In this chapter, the authors present a method that is useful for the evaluation of the importance of an explanatory variable. The method may be applied for several purposes. It includes model simplification, model exploration, domain-knowledge-based model validation and knowledge generation. The methods for assessment of variable importance can be divided, in general, into two groups: model-specific and model-agnostic. For linear models and many other types of models, there are methods of assessing explanatory variable’s importance that exploit particular elements of the structure of the model. The plots of variable-importance measures are easy to understand, as they are compact and present the most important variables in a single graph. The bars in the plot indicate the mean values of the variable-importance measures for all explanatory variables. Variable-importance measures are a very useful tool for model comparison. In this chapter, the authors use the dalex library for Python. It is available on pip and GitHub.