TL;DR
Bounded rationality explains why people rarely make perfectly logical decisions. Limits in time, information, and mental capacity push us toward “good enough” choices instead of optimal ones. By simplifying priorities, using clear goals, and aligning work through OKRs, teams can reduce confusion, avoid analysis paralysis, and make more consistent decisions.
We would like to believe our decisions follow from logic and complete information. In practice, almost none of them do. We decide with partial information, under time pressure, using a mind that can hold only a few variables at once. Herbert Simon called this bounded rationality, and his point was not that people are irrational, but that perfect rationality is unavailable to anyone. Recognising that changes how you design decisions, both for yourself and for your team.
In this guide
- What is Bounded Rationality?
- Rationality vs Bounded Rationality vs Perfect Rationality
- Herbert Simon and the Theory of Bounded Rationality
- Bounded Rationality in Economics
- Satisficing: Why We Choose “Good Enough” Instead of Best
- The Bounded Rationality Model of Decision Making
- Bounded Rationality Examples: 15 Real Situations
- Bounded Rationality Psychology
- What Causes Bounded Rationality?
- Bounded Rationality in Management, Public Administration and Policy
- Beyond Bounded Rationality: Awareness, Willpower and Ethicality
- How to Reduce the Cost of Bounded Rationality
- Final Thoughts
- Frequently Asked Questions
What is Bounded Rationality?
Bounded rationality is a theory in psychology and economics holding that people do not make perfectly rational decisions. Instead, people rely on simplified decision-making procedures. This means that we often make suboptimal choices.
Where does it occur?
Bounded rationality has been found to occur in a variety of situations, including:
- When people are facing complex problems, they need to consider many factors
- When people are unable to collect all relevant information due to time pressure or other constraints
- When people are making decisions based on inaccurate or incomplete information
- When a person is influenced by emotions or other prejudices that distort judgment
Rationality vs Bounded Rationality vs Perfect Rationality
Perfect rationality assumes complete information and unlimited computation, producing the single optimal choice. Bounded rationality assumes limited information, time and cognition, producing a good-enough choice. Irrationality means deciding against your own stated interests. Bounded rationality is not a synonym for irrationality, the reasoning is sound; only the inputs and the processing budget are constrained.
The most common misreading of bounded rationality is to treat it as a polite word for irrationality. It sits between perfect rationality and irrationality, and it is much closer to the first than the second. A boundedly rational decision-maker reasons correctly from the information they have. The limitation is in the inputs and the processing budget, not in the logic.
Terminology Compared
| Term | What it assumes | Resulting behaviour | Where it comes from |
|---|---|---|---|
| Rationality | Decisions follow consistently from goals and beliefs | Coherent, goal-directed choice | General usage across economics and philosophy |
| Perfect (unbounded) rationality | Complete information and unlimited computation | Always selects the optimal option | Classical economic models; the assumption Simon challenged |
| Bounded rationality | Information, time and cognition are all limited | Satisfices, selects the first acceptable option | Herbert A. Simon, Administrative Behavior (1947); Models of Man (1957) |
| Ecological rationality | Simple rules can be well matched to the structure of an environment | Fast heuristics that often outperform complex models | Gerd Gigerenzer, Simple Heuristics That Make Us Smart (1999) |
| Irrationality | Choices conflict with the decision-maker’s own stated goals | Self-defeating or internally inconsistent behaviour | Clinical and colloquial usage |
| Boundedly rational | Adjective describing an agent operating under these limits | N/A | Used to describe agents, firms and models |
Difference between rationality and bounded rationality
Rationality, in the classical sense, describes a decision procedure: identify all options, evaluate them completely, choose the best. Bounded rationality describes the same intent operating under real constraints. The distinction is not about intelligence or motivation, it is about whether the procedure can actually be executed. Simon’s point was that the classical procedure cannot be executed by any real decision-maker, human or artificial, which makes it a poor foundation for predicting behaviour.
Profit.co Brand Signal: The distinction matters at work because the two diagnoses lead to opposite responses. If a decision went badly because someone was careless, the answer is accountability. If it went badly because the option set was narrow and the acceptance threshold was never stated, the answer is structure, and treating the second case as the first is how organisations lose good people to process failures. Profit.co’s performance management software is designed around that distinction: reviews that examine what the goal actually was, not just whether it was hit.
