Two models of international country segmentation

The growth of global competition has established international segmentation as a key issue in developing, positioning and selling products throughout the world (Ter Hofstede, Steenkamp and Wedel 1999). Many international segmentation studies have used macro-level, secondary data to identify country clusters based on similarities in political, economic, geographic or cultural variables. As a result of extensive review, we identify three major gaps in the international country segmentation literature. First, no study so far has accounted for the influence of time. While researchers suggest that longitudinal analysis provides additional insight into whether situational characteristics of countries change over time (Cavusgil, Kiyak, and Yeniyurt 2004; Helsen, Jedidi, and DeSarbo 1993; Sethi 1971; Steenkamp and Hofstede 2002,), a major limitation of this body of work is that most studies address country-level segmentation at a single point in time. However, bases of segmentation are considered to be dynamic in nature (Hassan, Craft, and Kortam 2003) and global and country-specific changes in economic development are likely to result in variations in segment membership over time. We investigate the stability of factors and the stability of segments over time by performing cluster analysis at two points of time. Second, most studies use ad hoc variables without theoretical basis which may result in accidental generalizations. Instead of suggesting a proliferation of random variables, which are considered influential in the decision making process without any empirical or theoretical evidence, we propose a theoretical basis for country segmentation. We use institutional theory to distinguish between heterogeneous groups of countries. Finally, there is the issue of providing "one size fits all" solutions.