The measurement of customer loyalty has been a hot topic lately. With the latest critiques of the Net Promoter Score coming in from the both practitioners and academic researchers, there is much debate on how companies should measure customer loyalty. I wanted to formally write my thoughts on this topic to get feedback from this community of users. Much of what I will present here will be be included in the third edition of my book, Measuring Customer Satisfaction. I welcome your thoughts and critiques. Due to the length of the present discussion, I have broken down the entire discussion into several parts. I will post each of them weekly. Below is Part 5 of the discussion. If you missed them, read Part 1, Part 2, Part 3 and Part 4.
Customer Loyalty 2.0 represents this advancement in the measurement and meaning of customer loyalty. The purpose of the following analyses is to provide additional validity evidence regarding the measures of loyalty, Advocacy Loyalty Index (ALI) and Purchasing Loyalty Index (PLI). A common approach to establishing the validity is to show how the ALI and PLI are related to attributes of service and product quality. Below, I employ a common method, loyalty driver analysis, used by companies to identify the key drivers of customer loyalty. This method includes analysis that shows the relationship between business attributes and customer loyalty. In the process of establishing validity evidence, I will provide real-life illustrations regarding the merits of conceptualizing customer loyalty in this multidimensional framework that can help companies increase growth through new and existing customers.
Loyalty Driver Analysis
Loyalty driver analysis enables companies to identify the business attributes that are important to ensuring customer loyalty. Companies can allocate resources to important business areas that have the greatest impact on increasing customer loyalty. While each response from a survey can be (and should be) examined to deal with specific causes of customer loyalty/disloyalty at an individual, customer level, driver analysis can be thought of as a macro look at the customer base (or customer segment). By analyzing data from a large segment of customers, driver analysis helps companies to identify the common causes of customer loyalty across these customers. Consequently, based on the results of the driver analysis, companies can make organization-wide improvements to their business processes that will have an impact on the customer segment of interest.
Two pieces of information are examined in driver analysis: 1) derived importance: the degree of impact of each business attribute on customer loyalty and 2) performance: the level of performance of each business attribute.
The degree of impact each business attribute has on customer loyalty is determined. This degree of impact is indexed by a correlation coefficient (sometimes, referred to as “derived importance”) between ratings of a business attribute and a customer loyalty index (either ALI or PLI). Each business attribute has a corresponding “derived importance” that indicates the impact that the business attribute has on customer loyalty. The higher the correlation, the greater the impact that business attribute has on customer loyalty. The degree of impact (e.g., correlation coefficient) can vary from 0.0 (no impact) to 1.0 (perfect impact).
Using the Wireless Service Provider and PC Manufacturer studies, we can identify the derived importance of each of the business attributes. Given we have multiple measures of customer loyalty, each of the business attributes will have multiple derived importance, one for advocacy loyalty, one for purchasing loyalty and one for retention loyalty (Wireless study only). In both studies, respondents were asked to indicate the degree to which they agree or disagree with statements regarding their customer experience on a scale from 1 (strongly disagree) to 5 (strongly agree). Statements reflected business attributes that ranged from product quality (e.g., reliable service, PC reliability) to service quality (e.g., customer service reps, technical support reps). These ratings were correlated with each of the measures of customer loyalty, ALI, PLI and RLI.
Next, company’s performance on each business attribute (customer ratings) is calculated. Performance of a given business attribute is simply the average rating of agreement. Higher scores reflected better performance (better customer experience). Possible performance scores could range from 1 (worst customer experience) to 5 (best customer experience). Business attributes that have low performance ratings have ample room for improvement. Business attributes that have high performance ratings have little room for improvement.
Wireless Service Provider Driver Analysis
I calculated the derived importance and performance for one of the Wireless Service Providers in the study. Table 3 contains the results for this particular Wireless Service Provider. The column labeled "Mean" reflects the performance for each of the business attributes; the columns labeled "Derived Importance on ALI," "Derived Importance on PLI," and "Derived Importance on RLI" reflect the importance for each of the business attributes on advocacy loyalty, purchasing loyalty and retention loyalty, respectively.
Table 3. Descriptive Statistics and Derived Importance of each Business Attribute for a Wireless Service Provider
In driver analysis, both the performance and derived importance are examined simultaneously to understand where improvements would have the greatest chance to improve customer loyalty. If business attributes that have a large impact on customer loyalty (high derived importance) and have low performance ratings, companies might consider allocating resources to these business attributes in order to improve customer loyalty. If ratings of business attributes are high, however, companies can promote these business attributes as strengths and best practices. Using both the derived importance of each business attribute and the performance (e.g., rating) of each business attribute, we can create a Loyalty Matrix (see figures below) that allows us to visually examine all business attributes at one time.
The abscissa (x-axis) of the Loyalty Matrix is the performance rating (agreement, performance, satisfaction) of the business attributes. The ordinate (y-axis) of the Loyalty Matrix is the impact (derived importance) of the business attribute on customer loyalty. The Loyalty Matrix is divided into quadrants using the average score for each of the axes. Key drivers appear in the upper left quadrant and are often referred to as Key Drivers. Key drivers reflect business attributes that have both a large impact on customer loyalty and have low performance ratings relative to the other business attributes (these key drivers appear in red). Because we have three customer loyalty indices, we can calculate three separate Loyalty Matrices, each for the specific customer loyalty index. Below are the Loyalty Matrices for a particular Wireless Service Provider.