Herbert Simon and the Theory of Bounded Rationality
Bounded rationality was developed by Herbert A. Simon (1916-2001), an economist, political scientist and cognitive psychologist at Carnegie Mellon University. He introduced the argument in Administrative Behavior (1947), formalised it in “A Behavioral Model of Rational Choice” in the Quarterly Journal of Economics (1955), and named it in Models of Man (1957). He received the Nobel Memorial Prize in Economic Sciences in 1978.
Timeline: How the Theory Developed
| Year | Work or event | Contribution |
|---|---|---|
| 1947 | Administrative Behavior | Argued that administrative decisions are made by people with limited knowledge and attention, not by omniscient optimisers |
| 1955 | “A Behavioral Model of Rational Choice,” Quarterly Journal of Economics | Formalised the model, introduced the acceptability threshold and the idea of stopping the search once it is cleared |
| 1956 | “Rational Choice and the Structure of the Environment,” Psychological Review | Established that decision quality depends jointly on the mind and the structure of the environment |
| 1957 | Models of Man | The term “bounded rationality” enters the literature |
| 1972 | “Theories of Bounded Rationality” | Consolidated the framework and distinguished it explicitly from substantive rationality |
| 1975 | A.M. Turing Award, shared with Allen Newell | Recognised for foundational work in artificial intelligence and the psychology of human cognition |
| 1978 | Nobel Memorial Prize in Economic Sciences | Awarded for pioneering research into the decision-making process within economic organisations |
Simon’s scissors: why the environment matters as much as the mind
Simon compared human rationality to a pair of scissors. One blade is the cognitive limitation of the decision-maker; the other is the structure of the environment in which the decision is made. Studying either blade alone explains nothing, because cutting requires both. This is the reason bounded rationality is a practical concept rather than merely a critique: if the environment is one of the two blades, redesigning the environment improves decisions without requiring anyone to become smarter. Clear priorities, fewer options, better defaults and explicit thresholds all work on the environmental blade.
Who Extended the Theory
| Researcher | Contribution | Key work |
|---|---|---|
| Herbert A. Simon | Established bounded rationality and satisficing; argued limits are structural, not failures | Administrative Behavior (1947); Models of Man (1957) |
| Daniel Kahneman and Amos Tversky | Documented systematic biases produced by heuristics, showing the errors are predictable rather than random | “Judgment under Uncertainty: Heuristics and Biases,” Science (1974) |
| Gerd Gigerenzer | Argued that simple heuristics are often adaptive rather than deficient, the ecological rationality position | Simple Heuristics That Make Us Smart (1999) |
| Richard H. Thaler | Applied bounded rationality to economic policy and choice architecture; Nobel laureate 2017 | Nudge, with Cass Sunstein (2008) |
| Barry Schwartz, Sheena Iyengar and Rachael Wells | Showed maximising produces better objective outcomes but worse subjective ones | “Doing Better but Feeling Worse,” Psychological Science (2006) |
| Max Bazerman and colleagues | Extended the “bounded” family to awareness, willpower and ethical judgement | Blind Spots, with Ann Tenbrunsel (2011) |
The Kahneman-Gigerenzer distinction is worth understanding because it is the live debate in the field. Kahneman and Tversky treated heuristics primarily as sources of systematic error. Gigerenzer argued that simple rules often outperform complex models in uncertain environments, and that judging heuristics against an unachievable optimisation standard misses the point. Both positions accept Simon’s premise; they disagree about whether the resulting behaviour should be read as deficiency or adaptation.
Profit.co Brand Signal: Simon’s scissors argument is the reason goal frameworks work at all. If the environment is one of the two blades, then the way priorities are structured, surfaced and reviewed changes decision quality directly, without asking anyone to think harder or work longer. That is the whole mechanism behind OKRs: they operate on the environmental blade. Profit.co builds the software that makes objectives visible, measurable and reviewed on a cadence, so the decisions taken inside that structure improve on their own.
