You’ve become the new manager of a hotel that is struggling. You want to find out what guests thought of their stay in order to make the hotel better. You come up with the following question…
How satisfied were you with your stay at our hotel?
It’s short…it’s sweet…and it’s almost entirely useless!
Okay, let’s start with what the positives about this question before we tear it apart. This question will give you an overall measure of how much guests liked your hotel. It can test whether it really is true that guests don’t like the hotel.
1) What do you do now? A satisfaction survey is likely being done because you want to take some sort of action to fix a problem if it exists. Just knowing someone is unsatisfied, however, won’t give you enough information to make an adjustment to improve your hotel. Was it that the room service food was slow? Or that the pool drains were clogged? Or that the check-in clerk was unfriendly? The possibilities are endless.
2) What were they thinking? Given the fact that satisfaction with a hotel encompasses a wide range of concepts and services, there is no way of knowing which of these your respondents were thinking about when they answered the question. Some respondents may think, “well, my room was comfortable, so I was very satisfied with the hotel”, while others may evaluate only the service personnel and the food at the breakfast buffet. You have no way of knowing whether people who answered “very satisfied” were all thinking of the same thing when answering the question.
How to fix it.
Don’t ask questions about general concepts or ideas; ask about specific concepts or ideas (i.e. being “a good person” is general, being “polite to waiters” is specific). A question that focuses on a specific concept is known as a “construct-specific” question (“construct” is really just a fancy word for concept).
So, for our example above, don’t make people guess what you mean by satisfaction! Tell them. Ask if the room service was prompt, if the swimming pool was clean, if the check-in clerk was friendly, if their bed was comfortable, if the breakfast buffet was tasty, and so on… You get the point.
Asking multiple specific questions instead of one general question will not only make your questions easier to answer for your respondents, but it will also make your data easier to analyze and act on.
How to be sure it’s fixed.
All you have to do is think about whether you can break it down into a more specific word or phrase or not.
For example, let’s say you’re a coffee shop, and you want to know if you should change coffee suppliers.
Your first thought is to ask your customers, “How satisfied are you with your coffee?” Not a construct-specific question.
So, let’s break that down into construct-specific questions… (The questions in the blue boxes are your final set of questions.)
Okay, but how do I create these out of thin air?
Creating a diagram like the one we did above will really help. Start with a general idea and then break it down more and more into its specific parts. So let’s take our hotel example again…
Then, all that’s left is to turn each box into a methodologically sound question with methodologically sound response options to go with it. Need help with that part? Check out some of my other blog posts.
Want us to do the heavy lifting for you? Download one of our methodologist verified templates straight from QuestionBank into your survey—for free! Any questions? Let us know in the comments section below…