Why Is Customer Satisfaction Research So Hot?

I'm really interested in getting readers perspectives on why customer satisfaction research is so hot?

One thing that has constantly amazed me since I became an Analyst at Forrester Research, is the overwhelming interest in all things concerning customer satisfaction research. Easily a third of my inquiries are about how to design such studies, how to improve what they have, what are the issues with multinational studies, and how to deal with new concepts such as NetPromoter.

Even in this dire market, it seems that customer satisfaction studies are one of growth area in market research (according to Inside Research).

This has led me to write quite a bit about customer satisfaction ("The Next Wave In Customer Satisfaction is CRM Integration, http://www.forrester.com/go?docid=47246, "Enhance Customer Satisfaction's Impact" http://www.forrester.com/go?docid=44166, and "Why Customer Satisfaction Studies Fail," http://www.forrester.com/go?docid=45043). But for those who are short on time, I'll net out a few key pointers:

  • Never assume you know all of the right questions to ask. Conduct some qualitative research upfront to determine not how well you are doing, but what metrics your customer base would like to use assess you.
  • Use a 10 point scale, never a five point scale.It is very common for companies to use a five point scale for satisfaction studies. When they get the results, they often look at the percentage who give them a 4 or 5, see that the percentage is quite high, and call it a day. The problem is, a 4-5 on a 5 point scale is all those who are somewhat satisfied and above (3 being neutral). Also, some people will never give a 5, leaving you to sort out which 4's are those who are really only somewhat satisfied, and which are those who won't rate anything as "perfect." It is far better to use a 10 point scale and look at the percentages that rate 9-10.
  • Always conduct a key driver's analysis. Not all the issues that you measure are likely to be equally important. If you conduct a key driver's analysis using a tool such as regression analysis, you can understand the impact of various attributes. It also will allow you to reduce the number of questions for future waves once you know what is important and what is not.
  • Consider moving from measuring to fixing. One major trend in the marketplace seems to be CRM integration of customer satisfaction metrics. This represents a move from measuring and assessing to uncovering problems and tasking resources with fixing them as quickly as possible.  

There is lots more. In fact, there is so much that I'm running a workshop on Customer Satisfaction at the end of this month! Its called "Driving Business Success with Customer Satisfaction," http://www.forrester.com/go?eventid=2183


re: Why Is Customer Satisfaction Research So Hot?

I firmly agree with many points in your post. It's so common for companies to rush into a survey without taking the time to understand the key drivers of customer satisfaction that they need to ask about. I also couldn't agree more about the need for regression or other statistical analysis to determine the importance of a category. Usually management teams will focus on the lowest scoring category, even if that category is relatively unimportant. The idea of improving a critical category that's just good and not great is foreign to a lot of people.

I think a 5 point scale can work in some cases, especially if you have a positive bias (1 negative, 1 neutral, and 3 positive choices). Seven point scales can also work in my opinion. The only problem I have with 10 point scales is that some people will assume 5 is "neutral", while others assume 5 is "failing", so you don't have consistency in your results.

I wrote an article that goes into more detail in this area if your readers are interested in some additional information. I also added a link to this article on my blog.



Jeremy Farkas