Most well-designed customer satisfaction surveys contain a series of "attributes," which are rating scales of a series of specific statements or questions (courtesy, accuracy, timeliness, etc.). Naturally, some "attributes" will be more important than others. There has been a long-standing controversy in the Research Industry about how to know the "importance" of attributes used in a customer satisfaction survey. There are two basic choices:

1. "Stated Importance," determined by asking customers how important an item is,

2. "Derived Importance," determined by calculating the relationship between attributes and satisfaction.

We favor the derived approach. Asking importance adds unnecessary questionnaire length (which irritates respondents) and provides answers which, in and of themselves, may lead you astray. The results of asking customers what is important are useful if combined with the derived approach, but could well result in erroneous customer satisfaction improvement strategies if used without also deriving importance. This can happen if what customers say is important is a prerequisite to competing in the marketplace. In other words, it is already being done well by your company and by your competitors. For example, bank customers may say that the most important item is to not lose the money in their accounts. As this generally is not a problem in the banking industry, it would be misleading to formulate a customer satisfaction improvement strategy around the "don't lose my money" attribute.

The derived approach will uncover items which are most important to the satisfaction of customers. These attributes will not always be the same attributes that a customer would identify as being most important. But they would be the ones which, if improved upon by a business, will result in higher levels of satisfaction. Taking the bank example a step further, it would not be unusual to find that the most important customer satisfaction attribute is something along the lines of "courteous treatment of customers," since banks vary quite widely in this regard, and most people respond very favorably to courteous treatment.

It is not difficult for someone who understands statistical analysis to calculate the relative importance of a series of attributes, provided that a questionnaire also includes a satisfaction measure of some sort. The basic process is to conduct a correlation analysis, eliminate attributes which have high correlation coefficients with each other and are saying "much the same thing," and then to run a regression analysis. There are a variety of PC statistical applications which can be used to perform these analyses, including SPSS. If you don't have a copy of SPSS or a similar package lying around, but you do know someone who can help with statistics, the more ubiquitous Microsoft Excel can be used to perform such analyses. We use the derived approach to calculate attribute importance weights in each of our customer satisfaction surveys.