Loading field map
How should a manager decide? The question seems straightforward, but the answers have shifted dramatically over the past century. At the heart of the subfield lies a persistent tension: managers are expected to make choices that are rational, efficient, and well-informed, yet the real world of organizations is messy, uncertain, and shaped by human limitations. The history of managerial decision making is the story of how scholars have wrestled with this gap between the ideal of rational choice and the reality of how decisions actually get made.
The earliest systematic framework for managerial decision making was the Rational Decision-Making Model, which took shape in the 1940s and 1950s. It assumed that a decision maker could identify a clear problem, gather complete information about all possible alternatives, evaluate each option against a set of consistent preferences, and then select the alternative that maximized utility. This model was deeply influenced by classical economics and statistical decision theory. It treated the manager as a fully informed, logically consistent optimizer. The model was prescriptive: it told managers how they ought to decide, not necessarily how they actually did. For decades, it served as the benchmark against which all later frameworks would measure themselves.
The first major challenge to the rational model came from Herbert Simon, who introduced the concept of Bounded Rationality in 1947. Simon argued that human cognitive capacities are limited: managers cannot process all available information, cannot foresee all consequences, and cannot hold all possible alternatives in mind. Instead of maximizing, they satisfice—they search for a course of action that is good enough to meet a minimum threshold. Bounded Rationality did not reject the idea of rationality altogether; it proposed a more realistic version that acknowledged cognitive constraints. This framework has proven remarkably durable and remains a foundational concept in the field today.
Simon's ideas were taken up and extended by the Carnegie School, a group of researchers at Carnegie Mellon University active from the 1950s through the 1970s. The Carnegie School treated organizations not as rational machines but as coalitions of participants with conflicting interests. Decision making, in this view, was a political process of bargaining and negotiation, not a purely analytical one. The school emphasized that organizations simplify complex decisions through routines and standard operating procedures. This was a direct narrowing of the rational model's scope: instead of assuming a single decision maker with unified goals, the Carnegie School insisted that organizational decisions emerge from conflict and compromise among multiple actors.
Building directly on the Carnegie School, the Behavioral Theory of the Firm, articulated by Richard Cyert and James March in 1963, provided a more detailed model of how organizations actually make decisions. It argued that firms learn from experience, adjust their goals over time, and rely on rules of thumb rather than exhaustive analysis. The theory introduced concepts like problematic search (search triggered by a performance shortfall) and organizational slack (excess resources that buffer the organization from uncertainty). This framework absorbed the Carnegie School's emphasis on coalitional politics and added a dynamic, learning-oriented dimension. It showed that organizational decision making is not a one-time event but an ongoing process shaped by feedback and adaptation.
While the behavioral frameworks had challenged the rational model's assumptions about information and cognition, they still assumed that decision making followed some kind of systematic process, however imperfect. A more radical departure came from Charles Lindblom's Incrementalism, introduced in 1959. Lindblom argued that in complex policy and organizational settings, decision makers do not compare comprehensive alternatives. Instead, they make small, incremental adjustments to existing policies—a process he called "muddling through." Incrementalism rejected the idea that decisions are made through a clear sequence of problem identification, alternative generation, and evaluation. It portrayed decision making as a series of small steps, each negotiated among stakeholders, with no single moment of choice. This framework coexisted uneasily with the behavioral models: both acknowledged limits on rationality, but Incrementalism went further by questioning whether any systematic decision process existed at all.
The Garbage Can Model, developed by Michael Cohen, James March, and Johan Olsen in 1972, pushed this skepticism even further. It described organizational decision making as a chaotic process in which problems, solutions, participants, and choice opportunities float around independently and come together only when a random confluence occurs. Decisions happen not because a rational process leads to them, but because a solution happens to be available when a problem arises and someone is paying attention. This model was a direct challenge to both the rational model and the behavioral models: it suggested that organizational decisions are often the result of timing and luck rather than intention or learning. The Garbage Can Model was most influential in the 1970s and 1980s, particularly in studies of universities and public organizations, but its radical implications—that decision making can be fundamentally anarchic—remain a provocative counterpoint to more orderly frameworks.
