Editor's note: Richard Garfein, Ph.D., is vice president, marketing research, American Express International.

In the early 1980s, consumers were passive in the face of deteriorating service. Today they are increasingly intolerant of poor service, and companies themselves, in their advertising, are more explicit in helping customers become more demanding. For example, a recent AT&T commercial showed a former AT&T customer getting poor service from his new long distance company and then deciding to switch back to AT&T.

Free trade agreements and heightened international competition will intensify the challenge of winning customer loyalty. This is a challenge that America needs to meet in competing globally during this decade and beyond. A more intimate understanding of customers and of what drives customer satisfaction is certainly called for.

With regard to American Express, we have always had an imperative to deliver the highest quality service. The American Express Card's higher annual fee - both in the U. S. and abroad - needs to be justified by higher quality service.

We have found that the meaning and nuances of customer satisfaction can vary significantly across international settings. In Latin America, for example, the combination of high inflation and slow postal systems makes the experience of owning a credit card quite different than we would ever imagine it to be in the U.S. Credit card bills are actually delivered to the cardholder's home or office by courier. And cardholders must then pay their bills in person. Good service from a credit card takes on a much different meaning in this kind of setting, and is based on considerations such as:

  • Does the bill arrive on time? (Given the high rate of inflation, a penalty is assessed for every day of late payment.)
  • Is the normal interval between receipt of bill and due date acceptable?
  • How convenient are the available options for paying the bill (e.g., bank branches, ATMs, etc.)?

Based on our experience at American Express, I would recommend the following as guidelines for producing high quality customer satisfaction research in international settings.

1) Insist on the highest quality. As a general rule, bear in mind that good market research is better than no market research, but no market research is better than bad market research. Bad market research means wasted time and money, as well as misleading information. A study is only worth doing if it can be done correctly.

2) Hire an in-house market research professional in each of your major markets. This is someone who would manage the market research process and have day-to-day interface with outside vendors and internal marketing people. Without the in-house professional, you will find that managing the market research process long distance is virtually impossible.

3) Manage the process closely. If you hire local vendors to carry out your studies (which we do most of the time), you need to spend a lot of time with the vendors-more than you would generally think of spending in the U.S. Make sure that safeguards are in place against low quality and possible fraud. Participate in interviewer briefings, monitor the interviewing, and always conduct pilot tests. These cautions not withstanding, we have found outstanding market research vendors in Latin America, and we have had a generally positive experience.

4) Find the ideal middle ground between centralization and decentralization. From the headquarters' perspective, "re-inventing the wheel" is inexcusable, and clearly, it is in a market's best interests to build on expertise gained in other countries. Also, between-market comparisons (e.g., France vs. Brazil) can be very meaningful, over and above what is obtained in your within market tracking studies. Nevertheless, an overly centralized and rigid approach can jeopardize the local relevance (and the local sense of ownership) of whatever you put in place. Flexibility and common sense are called for, and also the realization that what worked in one country might not work in another.

5) Focus on your weaknesses. There is a tendency to focus on strengths, and it is only natural that we prefer to hear good news. However, it is important to use research to identify and learn more about your weaknesses (and your competitors' strengths). Write the questionnaire as though you work for the competition. And remember, weaknesses can never hurt you in a study, but they can hurt you in the marketplace.

6) As a general rule, customer satisfaction studies ought to be set up as tracking studies, not as one-time-only events. The frequency of waves will depend on the market's stability over time and on its relative importance. Tracking studies should be changed (vitalized) regularly in moderation. Plan and design tracking studies carefully with every intention of having them last until the year 2000.

7) Be cautious in using customer satisfaction measurement to appraise management performance. Don't let it become a punitive exercise, be careful that it doesn't adversely affect morale, and keep in mind that there are error margins and sources of bias in all survey work. Customer satisfaction measurement should be used primarily as a management tool, and less so as a management report card. That is what true "empowerment" really means.

8) Assess the credibility and effectiveness of advertising claims. As noted earlier, a lot of companies are emphasizing service quality in their advertising. A good customer satisfaction research program will guide you as to what messages will be seen as credible, honest, and straightforward.

9) Focus your research on the key customer segments of your business. At American Express, those segments include new Cardmembers (who are more impressionable than longer-tenured Cardmembers), heavy users (who account for a disproportionate share of spending), and people who dropped their Amex Cards (to find out exactly what went wrong).

10) Customer satisfaction measurement can only be considered successful if it meets these four criteria:

a) The study is well conceived, executed and analyzed.

b) The results are effectively communicated to management.

c) The recommendations made in the study are implemented.

d) The actions taken have the predicted (or better-than-predicted) impact on the business.