Measuring the magnitude of customers’ satisfaction for some commercial and health activities is of practical meaning and general interest to business sectors and the public society. Traditional methods for this purpose require relatively complicated statistical procedures, expert knowledge, expensive sampling, and time-consuming data collection stages and are not open to public users. We introduce some basic concepts for judgement theory, and propose two statistical methods for the analysis of such data, one un-regularized which is more natural and simpler to use. However, when the observation points are not dense enough, the variation in the estimated curve can be large and the un-regularized estimation may not have enough smoothness. A common way to overcome this problem is to use a regularized estimation, to force the estimator to have desired smoothness.

For the un-regularized method, by using a carefully chosen population representation function, a very simple tool is constructed to measure the customers’ extent of satisfactory. Although the method is of general purpose, at current stage we implement and illustrate its use for measuring customers’ satisfactory for some penalty-awarding scheme for online purchasing activity.