The Science of Organizational Design
A contingency view of organization design
Simon writes:
The division of labor is quite as important in organizing decision making as in organizing production. From the information-processing point of view, division of labor means factoring the total system of decisions that need to be made into relatively independent subsystems, each one of which can be designed with only minimal concern for its interactions with the others. The division is necessary because the processors that are available to organizations, whether humans or computers, are very limited in their processing capacity in comparison with the magnitude of the decision problems that organizations face.
How should the big task be structured or partitioned, and what resources should be allocated to the particular task? For example, deciding between a functional and divisional structure is choosing the basis for breaking up the big task. Then you must choose how many departments or divisions you would like to have. For the divisional structure, you can choose to allocate private customers to one division and corporate customers to another, or you can base your divisions on types of products. Concurrent with the structure, you have made a choice of how to coordinate. Coordination mechanisms require information, communications, cooperation, decisions, rules, routines, trust, incentives, and leadership, among others.
Thompson analyzes the organization in terms of uncertainty and technology or work flows of pooled, sequential, and reciprocal relations. Miles and Snow analyze the design problem in terms of structure and process and develop a typology of four organizational prototypes based on a particular choice of strategy. Lawrence and Lorsch use the concept's differentiation and integration, which are similar to structure and coordination, and develop a contingency theory based on the particular type of environment. These approaches are single-contingency theories.
Using the information-processing concept, Burton and Obel developed the multi-contingency theory of organizational design and further developed these concepts in Burton et al. This view says that an organization's design should be chosen based on the particular context and further that the description of the context should be multi-dimensional, including both structural and human components. Structural components of organizational design include goals, strategy, and structure and tasks. Human components include leadership, work processes, and people. Coordination includes control systems, decision systems, information systems, and incentive mechanisms.
In the multi-contingency theory, the relationships between structural, human, and coordination components are represented as a series of interconnected design rules. Design rules are "what should be" relationships. They incorporate both feasibility of "what might be" and desirability for the organization. The development of design rules has originally been related to simple design rules focusing on one or a limited set of contingencies, such as Lawrence and Lorsch on the relationship between environment and organizational design, or Woodward on technology and organizational design. These design rules were based on observation of what is. Later, these simple design rules have been combined into a set of more complex design rules.
Design rules can be based upon "what is," using the logic that what has been successful in the past in somewhat similar conditions is likely to work for the future - even applied in circumstances going beyond what has been observed. Design rules can also be developed based on the theory of balancing the information-processing capacity with information-processing demand. Further experimentation using simulations, laboratory studies, and empirical research are the basis for design rules.