BUS603: Managing People

By science, we mean knowledge and understanding about the world through observation of what is, and experimentation of "what is" and "what might be". Design focuses on imagination and creation of what might be to achieve a purpose, moving toward "what should be".

Organizational theory provides the theoretical underpinnings for organizational design. Organization theory describes and explains for our understanding how the world works; in complement, organization design builds on this to understand how the world could possibly work. Organization theory is a positive science to explain and understand the structure, behavior, and effectiveness of an organization - what is; organizational design is a normative science to recommend what might be designs for increased effectiveness and efficiency.

According to Simon, an organization is an artifact that must be created in concept before it is created in reality. Romme, building upon Simon, argues that the "idea of a design involves inquiry into systems that do not yet exist - either complete new systems or new states of existing systems". Organizational design is thus prescribing how an organization should be structured in order to function effectively and efficiently. Organization design is a systematic approach to aligning structures, processes, leadership, culture, people, practices, and metrics to enable organizations to achieve their mission and strategy. The basic premise is that there is no one best way of organizing and that different organizations are not equally effective or efficient. This introduces the concept of contingency thinking, where the organization should be designed to fit the particular circumstance, which may be new and not experienced before.

How can we create information or knowledge about something that does not yet exist? We need knowledge-based experimentation and observation. Observation is an "as is" experiment or natural experiment where the researcher describes the "what is" situation and variations among variables that are manipulated by others or nature. Experimentation involves manipulation of variables to understand the effects.

We need to experiment to generalize from one study to another. We want to understand cause and effect in the science of organizational design through experimental manipulation of factors. There are a number of methodologies: simulation, laboratory studies, field studies, ethnographies, and large data analyses, among others. All utilize experimentation and observation to investigate and to understand the world of what is, as well as to explore the world of what might be.

Simulation modeling provides a powerful methodology for advancing theory and research on complex behaviors and systems. The modeling requires current knowledge, and the simulations allow experimentation with models of new designs with full control of the setup. Laboratory experiments introduce human behavior and require less formal modeling. Much of the power, beauty, and pleasure of models comes from inventing and elaborating models for exploring their implications in new domains. This is the basis for the experimentation of what might be in a way that the results can be generalized. Many experiments are done in the real world, but they may not be easily generalized. A triangulation approach using observation, simulation, and laboratory experiments will drive the knowledge and science of organization design forward.

The challenge for the science of organization design is to create predictive models of future organizational designs. Prototyping new organizational designs could happen either through simulations or in the lab - with new proposed organizational arrangements being tested for unanticipated consequences before being implemented. Levitt states that "organizational chemistry" (goal conflict, institutional differences, etc.) and "organizational biology" (individual learning, organizational learning, evolution and regeneration of networks of organizations) will eventually yield robust and accurate enough agent-based modeling, analysis, and validation so that simulation of these phenomena will become useful to managers. This is supported by Puranam who states that we need models "which goes well beyond providing general advice to prototyping new organizational designs".

In an analysis of novelty in forms of organizing, Puranam et al. argue that a new form is one that embodies new solutions to the basic problems of organizing - the division of labor and the integration of effort - in contrast to the solutions used by existing organizations.

An organization is a social unit of people that is structured and managed to meet a need or to pursue collective goals. All organizations have a management structure that determines relationships between the different activities and the members and subdivides and assigns roles, responsibilities, and authority to carry out different tasks. The activities must then be coordinated to obtain the collective goals. Structure and coordination are thus the fundamental choices in organizational design.

Structure includes the assignment of tasks to individuals or subunits, the apportionment of resources to these units, the designation of customers and markets to units, and generally the breakdown of the larger problem for smaller units. Coordination is bringing the units together through communications, IT, leadership, culture, incentives, routines and procedures, and generally what we call management.

The structure and coordination choices are not independent. Once a structure is chosen, the coordination choices are limited in order to achieve a good fit. Further, coordination requires much more information processing than finding the structure, and designing the coordination has a different time perspective than designing the structure. The structure issue is a decision problem or analytical problem, while coordination is a management issue. Coordination is done in real time, as it must be done when activities are under way. But, the coordination mechanism is a design problem.

Within the framework of structure and coordination, organizational design boils down to who does what when, or how to allocate tasks, resources, customers, etc. to each of the small problems and how to coordinate these small units and tasks. This gives the framework within which experimentation should be done. However, experimentation also requires a theoretical framework with which the experimentation should be done to allow the required ability to generalize. One such framework is the information-processing paradigm.