How to develop a Survey to capture Voice of Market

This is an excerpt from the book Best in Market authored by Dr. Eric Reidenbach. The book incorporates the overall essence of Six Sigma in marketing today. It takes the DMAIC structure and applies it in a simple easy to read structure that will be beneficial to a Black Belt that may not be all that familiar with marketing and/or to a marketing team that has limited exposure to Six Sigma. It is a readable book that it is not filled with Six Sigma terminology and methodology that typically takes away from the message. The points that are made on value and quality through out the book and on how to acquire them through proper techniques will provide additional insights into your own marketing methods.

The previous value models were generated from survey information collected from buyers in the targeted product/market. It might be useful to take a few moments and review the process. The purpose of this review is not to make market researchers out of manufacturers but rather to make them better buyers or specifiers of market research information. Too many research companies sell off – the – shelf research to companies. Good research is driven by the information needs of the client, not the bottom line of the seller.

Step one: Asking the right questions

The first step in any survey design is to make sure you are asking the correct and relevant questions. This is where interviews or focus groups can be particularly helpful. Both of these techniques are best classified as “exploratory” in nature rather than definitive. They can be very useful in eliciting the right types of questions to ask in a survey.

Focus groups usually are comprised of about 8 to 12 individuals who are part of the targeted product/market. These individuals are asked to talk about their experiences, likes, dislikes, problems regarding the product in question. It is essential that this discussion should focus on how they define quality and value. This is where the more comprehensive definitions of quality and value come in. Relying on the interaction among the group can be very beneficial and revealing as one respondent will play off another. Your attention should be directed toward compiling a list of attributes that define quality and value. During this process it is essential that a level of granularity be achieved, since the more granular the information the more actionable it is.

For example, a respondent might say “quality of the product” is important. Your next set of questions should be what does quality mean? Can you give me an example of good quality? Can you give me an example of bad quality? If you were designing a product like this, how would insure that it is a quality product? Allowing these questions to be part of a group discussion can be very enlightening. In fact, I have found it quite useful to have a group of manufacturing executives behind the one way glass to listen and view the groups being conducted. They will learn a lot just from this exposure.

The tendency of the uninitiated research user is to rely too heavily on focus group information. Some people tend to take the information as gospel. Keep in mind that the responses come from just 8 to 12 individuals and do not reflect a statistical sample of actual buyers. In fact, it is probably wise to conduct at least two or more focus groups to corroborate the information. These sessions can be videotaped and replayed for other individuals in the company exposing them to how the market defines quality and value. They provide good discussion guides.

Step two: Questionnaire development

The actual form of the questionnaire will depend upon what data collection technique you are using. However, most types of questionnaires have some basic elements in common.

Once a list of quality and value attributes has been generated, they can be assembled into a questionnaire. The questionnaire has three basic sections: a screening section, a body, and a demographic section. The screening section contains questions that make sure you are reaching the proper person. Ask the wrong person the right questions and you run the risk of getting bad information.

The body of the questionnaire contains the attributes that you have generated from the focus groups. These should be randomly listed (as most professional survey companies do). These questions should be answered using a scale anchored by 1 = poor performance and 10 = excellent performance. There are other types of anchors but the real issue you are trying to assess is the performance of the different competitors.

Finally, any information that you want to collect regarding who is answering the questionnaire should go last. Typical demographics include age, gender, geographic location, size of company (revenues or employees), and any other information that will help you segment or target buyers. This information goes last so that in the case of a respondent dropping out you will still have the attribute information.

There are several other aspects of questionnaire development that are important and can and should be worked out with whoever is doing the research. Good competent research companies will be able to address these issues.

Step three: Fielding

Once the questionnaire has been developed and tested, it is fielded. This again is done by the research company and depending on the nature of the buyers can be handled by telephone, internet or mail. In some cases a personal interview format can be useful. Each format has its advantages and costs and should be discussed with the research company.

Step four: Analysis

When the data is collected and edited it is subject to an analysis. There are numerous ways of surfacing information. The models shown earlier were the result of a multiple analytic approach. First the items were factor analyzed which permits the surfacing of the CTQs. These are then regressed against a composite of value questions. The regression model generates a goodness of fit measure that tells you how good the independent variables (CTQs) are in explaining the dependent variable (value) and how important each element of the model is in capturing the meaning of value.

There are other ways of analyzing the data but any technique(s) chosen should identify and prioritize the drivers of value (quality, image and price) and should do the same for the CTQs. Clearly, the more focused the CTQs the more actionable the resulting information.

The model should also yield information that permits you to identify your competitive value proposition with that of your key competitors. This will require you to not only survey your customers but also those of your competitors.

Again, many types of analytics can be used but you will want to be able to break down the respondents into your customers and be able to identify other customers by brand or competitor. This is essential.

A complete discussion of data collection and the various types of analyses are beyond the scope of this book. Other books, such as Listening to the Voice of the Market: How to Increase Market Share and Satisfy Current Customers, delve into the research process in significantly greater detail.

Generating this kind of information is not cheap. But the returns from good market information are invaluable in becoming best in the market. And, being best in the market has its own longer lasting returns.