Improving customers’ satisfaction means understanding which attributes drive satisfaction. Improving customers’ satisfaction in a cost effective way also requires setting priorities and finding the best “leverages”.
Researchers know well how to find the importance of attributes by using traditional statistical techniques such as correlations, multiple regressions models, etc. Unfortunately these methods give limited actionable outputs because they all assume satisfaction is linear. This is rarely the case. Some attributes have a big impact as long as they haven’t reached a minimum level and almost no impact afterwards. In fact, it is critical to distinguish the impact of an attribute on overall satisfaction from its impact on overall dissatisfaction. As an example, if one uses a regression analysis to evaluate the impact of the many satisfaction attributes which new car owners rate during the first months they use their car, the attribute “my car starts” will be placed at the bottom of the list! For the majority of customers whose car starts normally, enhancing this attribute won’t increase overall satisfaction. Does it mean that this attribute is not “important” for them? Yes, as long as the car starts OK. However, for those owners whose car doesn’t start, making it start will become an absolute priority.
This car example is somewhat simplistic but demonstrates that the impact of an attribute on overall customer satisfaction can be asymmetric. As long as the car doesn’t start, users are totally dissatisfied with it and nothing else can compensate. As soon as it starts, other items become much more important to them than simply making the car start even better.
Previous research has demonstrated that attributes can impact satisfaction in four different ways. These are usually classified as follows:
Maslow (1954), in his well-known Hierarchy of Needs, demonstrates that physiological needs must be fulfilled before a human being starts to care about safety and security, which in turn must be fulfilled before working on social connection, … This is also the case with the four categories of attributes mentioned above.
Basic attributes must be at an “acceptable” level before Performance attributes start to matter. And it’s only when both the basic and performance attributes are optimized that bonus attributes show their full potential.
David Perroud exposed the results obtained with this method on a large airline passenger satisfaction study at the ESOMAR 06 leisure conference. Both the conference program committee and the participants elected this contribution best paper of the conference for “an innovative approach that gives a concrete contribution to the decision making process”. The paper ran for the best presentationat the ESOMAR event in 2006/2007.For more information on the subject download the paper presented at ESOMAR (including a case study on airline satisfaction)