One of the primary indicators of successful business performance is Customer Satisfaction Measurement. Companies benefit from listening to their customers and taking notice of what they say about their strengths and weaknesses. It is not enough to just hear what customers are saying. Companies need to be able to understand the composition of their customer base. What is the actual percentage of customers who are delighted with the service they receive? What is the percentage of the customer base that is less than satisfied? How many are in between? To what extent is the business vulnerable to customer drift? In an effort to maintain a competitive edge and because not all customers experience superior service, it is imperative to measure, identify the gaps and correct course, when necessary. Not all customers are 100% satisfied.
The purpose of the Customer Satisfaction program is to gauge the difference between what the various groups of customers are saying. While many companies believe that their customers would tell them if there are issues or problems, unfortunately, this is not always the case. There is usually a deeper meaning to what customers are saying and it is important not to just be hearing the part that we want to hear. We need to be sure that research operates in a listening mode to uncover the totality of the message. There are always a number of customers who have important messages to report, to let you know that things don’t always go right and that there is always room for improvement. If you don’t want to hear those comments, you won’t. If you listen, you will.
The key is to be knowledgeable about the required process and procedures to use.
Science – First, there needs to be recognition of the fact that Customer Satisfaction research relies on scientific methodology. The research methodology requires reliability to predict the degree of sample representation compared to the total population measured. Statistical reliability and confidence levels are the key components of research findings when significant differences are reported between distinct populations. From a practical standpoint, the overall reliability of the data collected is fully dependent upon specific procedures to ensure validity.
Objectivity – A common problem involves removing bias and the best way to ensure this is to involve a third party. It is impossible to pretend to be or to role play the part of a third party. When the researcher is too close to the problem, there is an initial flaw that is impossible to overcome. It is important to have a view from the outside looking in. The researcher must not have a vested interest in the outcome. There is way too much at stake to allow undue (or inside, in this case) influences on the interpretation of the data reported.
Adequate Sample Size – In the final analysis, the results of the survey are only as good as the reliability of the sample selected. Adequate sample size ensures that statistical calculations are based upon a representative subset of the total population. This is crucial to the validity of the data collected. When ratings are to be generalized to total company performance, they must not be taken from a subset only representing the favorite or best customers. The entire customer database should be used and sampled across the board.
Representation – Sample strategy depends upon proper representation of the sample frame or universe. Everyone must be included for a sample to be reliable and projectable. Under ideal circumstances, a census sample would always be attempted to ensure that everyone gets to have a voice and provide feedback. However, except in the case of measuring small populations, a census sample is not always necessary. Fortunately, it is possible to utilize proven sampling methods to allow a smaller subset of the universe to fulfill this requirement. Statistical methods of sample stratification, balancing, weighting and other principles must always be considered as part of the study design.
Equal Opportunity Approach – All respondents must have an equal opportunity to participate in the survey. Everyone must be invited to the survey using the same criteria. If the survey is to be conducted via email invitation, all potential respondents must have email contact information. If the survey is to be conducted by telephone, all potential respondents must have telephone contact information. If some only have one or the other, a mixed methodology should be considered. Measurement must be performed consistent with the Voice of the Customer to ensure clarity and understanding. To understand the customer experience, it is important to speak the same language.
Questionnaire Design – The right questions to be asked, the way to ask them and the order of presentation are all key elements to the reliability of the assessment. The questionnaire instrument must have a foundation in The Voice of the Customer and reflect the language that customers use. Questionnaire design is a process that begins with an analytic plan. First you need to know what you need to measure and then you need to take extra care to make sure that the design reflects the business issues surrounding the services performed.
Always start with an outline and construct a questionnaire that is suitable for the scope of business. You need to know the type questions that need to be asked, the type of rating scale to use and the techniques used to eliminate response bias. Make sure that the questionnaire addresses all aspects of the analytic plan. Take time to make sure all instructions are clear, to socialize the questionnaire among a test population and to revise before using.
Quality Control – Quality does not happen by accident. Quality controls must be in place to ensure that all of the above principles are planned, provided for and in effect. Initial as well as periodic monitoring of the study should also be performed during the data collection process.
Benefits – Remember that the outcome of a successful program is more than just a measure of Customer satisfaction. The results can also be tied to Brand Equity and an assessment of standing within the competitive environment. It is also recommended to be sure to evaluate findings against competitive benchmark data for a comparative viewpoint.