The problem of assessing customers satisfaction through questionnaires

Abstract

The task of measuring customers satisfaction is a complex one. Questionnaires are generally used, not just to assess the level of customer satisfaction in a given moment, but to grasp its evolution over time as well. In this case, questionnaires must be repeated periodically. These questionnaires are generally based on a number of questions or statements whose answers should be fitted in a Likert scale. The present article addresses two issues related to the employment of this type of questionnaires: (i) the use of control charts in monitoring the results and (ii) the interpretation of significant changes in the satisfaction profile. It is concluded that the modified p and the 2 control charts should be used together, since they are not able to detect statistically significant changes in the satisfaction profile under some circumstances. Considering the difficulties that arise in interpreting the occurrence of an out-of-control point, a new method based on value concepts is hereby referred as a tool to help decision-makers.