Customer Satisfaction Survey Design

Customer satisfaction surveys are often one of the first steps in a customer satisfaction initiative. While there is no single best design, there are a number of considerations that go into any well designed customer satisfaction survey program. Following is a brief explanation of some of these.

1. There are at least four categories of independent variable attributes to include: product, service, price, and image. And the format for understanding each of these is somewhat different.
    
2. Regardless of whether you adopt the Bradley Gale model, you must also include the value concept in your analysis. Is your product or service worth what customers must pay?
    
3. In addition, you must include loyalty measures, such as likelihood to continue doing business or willingness to recommend. The analysis should demonstrate how high the bar must be set to achieve the desired bottom line results.
    
4. A customer segmentation should be superimposed on the customer satisfaction analysis. You may want to differentiate between primary and secondary customers, key accounts and smaller accounts, decision-makers and end-users, distributors or end-users, or different need segments. Everyone is not equal.
    
5. In larger or more complex organizations, you may need to interview multiple people within a client organization. When you ask customers about customer service, are you talking with the person who makes the purchase decision or the person who actually calls customer service? Carefully consider which potential respondents will be asked which set of questions.
    
6. Competitor data must also be captured. Ideally, a market sample, rather than a customer sample, should be used. But this is not always feasible for budgetary reasons.
    
7. Whenever possible, capture information on standards of performance.
    
8. Probe for knowledge and understanding of how to use your product or service.
    
9. Analyze the data to determine key drivers of satisfaction. The most accurate approach for service attributes is some type of inferred importance measure. For product attributes, a trade-off approach is often preferred. Use factor analysis to eliminate multi-colinearity.
    
10. Assume that you will need follow-up qualitative research to make quantitative results more actionable. While customer satisfaction surveys seek to be actionable by using operational attributes on multiple levels, no quantitative survey can provide the depth of information needed for action planning. Successful implementation lies in the details.

Obviously, how the information is communicated and acted upon is as important as how it is collected. The best survey program represents only about 2% of the required effort to improve.