The Customer Service Survey
Vocalabs U, Part Two
Thu - November 16, 2006 12:19 PM in
I'd like to offer some added encouragement in response to Peter's blog entry about conducting a seminar on survey techniques and methods. Some years ago when I began work at VocaLabs, I had the vague and commonly shared suspicions about the validity of surveys. But I couldn't define my suspicion any more deeply than simple distrust.
Today, I can say that it is reasonable to be suspicious of certain survey methods and can list dozens of bias and other barriers to truly honest study results.
There isn't much point in educating on bias to those trying to create a survey with a pre-determined outcome. (Would you vote for candidate X who loves puppies or candidate Y who doesn't drink nearly as much as she used to?") But it is important to understand that there are real yet subtle issues that influence the outcome of a survey even though the study designer was trying to be objective.
An example of unintended bias that we use often is if you ask "Do you like milk?" the answers will differ if the previous question was "Do you like cookies?"
If you ask: "How satisfied were you?" results will look better than if you asked: "How satisfied or dissatisfied were you?"
Each of us would like to see improvement in the quality of the customer care we receive. One of the barriers is that companies believe they are doing better at service quality than an objective review might reveal. And bad survey technique with unintended bias that makes a company's service look ok is one significant reason poor service is so common.
If Vocalabs U causes better surveys to be done and raises the reliability of customer feedback data; and as a result of knowing the truth, companies do improve; then we all benefit.
Tell us what you think.
Posted by Rick Rappe'
There isn't much point in educating on bias to those trying to create a survey with a pre-determined outcome. (Would you vote for candidate X who loves puppies or candidate Y who doesn't drink nearly as much as she used to?") But it is important to understand that there are real yet subtle issues that influence the outcome of a survey even though the study designer was trying to be objective.
An example of unintended bias that we use often is if you ask "Do you like milk?" the answers will differ if the previous question was "Do you like cookies?"
If you ask: "How satisfied were you?" results will look better than if you asked: "How satisfied or dissatisfied were you?"
Each of us would like to see improvement in the quality of the customer care we receive. One of the barriers is that companies believe they are doing better at service quality than an objective review might reveal. And bad survey technique with unintended bias that makes a company's service look ok is one significant reason poor service is so common.
If Vocalabs U causes better surveys to be done and raises the reliability of customer feedback data; and as a result of knowing the truth, companies do improve; then we all benefit.
Tell us what you think.
Posted by Rick Rappe'
Posted at 12:19 PM | | | | |