Keep on Measuring Customer Satisfaction

n his column, "The Last Word," Paul Light argued that the federal government's recent customer satisfaction survey was unnecessary ("Which Customer Is Right?" February). We think some of his criticisms of the survey are valid but that his conclusion is wrong. In this distinction lies a lesson for the National Partnership for Reinventing Government on how to improve the next round of surveys.

Several years ago, a state government executive was diligently promoting customer service in his organization, which included the centralized motor pool of state vehicles. The executive visited fleet headquarters to review customer service initiatives. The fleet director told the story of a state car that was returned to the motor pool strewn with refuse from a visit to McDonald's. "What did you do?" inquired the executive, expecting-or at least hoping-to hear that the car was thoroughly cleaned and readied for another customer. "We gathered up all the junk from McDonald's," proudly replied the director, "boxed it, and sent it to the guy who made the mess, informing him that the vehicle is state property and should be treated with more care."

Who is the motor pool there to serve? Before one can pursue a strategy of customer service, one must decide who is the customer.

In choosing a customer to be surveyed, some federal agencies missed the chance to measure their largest group of constituents or the ones most central to their mission. Our friend Paul Light complains that some agencies picked easy-to-please customers, noting for instance, that the Education Department picked primary users of department publications, not teachers or parents, and the Environmental Protection Agency picked reference librarians. On the other hand, as he points out, the Federal Aviation Administration and the Occupational Safety and Health Administration picked harder-to-please customers-the principal targets of their regulatory activities. Others chose to survey the principal beneficiaries of their service. The Veterans Benefits Administration, for example, surveyed a random sample of veterans, knowing that in many cases the VBA has to say no to veterans' requests. It's no surprise that Education and EPA received much higher customer ratings than FAA, OSHA and VBA.

Yet scrapping the customer survey is not the answer. On the contrary, we think it should be enhanced and vigorously pursued. Three lessons can improve the survey process. All three involve making a more careful choice about what customers to survey.

LESSON NO. 1:
Distinguish between an agency's stakeholders, on the one hand, and its customers, on the other. Stakeholders are people or groups that hold a vital interest in how the agency does its job. These include constituent interest groups, congressional oversight committees, and the general public, as well as customers and employees. But customers are the people who actually engage in transactions with the agency. To evaluate agency performance, survey its customers.

LESSON NO. 2:
Survey your main customers. Paul Light says a flaw in using customer satisfaction to measure performance is that there's no such thing as a single government customer. Heck no, that's not the flaw; the flaw is picking just a single government customer to survey.

Most agencies have many groups of customers, often with conflicting interests. The U.S. Forest Service, for example, simultaneously serves recreational, forestry and mining interests. Veterans Benefits surveys customers for all its services, from home loans to vocational rehabilitation services. It even surveys customers who were turned down for benefits. And Student Financial Assistance is expanding its use of the American Customer Satisfaction Index (the survey instrument the National Partnership for Reinventing Government used). Last year they surveyed students who applied for loans-their principal intended beneficiaries. Next year they'll also survey college-aid officers, high school guidance counselors and financial institutions that make student loans.

LESSON NO. 3:
Distinguish between customers to whom an agency is primarily delivering a service-like students getting aid-and customers to whom an agency delivers an obligation-like taxpayers obliged by the Internal Revenue Service to pay their taxes. Many agencies deliver both an obligation and a service. Treating the individual citizen respectfully and responsively (i.e., with great service) is critical. Doing so is the most effective, least expensive way to deliver the obligation. So for sure, survey those who are obliged, whether by the IRS, OSHA, EPA, FAA or the other agencies. But we might expect satisfaction ratings lower than those from customers who are using a service of the federal government.

Identifying customers and regularly measuring their satisfaction serves three important purposes:

• It gives agencies a clear focus.
• It provides a strong vehicle for holding agencies accountable.
• It provides agencies with feedback crucial to improving their services.

  Focus, accountability and feedback are strong levers for reinventing government. The federal government should continue this worthwhile initiative. And agencies should take seriously the choice of customers to survey.

Babak Armajani is CEO of the Public Strategies Group Inc. Bob Stone, now a partner in PSG, was project director and energizer in chief of the National Partnership for Reinventing Government from 1993 to 1999.