Customer satisfaction and loyalty assessment by means structural equation models

Customer satisfaction is a key issue for the organizations in the today's competitive market. As such, much research and a lot of revenues have been invested in developing accurate ways of assessing consumer satisfaction at macro (national) or micro (organizational) level, facilitating comparisons of the performances between industries.
To this purpose since 1994 different national customer satisfaction indexes (CSI) have been proposed. American Customer Satisfaction Index (ACSI) and European Customer Satisfaction Index (ECSI) are the two most popular indexes. Their implementations are usually developed at the macro level, and suitable applications at micro level have not yet been proposed. This absence is the main cause of a large proliferation of different models that, in many cases, do not make possible the comparison of CSI within and between industries.
To enhance the comparison properties of this index an adaptation of ACSI and ECSI models to a specific economic sector is proposed. Furthermore, according to [18], the antecedent "expectation", not significant at the micro economic level [13], was removed and two new antecedents, "belief" and "leadership" were introduced.
The goodness of the estimated model was confirmed by an application to a network of 250 firms operating in the building retailers. The PLS (Partial Least Squares) approach has been proposed for the estimation phase. Data were collected during the summer of 2009 by means of a 15 items questionnaire given to 250 managers.