This article critiques Baron and Kenny's (1986) mediation analysis framework, which has been widely used in social sciences, including consumer research. While their method is well-known and often reflexively applied, it has limitations that have not been widely recognized by practicing researchers. The authors argue that Baron and Kenny's approach is flawed and that its strict criteria can lead to the premature rejection of promising research projects and the rejection of papers that could contribute to theory development. They propose a more nuanced understanding of mediation, emphasizing the importance of considering both direct and indirect effects, and suggest that the Sobel test for indirect effects is less powerful than alternative methods like the bootstrap test. The article highlights that mediation is not always characterized by the absence of a direct effect, and that the presence of a direct effect can provide valuable insights for theory building. It also challenges the necessity of a significant "effect to be mediated" (zero-order effect of X on Y) in establishing mediation, arguing that this is not a requirement. Instead, the focus should be on the significance of the indirect effect. The authors present a typology of mediation and nonmediation, distinguishing between complementary and competitive mediation, and emphasize the importance of interpreting the sign of the direct effect for future research. They also advocate for the use of bootstrap tests over the Sobel test for more accurate and powerful mediation analysis. The article concludes that a more comprehensive understanding of mediation, incorporating both direct and indirect effects, is essential for advancing theory and research in social sciences.This article critiques Baron and Kenny's (1986) mediation analysis framework, which has been widely used in social sciences, including consumer research. While their method is well-known and often reflexively applied, it has limitations that have not been widely recognized by practicing researchers. The authors argue that Baron and Kenny's approach is flawed and that its strict criteria can lead to the premature rejection of promising research projects and the rejection of papers that could contribute to theory development. They propose a more nuanced understanding of mediation, emphasizing the importance of considering both direct and indirect effects, and suggest that the Sobel test for indirect effects is less powerful than alternative methods like the bootstrap test. The article highlights that mediation is not always characterized by the absence of a direct effect, and that the presence of a direct effect can provide valuable insights for theory building. It also challenges the necessity of a significant "effect to be mediated" (zero-order effect of X on Y) in establishing mediation, arguing that this is not a requirement. Instead, the focus should be on the significance of the indirect effect. The authors present a typology of mediation and nonmediation, distinguishing between complementary and competitive mediation, and emphasize the importance of interpreting the sign of the direct effect for future research. They also advocate for the use of bootstrap tests over the Sobel test for more accurate and powerful mediation analysis. The article concludes that a more comprehensive understanding of mediation, incorporating both direct and indirect effects, is essential for advancing theory and research in social sciences.