A clientric, or client-centered, business is one that knows the names of all or nearly all of its clients and has the goal of lifelong service to them. I have written about the distinction between trade, industry, and clientric business at much more length in Gods of Commerce (Clear Glass, 1997).

I've been a manager, have advised managers, and have taught management for years. I have worked with over a thousand small businesses, and as a result I have high degree of skepticism about the management fads that come and go every few years.

I have written and taught about why institutions can't and don't change. This failure to change is based on a mechanism known as social sorting. People who become part of any institution proceed through three stages of social sorting:

* People are first sorted by the flag, which is the accumulation of the knowledge, opinions, data, rumors, and images that attract a select group of people to try and join an institution in the first place.

* These people are then sorted by the screen, the process by which the institution selects the people it prefers.

* These screened people then sort themselves in a dynamic known as overflow, in which the experience of being a part of an institution for several years leads some to stay and others to leave.

After this process of social sorting, an institution has a degree of homogeneity that perpetuates itself against all changes. The rules of social sorting apply to most institutions, including fraternities, corporations, government agencies, political parties, and local PTA's.

Also important to my view of management are the ideas, data ,and research of Henry Mintzberg, management expert from MIT who has demonstrated that businesses have common managerial structures in each distinct industry, because the function of management is to process and reduce the information that is available in the market of that industry (Mintzberg on Management: Inside Our Strange World of Organizations, New York, Free Press, 1989).

Another researcher who has added to my thinking about management is Milton Moskowitz, a San Francisco Bay Area corporate researcher, who examined the life of employees in hundreds of corporate environments. He found that the style of management a business follows has nothing to do with the satisfaction of the employees. Companies with a policy of "dress code, get promoted or fired, and work on your own time until midnight" had employees who were just as satisfied and loyal as those in companies with "casual dress, employee gardens, in-plant child care, company retreats, and long paid vacations."

What can we say about management if we believe that little or no change can occur in the organization, that its structure is largely dictated by market and industry factors, and that management style is uncorrelated with employee satisfaction? Plenty.

Let me clarify what management really is, a distinction that is not often understood. Management is not supervision. The supervision of employees is a well-worked-out field in which classes, textbooks, and on-the-job training yield high rewards. A good supervisor can produce results that are measurably superior to an average supervisor. However, he or she is not a manager. Encouraging employees to do their best is supervision, and despite its title, the popular book The One-Minute Manager is about supervising employees, not management.

Hiring employees is a basic management issue. Each person you hire should add something positive to the organization as well as maintaining the existing level of quality. I had a client who interviewed 100 people before hiring his first employee and at least thirty for each one after that. He was wise to do so, because he got the best he could and in the process came to understand the market of potential employees much better. He hired employees who were a perfect match and who stayed with the company for a long time. I have another friend who spends six hours, over several days, interviewing each of his final screened selection of potential employees. He has great employees.

Neither of these employment approaches should be considered extreme; they are reasonable if you understand the significance of this organizational front door. Few if any hiring decisions at the management level can be delegated.

One friend has looked at the hiring process as a chance to find appreciative employees. After earlier efforts to be generous and beneficent with unappreciative employees, he chose to hire only people who had previously worked in terrible environments. These employees appreciate him.

What hierarchy of management should be used? This is not an arbitrary or capricious question. The model we use in the United States is the military. Some managers, like officers, are on the management level from the start, while others are, like enlisted personnel, occasionally promoted into the ranks of management.

In Japan, only a few people are brought in at the enlisted level, while everyone else is groomed for potential management. When the bottleneck of too many aspiring to the top position occurs, this is handled by early retirement, promotions to positions on the board of directors, and the creation of outside suppliers run by top-run aspirants.

In a clientric business, the bias is toward the Japanese model because the potential for a client to interact with the people in the organization is fairly high. Using high-quality employees is an effective way to stifle bureaucratic tendencies. In Japan, I've found many government agencies with excellent personnel that pride themselves on having small staffs. Efficiency and imagination are fostered with this model, as less time is spent delegating work to low-level workers.

When someone, or everyone, is considered future management, they get the broadest training and experience possible through rotation, holding responsibility, and being given challenges. All of this is desirable for the effective working of a healthy organization, one that has a high learning capability and a knowledgeable staff.

There are three sub-rules governing hiring and attrition:

First, everyone who comes to an organization seeking employment should be well treated. It is popular to view this as a selling or marketing opportunity. This is certainly true, but it is a very limited view of the situation. Hiring is an opportunity to create a future steady stream of the right job seekers and to elevate the importance of being hired to the existing employees. They should feel privileged to be part of the organization.

Using each job interview as a way to promote a future stream of desirable employees requires that you explain the job and workplace environment in much greater detail than just describing the wonders of the company. If you want good listeners and people who work long hours, it is necessary to show how these qualities are rewarded in the company and to show examples of successful outcomes. The message will communicated to future applicants.

Second, candid explanations of what an organization needs, and constant candor in telling applicants what parts of their experience would be valued and appreciated and the scales on which they are being weighed, will help everyone involved.

Employees of other companies who have relevant skills, knowledge, and innovative ideas should be constantly sought. They will create change and initiate improvements because they know what kind of tactics worked in their previous setting.

Third, former employees who have left the organization on good terms should be treated as alumni, included in the organization's social events and regular reunions. McKinsey and Company, a leading international consulting company, has done the best job of this. Their alumni have have helped each other to populate the top ranks of American business&emdash;all to the benefit of McKinsey. (McKinsey is often hired for consulting work by former employees).

There are four ways that an organization adapts to a changing market:

First, is to bring in consultants who can train employees on cross-industry innovations. One such example is the use of facilitated meetings (pioneered by Michael Doyle and David Strauss). An organization should always be on the lookout for new management tactics that have proved successful in other industries. These innovations should be fully supported by the top managers, with direct participation when possible&emdash;for if something doesn't work for the top people in an organization, it won't work for anyone else in the group.

Second, an organization can hire consultants who have developed a proven innovation within the industry. Innovations that work can seldom stay secret, as suppliers usually pass on information about them. Management should constantly be alert for this type of development, and the notion of rejecting ideas from other sources will be effectively overcome with constant evidence that top managers respect and use outside help in a meaningful way.

Thirdly, innovations can arise within different departments of an organization based on their outside contacts. This is most likely to occur when employees are encouraged to be active in their relevant specialties. An organization profits when as many of its employees as possible are participants in outside business activities, regardless of their occupational level or status in the organization. Janitors' conventions, for example, can be beneficial for both janitors and their bosses.

Fourthly, it should acknowledged that whistle-blowers often have valuable insights to offer on how organizations can change and grow. Mechanisms for secret whistle-blowing should be set up, and protection for whistle-blowers should be established, along with a termination appeals system. Everyone who deals with whistle-blowers should see their information as an opportunity to make the organization function better.

Finally, one of the most important management techniques is the establishment of an open-door and open books policy. Every person involved in an organization needs to be thoroughly informed about the workings of the company. Everyone needs to be able to see the whole in order to make his or her part effective.