Absorption correction is today a highly automated process performed in the course of data reduction, on most diffractometer software suites directly after the measurement. In the utmost cases, this process is not handled with the required care, and by this considerable systematic errors can be introduced into the data sets. The impacts of missing or wrong absorption corrections are subtile but inflict all structure parameters, from standard deviations of atomic parameters and residual electron densities to R values. It is of high importance to rationalise functionality and methods of absorption correction in order to avoid systematic errors. A brief overview is given on the theoretical background and different methods of absorption correction algorithms. Some examples from our research on amalgams showing unusually high absorption effects are given to illustrate problems of absorption and how to overcome them.