Not All Service Attributes Are Equal: Ranking Service Attributes by Their Correlation to Loyalty
Research without a call to action may be informative, but not very useful. One way to build a call to action element into your customer experience research is to add a measure of customer loyalty. Loyalty can serve as a basis for evaluating which elements of the service mix are most important in terms of driving customer loyalty, and as result, have more potential ROI.
Measuring customer loyalty, however, in the context of a survey is difficult. Surveys best measure attitudes and perceptions. Loyalty is a behavior. Kinesis has had success with a model for estimating customer loyalty based on two measurements:
- Promoter: This is measured with the likelihood of referral to a friend relative or colleague, using a numeric scale.
- Trust: Trust is measured by capturing agreement with the statement, “the company cares about me, not just the bottom line.” Again answered in a numeric scale.
These two measures are combined together to calculate a loyalty index, which visually is the linear distance of the plot of these two measurements from the highest possible value for each scale (cases where promoter and trust received the highest possible rating).
Mathematically, this index can be calculated with the following equation:
T = Trust rating
P = Promoter rating
ST = Number of points on the Trust scale
SP = Number of points on the Promoter scale
Note this index measures the distance from the ideal or most loyal state. Lower values estimate stronger loyalty.
Calculating a loyalty index has value, but limited utility. A loyalty index alone does not give management much direction upon which to take action. One strategy to increase the actionably of the research is to use this index as a means to identify the service attributes that drive customer loyalty. Not all service attributes are equal; some play a larger role than others in driving customer loyalty.
So…how does the research determine an attribute’s role or relationship to customer loyalty? One tool is to capture satisfaction ratings of specific service attributes and determine their correlation to the loyalty statistic. The Pearson correlation coefficient is a measure of the strength of a linear association between two variables.
The following table contains a hypothetical list of service attributes and their correlation to the loyalty index. Note lower values of the loyalty index indicate stronger loyalty, so the Pearson correlations to the attribute satisfaction ratings are negative. The closer the correlation is to -1 equates to a stronger relationship to loyalty.
|Pearson Correlation to Loyalty Index|
|Perform services as promised/right the first time||-0.62|
|Show interest in solving problems||-0.61|
|Problems resolved quickly||-0.56|
|Willingness to help/answer questions||-0.55|
|Perform services on time||-0.54|
|Employees instill confidence in customer||-0.52|
|Questioning to understand needs||-0.45|
|Appearance/cleanliness of personnel||-0.42|
|Knowledgeable employees/job knowledge||-0.41|
|Appearance/cleanliness of physical facilities||-0.37|
As this table illustrates, the service attributes with the strongest correlation to the loyalty index are: perform services as promised/right the first time (-0.62), show interest in solving problems (-0.61), and employee efficiency (-0.58). Under this hypothetical example, the hypothetical managers can conclude that of the attributes measured, these three are the strongest drivers of customer loyalty. They now can use this research to make informed judgments as to where investments in the service mix will yield the most ROI.
Correlating service attributes to loyalty is not the end of the analysis; the next step is to further put this research to action by layering in the overall performance of each attribute relative to its relationship to loyalty.