The Decision-Making Process According to Herbert A. Simon

Every organisation – whether a government ministry, a municipal corporation, or a district collectorate – is constantly making decisions. Some are routine; others are consequential and complex. Yet for much of the twentieth century, administrative theory had surprisingly little to say about how decisions actually get made. It took Herbert A. Simon – Nobel laureate, political scientist, and one of the most original minds in the study of organisations – to place decision-making at the very centre of administrative analysis. His framework breaks the process down into three clear, interconnected phases: intelligence, design, and choice. Far from being abstract theory, this model offers a practical map for anyone trying to understand how organisations move from recognising a problem to acting on it.

Simon and the study of decision-making 🔗

Herbert Alexander Simon (1916-2001) was an American economist and political scientist whose primary interest was the human decision-making process within organisations. He received the Nobel Prize in Economics in 1978 and the Turing Award in 1975, spending most of his academic career at Carnegie Mellon University. His seminal work, Administrative Behavior, first published in 1947, was described by the Royal Swedish Academy of Sciences – the body that awarded him the Nobel Prize – as “epoch-making.” The book asserted that decision-making is the heart of administration and that the vocabulary of administrative theory must be grounded in the logic and psychology of human choice.

What made Simon’s contribution distinctive was his rejection of the classical assumption that decision-makers are perfectly rational. Earlier theories depicted the so-called “economic man” – an ideal actor with access to complete information, unlimited cognitive capacity, and the ability to identify the single best solution from all possible alternatives. Simon found this picture deeply unrealistic. In its place, he proposed the concept of bounded rationality: the recognition that real decision-makers operate under constraints of limited information, limited time, and limited cognitive capacity. As a consequence, rather than maximising outcomes, they satisfice – a term Simon coined by merging “satisfy” and “suffice” – meaning they search for a course of action that is good enough to meet their goals, rather than exhaustively optimal.

This realistic foundation made Simon’s framework especially relevant for understanding public administration, where officials routinely face complex, underdefined problems under resource and time pressures. Building on this foundation, Simon articulated a structured, three-phase model of the decision-making process.

The three phases of Simon’s decision-making process 🔗

Simon argued that a systematic decision-making process involves three major phases: intelligence, design, and choice. These phases are sequential, but not rigidly so – at any point, the decision-maker may loop back to an earlier stage for further validation or refinement. Simon was aware of this interdependence and provided examples of feedback from one stage into another, noting that each phase can itself be approached as a mini decision-making process.

Phase 1: Intelligence 🔗

The word “intelligence” here does not mean cognitive ability – it refers to the act of gathering intelligence about a situation, much as a military commander surveys the terrain before acting. The intelligence phase involves scanning the environment, either intermittently or continuously, to identify problem situations or opportunities.

In this phase, the decision-maker attempts to determine whether a problem exists, identify its symptoms, gauge its scale, and define it clearly. This is harder than it sounds. Often, what appears to be a problem – such as excessive costs – may only be a symptom of a deeper issue, such as improper inventory levels. Because real-world problems are usually complicated by many interrelated factors, it can be difficult to distinguish between symptoms and the actual problem.

The intelligence phase involves two key activities: problem searching and problem formulation. In problem searching, the decision-maker compares the actual state of affairs against expected standards or targets, measuring the gap between what is and what should be. In problem formulation, the problem is defined with sufficient precision to allow for solution development. A common risk at this stage is solving the wrong problem – and drawing on analogies with previously solved problems can help avoid this trap.

Consider a district administration that receives a surge of complaints about delays in issuing birth certificates. The intelligence phase involves collecting data on the volume and nature of the complaints, identifying whether the delay lies in staffing, digitisation gaps, or inter-departmental coordination, and framing the problem precisely before moving to solutions.

Phase 2: Design 🔗

Once the problem is clearly understood, the process moves into the design phase. The design phase involves developing a model of the problem situation on which different decision alternatives can be generated and tested. At this stage, the decision-maker is not selecting a solution – the goal is to invent, develop, and analyse possible options before evaluating them.

