Editor's note: Chris Van Derveer is president of Van Derveer Industrial Research.

The past few years have been one of rapid advancement for marketing research as it is applied to industrial products. In the past, engineers would design products and introduce them to the market, rarely pre-testing users' attitudes about product features and per-formance. As long as sales were not disastrous, management would assume the product was OK. What could not be measured was what profits were actually lost because products were not designed properly or tested once they were in the field.

As in most segments of business, customer satisfaction is the hot topic in industrial research. Since industrial product purchasing is less motivated by emotion, the design of a questionnaire to test industrial attitudes looks relatively regimented by consumer standards. These questionnaires do not ask about how purchasers feel personally or what their lifestyles are. Unlike consumer buying this field is removed from a fully personal choice since several committees may also become involved in the purchase process.

In the ideal, if an industrial questionnaire is to rate attitudes properly it must do so with a mix of both open- and closed-ended questions. Questionnaires that include only open-ends usually indicate a writer who is not familiar with their markets and/or what responses are to be expected. Also, open-ends are more subjective and their tabulation can yield results which have a higher error rate. So where to begin?

Usually, such issues are evaluated by either a rating or a ranking question. Let us first cover ratings and then show a ranking example.

Ratings question

A typical ratings questionnaire begins with a question like the one shown below and is laid out in the following format.

Q.1 For the companies that you are familiar with, please rate them per the following features and factors.

After firms are read, ask "Are there any others?"

The rating scale would be: a "1" is excellent, a "2" is very good, a "3" is good, a "4" is somewhat poor and a "5" is poor.

The features/factors would be: product quality, price, product features, service, warranty, company reputation and advertising.

In constructing the questionnaire you should keep in mind the following when instructing interviewers:

1. After the introduction, the interviewer will read the names of the companies and circle the ones that the respondent is familiar with.

2. It is important that respondents be very familiar with the firms, otherwise they will deliver responses that are meaningless.

3. For circled firms and for one firm at a time, the interviewer should read the features/factors from left to right and elicit a response rating before proceeding to the next feature.

4. In general, don't present more than eight factors, it just becomes too tedious for the respondent.

When a respondent mentions a firm that falls in the "other" category, take down the name and rate it as you had the others. Include this category if you sell a product in a market that is expected to have new competitors.

The completion of a closed-ended question such as this provides the perfect opportunity to ask an open-ended question as a follow up. The types of questions are only as limited as your imagination. Typically, I suggest that the client ask open-ends about issues that are very, very important to them.

Below are some suggested open-ended questions. Remember, it is best to ask only one, as telephone interviewing time is limited to about 15 minutes.

1. In question #1, you rated the A and C companies as having an excellent level of quality. Why do you feel this way?

2. You said that you were familiar with companies A and B. Which of these firms is the best and which is the worst. Please describe exactly why you feel this way.

3. You mentioned that you were familiar with both the A and C companies. Could you describe the strengths and weaknesses that you attribute to both firms. How would you suggest these firms correct problems you experienced with them?

4. Which one of the four companies is most capable of introducing new technology to the market. Could you describe what that technology would be and how it would benefit to you?

Ranking question

The other option in customer satisfaction work is the ranking question. I recommend to my clients that they conduct a closed-ended ratings format, as described previously, and a rankings question. Developing customer satisfaction data from both formats is a good cross-check and can greatly increase the accuracy of the final project result.

Q.1 Could you please rank the following companies as to their product quality, pricing, service (or whatever)?

Which is the best, second best, third best or worst based upon their delivery of

(fill in from above) to your firm?

TO INTERVIEWER: Place a "1" next to best, a "2" next to second best a "3" next to third best and a "4" next to worst.

Note: You may again follow up a ranking question with an open-ended format. On our questionnaires, I try not to have open-ends after both the ratings and rankings sections. Usually one well structured, pertinent open-end is enough.

Conclusion

Well, there is the essence of a simple customer satisfaction questionnaire. Based upon the product and the numbers of surveys conducted per segment, other statistical techniques such as regression, conjoint analysis and the like can be employed. Please be advised that statistics should be used only in a focused manner and can greatly increase project cost.

As in all projects, always pre-test the questionnaire with a limited number of respondents to assure that it's sequenced properly and that the questions are well understood by the respondents. All participants should be thoroughly pre-screened to make certain that they can respond knowledgeably to the survey topics.

Don't make the ratings/rankings section too cumbersome. You will have up to 15 minutes to conduct your telephone interview. A good rule of thumb is that this section should consume about half of your survey time. In the other half you can ask them about how they use the product, what they will buy in the future, what they think of your advertising, or whatever else you need to know at that point.

Best of luck with your next customer satisfaction survey.