In die and mould manufacturing, the method used for quality control of finished surfaces is usually visual and tactile inspection of the finish, which is not easily quantifiable. In the present study, an evaluation of the connection between surface finish appearance and measured surface roughness was carried out using scale-sensitive fractal analysis to find functional correlations and to determine suitable cut-off limits for functional data filtration. A selection of ball-nose end-mills in combination with two different tool steels (hardness 39 and 47 HRC) were used to manufacture surfaces that were inspected and measured. It was found that the method developed in the present study for evaluating functional correlations and designing filters worked well. It was also found that there is a correlation between the surface roughness parameter Sq and the surface finish appearance and that this correlation is stronger in certain wavelengths on the surface.