Figure 4. Advocacy Loyalty Driver Analysis – Wireless Service
Figure 5. Purchasing Loyalty Driver Analysis – Wireless Service
Figure 6. Retention Loyalty Driver Analysis – Wireless Service
The results each of the driver analyses seem fairly consistent. We see that Customer Service Representatives (CSRs) have a relatively large impact on customer loyalty compared to the primary product offering attributes (Reliable service, Good coverage). Whether customers will be loyal to this Wireless Service Provider depend to a greater degree on the competency of the CSRs than on the product offering attributes.
Using the three Loyalty Matrices, we can draw some conclusions regarding how this particular Wireless Service Provider can increase advocacy, purchasing and retention loyalty. When employing the use of driver analysis, we typically focus on the the Key Drivers (those attributes in the upper left hand quadrant) as areas to focus if we want to improve customer loyalty. We do so because these are the attributes that have a large impact on customer loyalty and have much room for improvement. To improve loyalty, no matter how it is measured, the results of the driver analysis indicate that the company should focus on improving CSR attributes as these attributes have a relatively large impact on advocacy, purchasing and retention loyalty and have much room for improvement.
PC Manufacturer Driver Analysis
I calculated the derived importance and performance for one of the PC Manufacturers in the study. Table 4 contains the results for this PC Manufacturer. The column labeled "Mean" reflects the performance for each of the business attributes; the columns labeled "Derived Importance on ALI" and "Derived Importance on PLI" reflect the importance for each of the business attributes on advocacy loyalty and purchasing loyalty, respectively.
Table 4. Descriptive Statistics and Derived Importance of each Business Attribute for a PC Manufacturer
Below are the Loyalty Matrices for a particular PC manufacturer.
Figure 7. Advocacy Loyalty Driver Analysis – PC Manufacturer
Figure 8. Purchasing Loyalty Driver Analysis – PC Manufacturer
We see that there are differences in what drives advocacy loyalty and purchasing loyalty. With regard to advocacy loyalty, we see that both of the PC attributes (PC reliability, and PC features) have a big impact on advocacy loyalty, more so than the technical support attributes. Whether customers will be advocates of this PC manufacturer depend highly on the computer itself and less so on the quality of technical support. With regard to purchasing loyalty, however, we see that many of the technical support attributes (excellence, timeliness, understands needs, availability) have a relatively big impact on purchasing loyalty. Interestingly, PC attributes do not have a big impact on purchasing loyalty.
Using the two Loyalty Matrices, we can draw some conclusions regarding how this particular PC Manufacturer can increase advocacy loyalty and purchasing loyalty. To improve advocacy loyalty, the driver analysis seems inconclusive. While PC features are big determinants of advocacy, they are rated as relatively good. Consequently, there is not much room for improvement in these attributes. The technical support attributes, while rated as relatively low, do not have a large impact on advocacy loyalty. If this PC Manufacturer wants to improve purchasing loyalty, however, the results of the driver analysis indicate that they should focus on improving technical support attributes as these attributes have a relatively large impact on purchasing loyalty and have much room for improvement.
Missed Opportunities to Improve Customer Loyalty
When practitioners talk about "customer loyalty," they are usually referring to advocacy loyalty; many loyalty programs are, in fact, based solely on advocacy-related content. For example, the Net Promoter Score is based on the "likelihood to recommend" question. Additionally, the American Customer Satisfaction Index (ACSI) is based on the "satisfaction" question. If the PC manufacturer relied solely on this question, the results of the driver analysis suggests that they should focus ensuring that their PCs are reliable and has features their customers want. This driver analysis found that many of the technical support items were not important in improving advocacy loyalty (they appeared in the lower left quadrant). The use of advocacy-related loyalty questions as a way of measuring and defining customer loyalty limits improvements in acquiring new customers.
The driver analysis using the PLI painted an entirely different picture. We saw that many technical support items were now key drivers of customer loyalty. Expanding the definition of customer loyalty to include increased purchasing intentions clearly shows that loyalty can be improved beyond mere referrals of new customers. Companies can now identify business attributes that, when improved, would increase purchasing loyalty of existing customers. Improving technical support for the PC Manufacturer would increase revenue from existing customers through increasing their purchasing behavior (buy different products and increasing their purchasing frequency). Looking at both advocacy loyalty and purchasing loyalty, this PC manufacturer can maximize revenue through both new and existing customers.
The evidence from two separate studies shows that the Advocacy Loyalty Index (ALI) and the Purchasing Loyalty Index (PLI) measure two different types of loyalty. Even though the two types of loyalty are correlated (advocates tend to be purchasers), the relationship between the ALI and PLI is not perfect, suggesting that these loyalty indices measure unique constructs. We have good evidence that the loyalty indices are each measuring some unique aspect of customer loyalty.
The results of the present analysis show that, as expected, the measures of customer loyalty are logically related to the customer experience. Customers who have a better customer experience tend to have higher levels of customer loyalty. Furthermore, the impact that the business attributes have on customer loyalty depends on the customer loyalty index that is used.
For more information about the Advocacy Loyalty Index and the Purchasing Loyalty Index and more detailed information about the driver analyses in the studies reported here, you can download a free copy of executive reports on the two studies (Wireless Service Providers and PC Manufacturers) at Business Over Broadway.