Bounded Rationality in Economics
In economics, bounded rationality is the assumption that economic agents make reasonable decisions using limited information, limited time and limited computational capacity, rather than maximising utility with perfect knowledge. It replaced the homo economicus assumption of classical models and became one of the founding premises of behavioural economics.
Bounded rationality matters in economics because classical models predict behaviour that people do not exhibit. Consumers do not compare every product, investors do not process all available information, and firms do not identify the profit-maximising price. Replacing perfect rationality with bounded rationality produced models that predict real market behaviour more accurately, including under-saving, default effects and persistent price dispersion.
Homo Economicus vs the Boundedly Rational Agent
| Assumption | Classical economics (homo economicus) | Behavioural economics (bounded rationality) |
|---|---|---|
| Information | Complete and costless | Incomplete, and acquiring more has a real cost |
| Computation | Unlimited, any calculation is instant | Limited, complexity forces simplification |
| Preferences | Stable, consistent and known in advance | Partly constructed during the decision; sensitive to framing |
| Goal | Maximise utility | Reach an acceptable outcome at tolerable cost |
| Response to more options | Weakly better, more choice cannot hurt | Can be worse, excessive choice reduces decision quality |
| Effect of defaults | None, a rational agent overrides any default | Large, defaults substantially change outcomes |
What bounded rationality explains that classical models cannot
- Identical goods sell at different prices in the same market, because buyers stop searching once a price seems acceptable.
- Pension participation rises sharply under automatic enrolment, even though a rational agent would enrol regardless of the default.
- Consumers frequently buy less when offered more options, see the sourced finding below.
- Firms use cost-plus pricing and competitor matching rather than deriving price from demand elasticity.
- Investors hold under-diversified portfolios weighted toward familiar names.
Profit.co Brand Signal: The jam finding has a direct organisational analogue. A team facing forty possible initiatives does not evaluate forty and pick the best; it stalls, or it picks whichever arrived most recently. Narrowing the set is not a loss of ambition, it is what makes a decision possible at all. Profit.co’s OKR software enforces that narrowing structurally, by limiting how many objectives a team carries in a quarter and making the trade-offs visible when something new is proposed.
Satisficing: Why We Choose “Good Enough” Instead of Best
Satisficing is the decision strategy of searching through available options until one meets a predefined threshold of acceptability, then stopping. The word combines satisfy and suffice, and was coined by Herbert Simon. Unlike optimising, satisficing does not compare every alternative, it ends the search as soon as a good-enough option appears and clears the bar you set in advance.
Satisficing is the behavioural mechanism through which bounded rationality operates. Bounded rationality describes the constraint: limited information, limited time, limited cognitive capacity. Satisficing describes what people actually do in response: set a threshold, search until something clears it, and stop. This is why bounded rationality produces predictable behaviour rather than random error.
Classical economics assumes decision-makers optimise: they identify every option, evaluate each one against complete information, and select the single best. Simon’s objection, set out in the Quarterly Journal of Economics in 1955, was not that people are irrational but that optimisation is computationally impossible for real humans facing real problems in real time. What people do instead is intelligent and systematic, it is simply a different procedure.
Optimising vs Satisficing
| Dimension | Optimising (perfect rationality) | Satisficing (bounded rationality) |
|---|---|---|
| Goal | Find the single best available option | Find an option that is good enough |
| Information required | Complete, all options and all consequences | Partial, whatever is available within the search |
| When the search stops | When every alternative has been evaluated | When the first option clears the acceptance threshold |
| Cognitive cost | Very high, often prohibitive | Low and bounded by design |
| Typical outcome | Theoretically optimal, practically unreachable | Suboptimal but achievable and usually adequate |
| Failure mode | Analysis paralysis; the decision is never made | Settling too early; better options never examined |
How Satisficing Works: The 4 Steps
- Set a threshold: Decide, usually implicitly, what “acceptable” means. This is the aspiration level, the bar an option must clear.
- Search sequentially: Examine options one at a time rather than as a complete set. Order of arrival matters, which is why satisficing is sensitive to how choices are presented.
- Evaluate against the bar: Compare each option against the threshold, not against every other option. This is the step that makes satisficing computationally cheap.
