Comments or suggestions?

about this or future e-newsletters.

Increasing customer loyalty via advocate profiling

By Davis Klaila, Ceridian senior director of Strategy

Ceridian has a very high customer retention rate of over 90 percent in a business-to-business (B2B) industry that has an average retention rate of 74 percent. By learning what causes customer losses, we can capture information that helps us to respond to our customers' needs.

Ceridian has embarked on an enterprise-wide program to capture segment-level customer learning across the company to teach our customers about our industry and business and to amplify the voice and influence of our customer advocates. This will help us to provide accurate information in response to your business needs, therefore saving you time.

What is profiling?
Profiling is often associated with the work of law enforcement, intelligence agencies and consumer product companies. These organizations use profiling to help predict the behavior of individuals and groups in certain settings. Profile development is both qualitative and quantitative. Qualitative information may include observations of the occurrence of certain behaviors over time. Quantitative information is the measurable outcomes of that behavior. Different organizations may have vastly different subjects and constituents, however, they share similar research procedures and protocols.

Example: Every Thursday, Joe begins reading local restaurant reviews. At lunch on Thursday and Friday, he talks about food and solicits restaurant opinions. On Monday, Joe talks about the restaurant he went to on Friday night.

This behavioral information is qualitative. It cannot be added, subtracted or in any way treated as continuous data, but you are able to count the occurrences of the behavior and predict with some certainty when they may occur.

Quantitative information would be the amount Joe spent on restaurant dining. This information is suitable for many types of statistical analysis. One is easily able to imagine a consumer research firm putting this information to good use.

Conventional wisdom suggests that in B2B services, company profiling is ineffective -- as there are often multiple influencers and it is difficult to identify the key decision makers. But research conducted by Ceridian and others suggests that modeling and predicting group behavior is very effective when one focuses on meaningful attributes.

Understanding profiles
Profiling's roots are in conceptual dependency theory, scripts and event schemata. Simply put, these theories suggest that one walks through the world with a set of scripts -- as in a play. Our scripts help us interpret the actions of others and prepare us by managing our expectations. There are a great many shared scripts and certain scripts may even be "inherited." This suggests that if you are able to determine and appreciate the scripts one is "reading" from, you may know what is motivating them, what cues to provide to help them make a decision and possibly predict their behavior.

Creating profiles
The first step in creating a profile is to see though the eyes of your customers. Direct observation is important to avoid using your scripts as a means to study the behavior of others. Direct observation may be accomplished by observing individuals or analyzing data. The best results are obtained with a combination of the two.

The process begins by identifying what you want to profile. Here are a few examples:

• Researchers looking at the behavior of jurors seek to predict how the foreman is selected.
• Police in Southern California develop behavior models to determine who will emerge as a leader in street gangs.
• Market researchers create models of buying behavior in stores by strategically placing end caps and other product displays.

Understanding why customers leave
A recent study by the Corporate Executive Board and Ceridian looked at models of customer loyalty. One facet of that model is customer retention.

Asking customers why they are leaving is an interesting process -- especially because the customers who leave are generally satisfied. It has been proven in study after study that satisfaction does not increase the likely hold of retention.

Any conversation with a lost customer is a challenge as you try to understand why they left through the lens of any product or service issues -- while your customer is saying something very different. What is happening is you have an idea or "script" in your mind for why the customer may be leaving, and the customer has an entirely different set of reasons.

Before you ask, seek to understand. An effective way to gain insight into a customer's behavior is to determine the customer's value proposition for your products and services.

At Ceridian, our customer's value proposition is time and information. What our customers want from us, or any similar provider, is a set of products and services that save time and provide accurate information in response to their business issues. Ceridian arrived at this conclusion after analyzing several years' worth of data and interviews from our customers and then directly testing our assumptions with our customers.

A second tactic is to use an outside resource to interview lost customers. An outside resource has two advantages:

• The customer who still likes you may be more comfortable and perhaps more honest with a third party.
• The third party does not share your bias and is able to capture the basis of the ideas and factors that the customer may use. It is very important to hear the customer. Without the customer's true perspective, any intervention you put in place will miss the mark.

Deciphering the data
What did Ceridian find when we talked to our customers? We were able to determine 75 percent of the reasons for customer loss. This is not to say that other customers who left had no reason for doing so -- that reason may have been difficult to articulate or so unique it did not register.

We found that approximately 50 percent of those customers who left Ceridian did so for internal reasons not related to Ceridian. These reasons may include mergers, change in management's view of outsourcing, etc. These types of departures might be "managed" and perhaps mitigated for some risk, however, significantly influencing this type of decision requires a comprehensive relationship with the customer.

Customers also leave because they perceive they have "outgrown" the offering. This perception is often based on a set of beliefs that you may directly influence. For example, the customer may think that your offer lacks a certain set of features and functions. This belief may be a result of misinformation from a competitor or from a lack of appropriate training.

Success in a B2B service business is dependent, in part, on your customers knowing your offering. Considering the level of turnover in most firms, your current customer contact may never have been trained on your product. We found that approximately 20 percent of Ceridian's customers who left did so based on the perception that our services would not meet their current or known future needs. However, in most cases, those perceptions were incorrect.

Finally, customers leave because they have a better offer from a competitor (15 percent) or they are dissatisfied (10 percent).

Our model suggests customers leave for a set of reasons that are observable and if given the right information both predictable and open to intervention. We tested this assumption on our customers and those at other firms. We found that an estimated 25 percent of lost customers may have stayed had they been presented with a viable solution/alternative. In short, through the implementation of targeted interventions at each of the retention points (internal change, perception of features and functions, lower cost offering and dissatisfaction) one may expect to increase retention by influencing the customer's script.

Retention is key
When we tested the high value of our customer retention with other vastly different businesses, we found similar drivers for retention. The results suggest that models for retention may be developed and replicated because they appear to share some common attributes. The use of such models in understanding customer opinions for retention will result in a better customer experience!