click) bid of its competitors’ in internet housing service, we construct an computer simulation model of a artificial internet rental housing market in which internet users, search website, housing information websites interact with each others.

In the model, Internet users search housing information via search website and click on advertising banners of 5 housing information websites. After visit the website, users require for materials of favorite rental properties. The number of requests is called “conversion number” which is proportional to revenue of the housing information website. Activity of Internet users was captured based on data of their activities in 5 websites (date and time of visit, location, visited urn, etc.). Search website decides vertical position of advertising banners of 5 housing information websites based on CPC bids of them for powerful keywords of rental housing. Competition of those information websites in acquiring top positions in search website under powerful search keywords is modeled under 12 different scenarios. Analyses on simulation results imply that loosing top positions of more than 2 powerful keywords to its competitors could result in significant decrease in conversion number of the housing information website.

The research contributes a method of analyzing potential strategy in choosing keywords for keyword-targeted advertisement to administrators of Internet housing information websites. The future work is to build a gaming simulation model and provide it to company marketing staff as an education tool to study about website marketing strategy.

to Page top