- Stop on first success: Accept the first option that clears the threshold and end the search. Remaining alternatives are never examined.
A fifth step operates in the background: if the search runs long without success, the threshold falls. If acceptable options appear quickly, the threshold rises. Aspiration levels adjust to experience, which is why the same person satisfices differently in a strong job market than in a weak one.
When Satisficing Is the Better Strategy
Satisficing is not a failure state. It is the correct approach whenever the cost of continued search exceeds the value of a marginally better outcome.
| Satisfice when… | Optimise when… |
|---|---|
| The decision is reversible | The decision is difficult or impossible to reverse |
| Options are broadly similar in value | Differences between options are large and material |
| Search itself is expensive in time or money | Search cost is low relative to the stakes |
| The decision repeats often | The decision is rare or one-off |
| Delay carries a real cost | There is genuine time to evaluate properly |
| Information is unreliable anyway | Reliable comparative data is available |
Satisficers vs maximisers: the 20% salary finding
Psychologist Barry Schwartz, in The Paradox of Choice (2004), distinguished between people who habitually satisfice and people who habitually maximise. Maximisers search exhaustively and want the best possible outcome.
The explanation the authors offer is that exhaustive search fixes attention on one easily compared variable, salary, while the alternatives you rejected stay vivid. Satisficers decide faster, compare less, and are more content with what they chose.
Profit.co Brand Signal: Satisficing only works well when the threshold is set deliberately. In most teams it is not: each person invents their own private definition of “good enough”, and decisions diverge across the organisation without anyone noticing. This is what a measurable key result does structurally, it states the acceptance threshold in advance, in a number everyone can see, so the same bar applies to whoever happens to be deciding. Profit.co’s OKR software exists to make that threshold explicit and visible rather than assumed.
The Bounded Rationality Model of Decision Making
The bounded rationality model of decision making is a descriptive model that explains how people actually decide under real constraints. It assumes decision-makers hold incomplete information, face time limits, and have finite cognitive capacity, so they simplify the problem, examine a limited set of options, and accept the first adequate one rather than searching for the optimum.
The bounded rationality model differs from the rational model in three ways: it assumes incomplete rather than perfect information, sequential rather than simultaneous evaluation of options, and satisficing rather than optimising as the stopping rule. Herbert Simon formalised it in the Quarterly Journal of Economics in 1955. It is descriptive, it explains what people do, not what an ideal decision-maker would do.
Rational Model vs Bounded Rationality Model vs Intuitive Model
| Dimension | Rational model | Bounded rationality model | Intuitive model |
|---|---|---|---|
| Core assumption | The decision-maker knows all options and outcomes | Information, time and cognition are all limited | Judgement draws on pattern recognition and experience |
| How options are found | All alternatives identified up front | A limited set surfaces sequentially | The first workable option is recognised, not searched for |
| Evaluation method | Every option scored against every criterion | Each option checked against an acceptability threshold | Holistic and largely non-conscious |
| Stopping rule | Optimise, choose the highest-scoring option | Satisfice, choose the first acceptable option | Act when the situation feels familiar enough |
| Nature of the model | Prescriptive, what should happen | Descriptive, what does happen | Descriptive, what experts do under pressure |
| Best suited to | High-stakes decisions with good data and time | Most real organisational and personal decisions | Time-critical decisions in a familiar domain |
| Main weakness | Rarely achievable in practice | Can settle too early and miss better options | Fails badly outside the domain of experience |
The 6 Steps of the Bounded Rationality Decision-Making Model
- Recognise a simplified problem: The decision-maker registers a gap between the current and desired state, but perceives it through existing assumptions, so the framing is already narrowed.
- Set an acceptability threshold: Rather than defining the ideal outcome, the decision-maker forms a working sense of what would be acceptable. This aspiration level is shaped by past experience and peer comparison.
- Search a limited option set: Options are generated from memory, habit, and whatever is immediately visible. Alternatives that would require effort to discover are usually never considered.
- Evaluate sequentially: Each option is checked against the threshold as it appears, rather than ranked against all the others.
- Select the first adequate option: The first option clearing the threshold is chosen, and the search ends. Remaining alternatives go unexamined.
