Looking at Govaert's book and his breakdown of PCA: "It is often considered as the basic method of factor analysis, which aims to find linear combinations of the variables called components used to visualize the observations in a simple way." I think the idea of the simple way really sells how PCA could be considered the best statistical method. He also breaks down various others, a rather informative book! Processing the data I think comes down to what are we looking to come of the data we are processing. Do we need to show debts vs equity or are we showing a correlation of sales of generators and weather conditions?