The Customer Service Survey
The Satisfaction Question
Tue - September 26, 2006 03:29 PM in
How satisfied were you with your overall experience?
- Very Satisfied
- Somewhat Satisfied
- Neither Satisfied nor Dissatisfied
- Somewhat Dissatisfied
- Very Dissatisfied
This satisfaction question is a staple of consumer surveys, and even we at VocaLabs use it all the time (mostly because everyone expects it to be on a survey). Most people don't realize just what a bad question it is, though.
Let me count the ways.
So why do we continue to ask the Satisfaction Question?
Posted by Peter Leppik
Let me count the ways.
- By using the word "Satisfied" in the question but not the word "Dissatisfied," the question implicitly solicits a "Satisfied" response. A better wording would be:
How satisfied or dissatisfied were you with your overall experience?
- If this question is being read aloud (for example, in a phone survey), then there are too many choices for the participant to remember. For a verbal survey, it is best to have as few options as possible for every question, with two being the ideal number (where possible). If this was a phone survey, I would suggest:
Were you Satisfied or Dissatisfied with your overall experience?- Satisfied
-
Dissatisfied
- The word "Satisfied" is ambiguous and means different things to different people. It is better to ask about specific events or actions rather than the participant's emotional state. One variation which is becoming popular is:
Would you recommend this company to your friends?
- Companies have a bad habit of writing variations on this question, such as How satisfied you were you with the agent's friendliness? and How satisfied were you with the time it took to reach an agent? Most survey participants won't appreciate the distinctions between these different "satisfaction" questions, and will answer them all the same way. It is better to ask about the issue directly:
Was the agent friendly? and Were you able to reach an agent in a reasonable amount of time?
- Because this question tends to get more positive than negative responses, many companies only look at the percentage of responses in the "Very Satisfied" category. But if that's all you're going to look at, why waste the participant's time with the other four choices? Might as well just ask:
Were you very satisfied with your overall experience?
- Most companies don't really know what they want to measure with this Satisfaction question, other than some nebulous quality called "satisfaction" which means something different to everyone. It's much better to figure out exactly what you want to measure, and craft survey questions to measure those parameters directly. For example, if "satisfied" means "the customer's problem was solved," then you should ask:
Was your problem solved on this call?
If "satisfied" means that the call met the customer's expectations, then you should ask:
Did this call meet your expectations?
So why do we continue to ask the Satisfaction Question?
Posted by Peter Leppik
Posted at 03:29 PM | | | | |