Each alternative solution is evaluated after gathering relevant data about it. The evaluation is done on the basis of criteria to identify the positive and negative aspects of each solution. Quantitative tools and models are used at this stage, and solutions at this point are only outlines – meant for analysis of their suitability rather than immediate action. A great deal of creativity and systematic thinking is required here.

This phase is often underappreciated in practice. Organisations under pressure tend to jump from problem identification directly to selecting a familiar solution. Simon’s model insists on the importance of generating a genuine range of alternatives. In the birth certificate example above, the design phase might produce several options: adding staff at counters, linking the civil registration system with hospital databases, enabling online applications, or introducing a tracking mechanism for pending cases. Each is a distinct alternative with different cost, feasibility, and implementation timelines.

The design phase also involves testing alternatives against a model of the problem. The main goal of the design phase is to define and construct a model representing the system, by defining relationships between collected variables. Once the model is validated, criteria for the choice phase are established and several possible solutions are identified.

Phase 3: Choice 🔗

The choice phase is where the actual decision is made – one alternative is selected from those developed in the design phase. The “best” solution may be identified using quantitative tools like decision tree analysis or qualitative tools like force field analysis. This is not straightforward, because each solution presents a different scenario, and the problem itself may have multiple objectives, making the choice process a genuinely difficult one.

It is important to note that in Simon’s framework, “best” does not mean optimal in an absolute sense. Consistent with bounded rationality, the choice selected is the one that appears most satisfactory given the available information and existing constraints. The decision-maker is not an omniscient calculator but a pragmatic actor working within real limits. Rather than considering all relevant factors and alternatives to make optimal decisions, individuals limit their search and focus on only a few options to make decisions that are “good enough” to meet their aspiration levels.

Returning to the certificate example: after evaluating the alternatives designed in the previous phase, the administration might choose to integrate hospital databases with the civil registration portal and simultaneously set up a dedicated helpdesk. This may not be a perfect solution, but it is a workable, implementable one given available budget and technology.

The iterative nature of the model: feedback loops and implementation 🔗

Simon’s model is not a simple linear pipeline. There is a continuous flow of activity from intelligence to design, from design to choice; but at any phase, there may be a return to a previous phase – feedback – for validation, verification, and refinement. This iterative quality is one of the model’s most realistic and enduring features. If the design phase reveals that the problem was poorly defined, the decision-maker returns to intelligence. If the choice phase uncovers that no satisfactory alternative exists among those developed, the decision-maker returns to design.

Simon later added a fourth phase: implementation. In implementation, all the previous steps – intelligence, design, and choice – are put into action. Successful implementation results in a solution to the defined problem. On the other hand, failure brings the process back to an earlier phase. Implementation, in other words, is not the end of the story; its outcomes feed back into the intelligence phase of future decisions.

Even a theoretically sound decision can fail if poorly implemented. And when outcomes – successful or unsuccessful – are monitored, those results inform future decision-making, reinforcing successful approaches and modifying unsuccessful ones. This is why Simon’s model is sometimes described as cyclical rather than purely sequential.

Programmed and non-programmed decisions 🔗

Simon’s decision-making framework is closely tied to another of his important contributions: the distinction between programmed and non-programmed decisions. Programmed decisions are repetitive, routine, and governed by established procedures or rules – a treasury officer sanctioning a routine expenditure within pre-approved limits, for instance. Non-programmed decisions are novel, unstructured, and require fresh analysis – such as a state government deciding how to respond to an unprecedented flood of a scale not seen before.

The three-phase model applies to both, but its value is greatest for non-programmed decisions. When problems are new and solutions are not obvious, the intelligence-design-choice sequence provides a structured method that helps prevent hasty, intuition-driven choices. Simon observed that higher organisational levels typically deal with more non-programmed decisions, while lower levels handle more programmed ones. This has important implications for how administrative hierarchies are structured and where analytical capacity needs to be concentrated.

Relevance for public administration 🔗

Simon’s three-phase model has had a lasting influence on how public organisations are designed and how administrative decision-making is studied. Simon’s concepts of bounded rationality and satisficing heavily influenced classic public administration work, including scholarship on incremental policy-making and the budgeting process. His insight that the formal structure of organisations shapes individual decision-making – by assigning roles, developing decision premises, and establishing communication channels – fundamentally changed how administration is theorised.

