However, such measures themselves are useless unless these are related to some specific factors of service delivery. These could be:

-   Quality of Service- Range/suitability/pricing etc. (QS)

-   Quality of People - Operator assistance etc. (QO) ...and so on.

Thus, the perceived & expected quality of the above measures can also be obtained in comparison with the benchmark competitor. Once this is done, we can relate the overall satisfaction measures with each of the factors identified above. The one that is most correlated will obviously be the most important measure in determining overall satisfaction.

A great advantage of undertaking this analysis is that it identifies the factor that is critical to the success of a customer satisfaction program. Sometimes, such analysis throws up useful insight in customer behavior not overly identifiable. In one study on satisfaction with telephony services, it was identified that the overall satisfaction was related negatively with behavior of engineer. The root cause was that the service & sales departments operated independently to the detriment of company image. The salesmen promised certain performance characteristics while the service engineer explained the flaw in the promise and seemed to help the customer by visiting often. However, this led to greater dissatisfaction (implying, more happy is the customer with the engineer less happy is he with the company). The subjective parameters that may be used to determine quality of service among telcos may be as follows: (table to the right)