Procrustes problems have been studied since the second half of the 20th century, when the first problem was stated. Since then Procrustes analysis has been developed. However, some gaps still hold mainly in estimating missing values and the lack of tools for statistical inference. This work analyses the influence of putative values in the estimation of missing cells in ordinary Procrustes problems, reports and suggest new aspects for estimation algorithms in generalised Procrustes analysis and describes a decision method to allow inference features. Such method is illustrated with three practical experiments and two simulation studies. Inadequate putative values have shown to ability to lead to local minima and the decision method performed coherently, stably and efficiently in detecting similar products, associating a liberal and a conservative stage.