Purpose - Despite prior studies, little has been done to understand the advertising carry-over effect. The purpose of this study is to investigate the heterogeneous attributes of the carry-over effect by focusing on the differences between franchise and non-franchise firms. Design/methodology/approach - The data were retrieved from the Compustat database and annual corporate financial reports (10-K) for five representative franchise industries from 1980 to 2009. Ultimately, 185 firms were included and 1,592 firm-year observations were analysed. This study used a Panel VAR(Vector Autoregression) to examine the effects of advertising on firm performance. We can control endogenous effects using Panel VAR, which also allows us to control unobserved firm-specific heterogeneity. Findings - This study found that advertising had no effect on sales growth or brand equity in the long run for non-franchise firms and further confirmed that non-franchise firms incur agency costs. In contrast, the effect of advertising on sales growth and brand equity was significant for franchise firms. A carry-over effect of advertising on brand equity was detected for franchise firms, but sales growth rapidly decreased after two years. Practical implications - Even though franchise firms retain the carry-over effect of advertising on brand equity, franchisors should carefully monitor market trends and their sales growth because sales growth rapidly decreased after two years. Originality/value - This study incorporated the concepts of the delayed response effect and the customer holdover effect to better understand the advertising carry-over effect. From the results, this study proposed a more detailed concept of carry-over effects for further studies. From this perspective, this study makes both academic and industrial contributions.