I have been thinking a lot lately about surveys; which is a good thing, since that is my job. But not just surveys in general, but customer experience surveys, since that is our job at Maritz. And not just customer experience surveys in general, but bad customer experience surveys. And, if you look at them from the customers’ perspective, most of the industry’s current customer experience surveys are bad.
Why are they bad? Lots of reasons we all know. They are too long. They are too complicated. And they ask questions that are irrelevant to the customer. While we can debate the pros and cons of the Net Promoter Score movement, I believe a large part of the appeal of the philosophy is a reaction against these complaints. In my experience, few companies can really survive on only the “Ultimate Question”, but that doesn’t mean there isn’t a better way to write and deploy surveys.
Think about the current state of the art for survey construction. You start with a set of basic needs; for simplicity’s sake, let’s say that for a prototypical customer experience survey the needs are to know three things: 1) how the unit is doing at delivering the brand promise, 2) what they need to do to improve on their performance, and 3) how satisfied the individual customers are.
With these needs established, we write, test, refine and deploy a survey. Traditionally, a good survey is defined as one with good psychometric properties. It allows us to reliably and validly rank the channel from the best to the worst and provides key driver analysis and guidance on how to improve overall performance. And, when we do identify a customer who needs assistance, we can generate a hot alert and forward the details to the appropriate party for a response.
But, is that a good survey from the customer’s perspective? Was the length appropriate to the experience? Did we give the customer enough opportunities to tell his or her story? Was the survey relevant to the individual customer whose opinion we were soliciting? These first two issues have been raised before, but not much has been written about the third. As an example, let’s think about hotel surveys. There is a hotel just down the street from our Maumee, Ohio office, where I almost always stay when I am there. It is part of a chain and I have been there many times; it is exactly just like almost every other property in the chain. But, every time I get selected for that chain’s customer satisfaction survey, the questionnaire asks me five or six questions about the room design and layout. Have they redecorated it since the last time I was there? Is it different from the other properties in the chain that I have previously reviewed? If not, do they really think my opinions have changed? Why repeatedly burden me those questions? Cleanliness and condition of the room, yes. But design? Really? And, as a regular guest of that property, why should I get the same survey as someone who is staying there for the first time? Don’t I have other things to offer? Don’t they?
Let’s stay with my hotel analogy for a moment, although the concepts could be applied to just about any product or service. I am going to be staying again at the property in Maumee in just a few weeks. Imagine for a moment that my previous stay was less than satisfying. Say that my last survey generated a hot alert and hotel management sent me a note of apology and a coupon for a room upgrade and free drink on my next visit. I return and everything goes right this time. Then comes the survey for this visit. Will it be any different than the survey I received from the previous visit? Not under the current customer experience measurement paradigm. No wonder customers are abandoning the research process!
It doesn’t have to be this way. Databases and database management has come a long way in recent years, as has our ability to integrate and analyze data at a micro level. There is no reason why surveys in 2013 have to follow the same model as has been employed since the advent of CATI interviewing more than 30 years ago. Yes, individually tailored surveys—or as I like to think of them, intelligent surveys—will create new challenges for us. But, I believe far greater challenges await us if we don’t start doing a better job of relating to the individual customers whose opinions power our industry.