Another Loyalty Model

Another loyalty model was developed at Harvard Business School. It is referred to as the Apostle Model and uses both customer satisfaction and loyalty to define a 2x2 matrix with customer satisfaction plotted on the horizontal axis and customer loyalty on the vertical axis. The model suggests dividing up the matrix such that the four quadrants are named Loyalists, Hostages, Mercenaries and Defectors. Each of these groups is defined as follows:
1. Loyalists - have high levels of customer satisfaction and loyalty
2. Hostages - have high levels of customer loyalty and low levels of satisfaction
3. Mercenaries - have high levels of customer satisfaction and low low levels of loyalty
4. Defectors - have low levels of customer satisfaction and loyalty.
I think each of these groups are relatively easy to understand.

The logic of the model makes sense and I certainly understand that high loyalty and high customer satisfaction would seem to identify a loyalist (one who is truly loyal).

Jonathan Barsky and Lenny Nash are principles of Market Matrix LLC and applied this model to the hospitality industry. They plotted results from 140,000 customer surveys completed in 2006 and found the model provides a robust method of understanding and grouping customers. They found Ritz-Carlton and Four Seasons to be the only two hotels to fall into the Loyalists quadrant. Days Inn and Red Roof Inns were found to fall into the Defector quadrant and the authors suggest the hotel chains in this quadrant may not have evolved as quickly as customer expectations have changed. They conclude that the model provides advance warning of customer dissatisfaction. I think this might be a leap since they were only measuring satisfaction with no apparent measures of dissatisfaction. I have found that satisfaction and dissatisfaction are not opposite of one another. Satisfiers do not necessarily become dissatisfiers when the satisfaction level drops nor do dissatisfiers necessarily become satisfiers for similar reasons. More about this another time.

These principles at Market Metrix certainly have an excellent data set. I wonder how they decide just where to draw the lines which segments the matrix into customer loyalty into low and high and the same with customer satisfaction. I remember a study that was done about 10 years ago and published in an industry trade magazine where a market research company showed plotted data points of customer satisfaction against loyalty and drew a straight line to show the relationship was linear. When I fit a regression line to the data I found an r-square for the line of less than 50%. So much for believing the relationship was linear - at least for that set of data. I believe that as customer satisfaction increases, loyalty tends to increase also. I'll stay there for now.