In the last few years, concept factor analysis has been an object of study in the FCA community. Its main idea is to use formal concepts as factors to explain the data in a more concise way. We study factorisation of graded tabular data by means of well-structured families of concepts which have an ordinal character. This method enables us to obtain a smaller number of items which explain the data while they still have a clear and comprehensible meaning. We illustrate the method and its applicability on a sports data set.