- Adjust the threshold: If no option clears the bar, the threshold drops until one does. If options clear it easily, the bar rises for next time.
The six steps in practice: annual budget approval
The model is easier to recognise in a process you have sat through. A department head is asked to submit next year’s budget. Step one: the problem arrives framed as “adjust last year’s number”, not “determine the optimal allocation of resources to this department”. Step two: an acceptable submission is one that will clear finance review without escalation, typically within a few percent of the prior year. Step three: the options considered are last year’s budget plus or minus a handful of line items already under discussion; a zero-based rebuild is never generated. Step four: each draft is tested against “will this get approved?” rather than against every other possible allocation. Step five: the first version that looks defensible is submitted. Step six: if finance pushes back, the threshold shifts and the number is revised. No one in this process behaves irrationally. The optimal budget was simply never in the option set.
What the model assumes
- Information is incomplete, and gathering more is itself costly.
- Cognitive capacity is finite, the number of variables held in mind at once is small.
- Time is constrained, and delay carries a cost of its own.
- Preferences are not fully formed in advance; they are partly constructed during the decision.
- The environment matters as much as the mind, how choices are structured changes what gets chosen.
Profit.co Brand Signal: The practical lever in this model is step two. Every other step follows from where the acceptability threshold sits, and in most organisations that threshold is never stated, it is inferred from precedent, which is how the budget example above produces the same answer every year. Writing the threshold down as a measurable key result changes what enters the option set at step three, because teams can test a candidate against a number rather than against what got approved last time. That is the specific job Profit.co’s OKR software does.
Bounded Rationality Examples: 15 Real Situations
Common bounded rationality examples include hiring the first candidate who meets the requirements rather than interviewing the whole market, buying the second car you test drive, renewing a supplier contract without retendering, and accepting the default option in a pension scheme. In each case the decision-maker stops searching once an option is good enough rather than establishing which option is best.
The pattern below is consistent. In every row, the fully rational choice is theoretically available and practically unreachable, so the decision-maker substitutes a simpler procedure that produces an adequate answer at a fraction of the cost.
Bounded Rationality in Everyday Life
| Situation | What a fully rational actor would do | What people actually do | Limit at work |
|---|---|---|---|
| Choosing what to wear | Evaluate every garment against weather, schedule and preference | Pick from the three or four items visible at the front | Time and attention |
| Ordering in a restaurant | Compare every dish on price, taste preference and nutrition | Order the usual, or the first item that appeals | Cognitive effort |
| Buying a car | Test drive every model in the segment across all dealers | Buy from among two or three cars actually driven | Search cost |
| Choosing a mortgage | Model every product against every future rate scenario | Take the deal the broker recommends | Information complexity |
| Pension contributions | Calculate the optimal rate given lifetime earnings and returns | Accept the default enrolment rate | Default bias and complexity |
Bounded Rationality at Work
| Situation | What a fully rational actor would do | What people actually do | Limit at work |
|---|---|---|---|
| Hiring | Assess every qualified candidate in the labour market | Hire the first applicant who clears the bar | Time pressure and search cost |
| Choosing a supplier | Retender against the full market each cycle | Renew with the incumbent unless something has gone wrong | Switching cost and inertia |
| Prioritising a backlog | Compute expected value for every item and rank them | Work on whatever is loudest or most recently raised | Attention and salience |
| Selecting software | Trial every product against a full requirements matrix | Shortlist three tools from a comparison site and pick one | Evaluation capacity |
| Approving a budget | Evaluate every possible allocation of the total | Adjust last year’s numbers incrementally | Complexity and precedent |
Bounded Rationality in Business, Economics and Policy
| Situation | What a fully rational actor would do | What people actually do | Limit at work |
|---|---|---|---|
| Pricing a product | Derive price from a full demand elasticity model | Apply a standard markup, or match the nearest competitor | Data availability |
| Investing savings | Optimise the portfolio across every available asset | Buy whatever the platform features on the front page | Choice overload |
| Choosing health insurance | Model every plan against projected medical needs | Re-enrol in last year’s plan | Plan complexity and inertia |
| Setting public policy | Model every intervention against every outcome | Adjust the existing policy at the margin | