This paper discusses the development and application of goodness-of-fit indices (GoF) in partial least squares (PLS) path modeling. The authors introduce the GoF and the relative GoF (GoFrel), which are designed to assess the overall fit of PLS models. However, they find through simulations that these indices are not suitable for model validation, as they do not effectively distinguish between valid and invalid models. Instead, the GoF can be useful for evaluating how well a PLS model explains different sets of data. The paper also highlights the differences between PLS path modeling and covariance-based structural equation modeling (CBSEM) in terms of their objectives and fit measures. While CBSEM focuses on minimizing discrepancies between observed and implied covariance matrices, PLS path modeling aims to maximize explained variability. The authors conclude by suggesting that researchers should carefully evaluate path coefficients and their significance rather than relying solely on GoF and GoFrel for model validation.This paper discusses the development and application of goodness-of-fit indices (GoF) in partial least squares (PLS) path modeling. The authors introduce the GoF and the relative GoF (GoFrel), which are designed to assess the overall fit of PLS models. However, they find through simulations that these indices are not suitable for model validation, as they do not effectively distinguish between valid and invalid models. Instead, the GoF can be useful for evaluating how well a PLS model explains different sets of data. The paper also highlights the differences between PLS path modeling and covariance-based structural equation modeling (CBSEM) in terms of their objectives and fit measures. While CBSEM focuses on minimizing discrepancies between observed and implied covariance matrices, PLS path modeling aims to maximize explained variability. The authors conclude by suggesting that researchers should carefully evaluate path coefficients and their significance rather than relying solely on GoF and GoFrel for model validation.