In this research, the factors determining the purchase of a product that has been launched as the line extensions of different retail business formats are clarified not only through the purchases of competitor brands, but also through the views of the webpage of the parent brand.Research on the evaluation of line extensions has clarified such relationships as the success or failure of an extension brand and the strength of the parent brand (Reddy et al., 1994), trial and repeat purchases of the extension brand and purchase frequencies of the parent brand (Kim and Sullivan, 1998), the cannibalization of the parent brand by the extension brand (Lomax and McWilliam, 2001), and the interdependence of the extension brand and the parent brand (Sinapuelas et al., 2010). Previous analyses of the factors determining a line extension and their evaluations assumed that consumer purchasing behavior took place within the same retail business format and within a store; as a result, these analyses focused on consumer purchase history data. In this research, in addition to data on the purchase histories of consumers, an attempt was made to ascertain the factors determining a line extension for the horizontal actions of consumers across different retail business formats and stores by applying single-source data, supplemented by data on the access histories of targeted consumer webpages. Specifically, with the specialty coffee shops of Tully’s Coffee (subsequently, TS) and Starbucks Coffee (subsequently, SB) as subjects of the analysis, a binominal logistic regression analysis was applied to investigate the question of whether the purchase or non-purchase of products that these shops launched for convenience stores could be explained by the types of views of the webpages of each of the coffee shops that became the parent brands. The analysis employed single-source data on purchase histories and webpage access histories within i-SSP (Intage Single Source Panel), provided by Intage Inc. The investigation period was the 10 months, from October 2012 to July 2013. As a result of this analysis, the following were established: (1) For both TS and SB, purchases of the extension brands could not be promoted only from the views of the webpage of the extension brands; (2) for both TS and SB, the views of the product webpages of the parent brands promoted purchases of the extension brands; and (3) for SB only, the views of its shop webpage promoted purchases of its extension brand. These are findings that cannot be established using only data on either consumer purchase histories or webpage-access histories. In an analysis of a line extension, they can be ascertained only from webpage access that is related to the parent brand from the direction ascertained for past purchases of competitor products.

• Takashi Teramoto(Meisei University)