We have noticed a rather disturbing trend in some of the customer satisfaction questionnaires found in restaurants and other retail establishments. Some seem to be missing the most important question of all -- "satisfaction with the service experience." This probably is an error of omission. At least we hope so. The alternative is even worse.

There are some people who feel that a satisfaction score should be "built" or "estimated" from the other questions in the questionnaire. While it certainly is true that a mathematical relationship can be established between overall satisfaction and other questions in a customer satisfaction questionnaire, that is no excuse to not ask how satisfied a customer was. Those who choose to estimate satisfaction run the very real danger of ending up with an estimated score that misses a key variable, since the score can only be comprised of the finite set of questions included in the survey. Using the overall satisfaction question as the main measurement for tracking bypasses this danger, since a respondent's answer to this question is a function of all "customer satisfaction attributes", stated or otherwise.

Long story short, if you want to know how satisfied customers are, ask them.