Don't Expect What you Don't Inspect

One of my least favorite bosses used the phrase "don't expect what you don't inspect." It did not make me happy when he said it, but he made his point that it is important to measure those attributes which are important. Michael LeBoeuf noted as his platinum principle that people do what gets measured. It all points in the same direction. If we want to have loyal customers, we better have systems in place to let us know how well we are doing.

Even the best designed loyalty-based program will deteriorate unless an effective measurement system is established. Accurate, relevant measures that establish feedback loops are the foundation of organizational learning.

One way to avoid mistakes in managing a customer loyalty program is to track and understand the cash-flow consequences of changing customer loyalty - the defection of a target customer. John Goodman, of the Technical Assistance Research Program (TARP) has described three building blocks necessary for an effective customer-driven quality-improvement tracking system:
1. quantify the bottom-line impact of poor quality and customer problems
2. identify quality priorities on the basis of market/revenue impact, and
3. measure performance continuously in each priority area and have top management review the results.

A company's definition of quality must be driven by the desire to earn customer loyalty, and it requires an examination of products and srvices in terms of customer expectations. Achieving customer satisfaction, continued customer loyalty and long-term revenue growth depends on the consistent delivery of products and services throughout the customer's relatioinship with the company, and requires that all areas of the company have a direct or indirect effect on quality, customer satisfaction and continued customer loyalty.

Goodman has identified two rules for a successful measurement system:
1. Identify where and how often customr expectation have not been met:focus on expectations. He has suggested "presenting customers with a list of potential problems or questions across a broad range of transaction types and soliciting which and how many problems or questions they have experienced." Remember, whatever the customer says is a problem, is a problem!
2. Determine the company's ability to handle customer complaints and inquiries by obtaining a baseline indication of the extent to which the existing recovery process reduces defections or helps retain customers who otherwise might defect to the competition.

By examining the impact of problems on customer loyalty, the company can quantify the payoff for problem prevention. For example, in most industries, for every six customers who perceive thay have an unpleasant experience, the company risks losing some, if not all of the future revenue from at least one of those six customers. Further, about 50 percent of all customers with problems or questions never complain to anyone; about 45 percent complain to someone at a front-line level; and barely 5 percent contact the corporate office.

In general, customers who do not complain are about 15 percent less loyal to the company that customers who do complain. Loyalty for customers who do complain and can be satisfied through effective measures increase by about 20 percent. Finally, customers who are completely satisfied often exhibit no less and sometimes greater loyalty than customers who have not experienced a problem. An additional measure of payback and a useful predictor of customer loyalty for problem resolution is word-of-mouth behavior.

Goodman has identified five types of data required for calculating the market impact of improved qualtiy and service:
1. profit value of a customer,
2. number of customers who experience problems,
3. percentage of custoemrs who complain about problems,
4. impact of problems on loyalty, and
5. the impact of the service system on loyalty.

The objectives are (i) estimating the bottom-line impact of quality and service,
(ii)determining potential payback from preventing problems and/or improving customer response systems,and (iii)calculating the return on investment of a customer-driven quality program. Armed with data, quality improvement priorities can be established on the basis of emprirical data collected from custoemrs, NOT management intuition.

Companies can isolate the relative importance of different performance areas by determining the relationship between specific problems and customer loyalty. Once the company determines the percentage of its customer base at risk, priorites can be set accordingly. It should be noted that internal data about custoemr complaints often can conflict becasue custeomrs report eh problem to different people at different times. Multiple reports can lead to conflicting or inconsistent indications about the nature and frequency of the problems. In addition, data may need to be weighted based on where within the company the complaint was received. Companies can easily fall into a measuremnt trap with too much data, too little sensitivity about data interpretation and too complex data integration.

Often, management may find themselves in a situation where analysis assumes priority over action. It is important to use the key drivers of customer satisrfaction to organize the use of internal data. Internal data shold be used to corroborate the nature, frequency and seriousness of customer problems in the key areas, then inform management about the specific nature and solutions for the problems.

A final point about data. Failure to conduct routine follow-up measurements will jeopardize the execution of quality improvement processes.

When reports quantify bottom-line impact, set priorities and dianose potential solutions, an increased investment in measurement activities can be justified. A tracking system, designed to determine the effectiveness of its customer loyalty management system, requiress five criteria to ensure continued effectiveness according to Goodman.

1. transaction-based questionnaires,
2. cost-effectiveness with ease of implementation,
3. minimized response burden for customers,
4. manage by exception, and
5. useful data.

The bottom line is that most companies that have loyalty programs do not manage them and then they wonder what happened. Duh!
5