A dilemma in developing line extensions is that there is often a trade-off between meeting the changing needs of the category and maintaining a high level of similarity to existing products. To help manage this trade-off, this study examines the relationship between the line extension's level of difference to existing products and its market share performance. A consumer scanner data set containing 318 line extensions provided input for a regression model, aimed at explaining the market share performance of line extensions. The findings not only demonstrate that an inverted U-shaped relationship existed between a line extension's level of difference and its performance, but also that this relationship changes with each entry position. More specifically, it appears that the first two products in a brand's range play an important role in defining the boundaries of difference and with each subsequent line extension, the optimal level of difference declines. In addition, the study has demonstrated the significance of the Gower distance coefficient as a measure of product difference derived from marketplace data. For managers, the study provides numerous insights into the simultaneous effects of product difference and other explanatory variables on the performance of their line extensions.