In the Indian context, public administration scholars and reformers have drawn on Simon’s framework in several ways. Administrative reform commissions, performance management systems, and e-governance initiatives all embed elements of Simon’s model: systematic data collection (intelligence), structured appraisal of options (design), and rule-based or evidence-informed selection (choice). The Right to Information Act, 2005, for instance, can be understood partly as a mechanism for improving the quality of intelligence available not just to administrators but to citizens, enabling more informed scrutiny of government decisions.

Simon also influenced the development of decision support systems (DSS) – information technology tools designed to assist managers in each phase of decision-making. Simon’s model has been foundational in the field of information technology management surrounding the development of decision support systems. From data dashboards used in urban local bodies to policy simulation tools in central ministries, the practical footprint of Simon’s framework is wide.

Criticisms and limitations 🔗

Simon’s decision-making model is not without its critics. The most frequent objection is that the model, while analytically useful, is too general to serve as a guide for specific decisions. By focusing on the structure of the process, it says little about the substantive content of what counts as a good decision in any particular context.

A second criticism concerns the continued primacy of rationality in Simon’s framework – even bounded rationality is still rationality. Critics from sociology, political science, and organisational behaviour have pointed out that decisions in real organisations are heavily shaped by power dynamics, institutional culture, personal relationships, and political pressures that do not fit neatly into the intelligence-design-choice sequence. In the Indian administrative context specifically, factors such as hierarchical deference, inter-departmental rivalry, and political interference can significantly alter how each phase unfolds in practice.

A third critique relates to Simon’s fact-value distinction – his argument that facts and values are separable elements in decision-making. Scholars like Norton Long have challenged this, arguing that administrative decisions are inherently value-laden and that no neutral, fact-based process can be truly disentangled from questions of social justice, equity, and political priorities.

Despite these criticisms, Simon’s pioneering work on bounded rationality and decision-making has had a profound and lasting impact on public administration, shaping scholarship across decision-making, human performance, and organisational knowledge. His record-breaking citation count – approaching half a million on Google Scholar – is testimony to how deeply his ideas have penetrated multiple disciplines.

Why Simon’s framework still matters 🔗

The enduring value of Simon’s decision-making process lies not in its claim to describe what perfect decisions look like, but in its honest account of how real decisions happen. By articulating the intelligence, design, and choice phases, Simon gave administrators, researchers, and reformers a common vocabulary and a practical scaffold for improving organisational decision-making. He showed that good decisions do not emerge spontaneously from individual genius or institutional authority – they require deliberate, structured effort at every stage, from problem identification through to implementation.

For students of public administration, this framework offers more than exam material. It provides a lens through which everyday administrative events – a policy reversal, a delayed infrastructure project, a poorly designed welfare scheme – can be analysed with precision. In most such cases, a careful look reveals a failure in one of Simon’s three phases: a problem misdiagnosed in the intelligence phase, too narrow a range of options developed in the design phase, or a selection in the choice phase that ignored key constraints. The remedy, Simon would suggest, is not to demand perfection, but to build better processes.

What do you think? When an administrative policy fails – such as a social welfare scheme that does not reach its intended beneficiaries – which of Simon’s three phases do you think is most likely to have broken down, and why? Is it enough to have a sound decision-making process, or do factors outside Simon’s model ultimately determine the outcomes of public decisions?

References

1. https://www.nobelprize.org/uploads/2018/06/simon-lecture.pdf
2. https://www.informit.com/articles/article.aspx?p=2992600&seqNum=2
3. https://onlinelibrary.wiley.com/doi/10.1111/puar.13540
4. http://managementstudyonline.blogspot.com/2014/02/herbert-simon-decision-making-model.html
5. https://unitfly.com/insights/decision-making-process/
6. https://banotes.org/administrative-thinkers/organizational-decision-making-herbert-simon-models/
7. https://www.researchgate.net/publication/228887070_Practical_Decision_Making-From_the_Legacy_of_Herbert_Simon_to_Decision_Support_Systems
8. https://onlinelibrary.wiley.com/doi/10.1111/padm.13051?af=R