There is a need in the development and validation of inverse methods for NDE inspection problems to demonstrate capability using quantitative performance metrics. An overview of metrics is presented addressing the condition of the ill-posed problem, the quality of feature extraction algorithms, and performance of inversion algorithms. In particular, the relationship between metrics found in estimation theory, linear algebra and statistics are presented including Fisher Information, Cramer-Rao Lower Bound (CRLB), covariance, singular value decomposition (SVD), and variance inflation factors (VIFs). The connections and utility of these metrics are illustrated through examples of estimating material loss (thickness), conductivity, and lift-off.