While the process-oriented frameworks focused on organizational and political dynamics, a parallel line of research turned inward to the individual decision maker's mind. The Heuristics and Biases program, launched by Daniel Kahneman and Amos Tversky in the early 1970s, drew on cognitive psychology to identify the specific mental shortcuts (heuristics) that people use when making judgments under uncertainty. These heuristics—such as availability, representativeness, and anchoring—often lead to systematic errors (biases). This framework provided a micro-level explanation for Bounded Rationality: it showed not just that people are limited, but exactly how they are limited and what kinds of mistakes they tend to make. The Heuristics and Biases program was experimental and laboratory-based, and it remains highly active today, influencing fields from behavioral economics to public policy.
Prospect Theory, introduced by Kahneman and Tversky in 1979, refined the Heuristics and Biases approach by focusing specifically on decision making under risk. It showed that people evaluate potential gains and losses relative to a reference point, not in absolute terms, and that they are loss-averse: losses hurt more than equivalent gains please. This asymmetry leads to risk-seeking behavior when facing potential losses and risk-averse behavior when facing potential gains. Prospect Theory absorbed the Heuristics and Biases framework's emphasis on systematic deviation from rationality and added a precise model of how framing and reference points shape choices. It coexists with Heuristics and Biases as a complementary account of cognitive limitations, with Prospect Theory specializing in risky choice and the broader heuristics program covering a wider range of judgment tasks.
By the 1990s, a growing number of researchers felt that the laboratory-based cognitive models, while powerful, missed something essential about how decisions are made in real-world settings. Naturalistic Decision Making (NDM) emerged as a direct critique of the Heuristics and Biases program's methods. NDM researchers studied experts—firefighters, nurses, military commanders, pilots—making time-pressured, high-stakes decisions in their natural environments. They found that experts do not typically compare alternatives or calculate probabilities; instead, they recognize patterns from past experience and match the current situation to a familiar prototype. This process, called Recognition-Primed Decision Making, is fast, intuitive, and context-dependent. NDM did not reject the idea of cognitive limitations, but it argued that the heuristics-and-biases view overemphasized errors and underemphasized the adaptive expertise that experienced decision makers develop. The two frameworks remain in a productive tension: Heuristics and Biases focuses on universal cognitive mechanisms and their pitfalls, while NDM emphasizes the role of domain-specific experience and situational awareness.
Today, managerial decision making is a pluralistic field. Bounded Rationality remains a foundational infrastructure that most frameworks accept as a starting point. The Heuristics and Biases program and Prospect Theory continue to generate research on cognitive biases and their implications for management, finance, and policy. Naturalistic Decision Making has carved out a strong niche in fields like emergency management, military command, and healthcare, where expertise and real-world context matter most. The older organizational frameworks—the Behavioral Theory of the Firm, Incrementalism, and the Garbage Can Model—are less active as distinct research programs, but their insights have been absorbed into broader theories of organizational learning, strategy, and change.
What do the leading frameworks agree on? They all reject the classical rational model as an accurate description of how decisions are actually made. They all acknowledge that cognitive, organizational, and contextual factors constrain decision making. They all see decision making as a process that unfolds over time, not a single moment of choice. Where they disagree is on the level of analysis (individual cognition vs. organizational process vs. expert intuition), the methods they favor (controlled experiments vs. field observation vs. computational modeling), and the extent to which deviations from rationality are seen as errors or as adaptive responses to complexity. The cognitive frameworks tend to emphasize universal biases and the potential for debiasing interventions, while the naturalistic frameworks emphasize the irreplaceable value of experience and the limits of formal analysis. This disagreement is not a sign of weakness; it reflects the richness of a subfield that has learned to ask different questions about the same fundamental problem: how managers can make better decisions in a world that never gives them all the information they need.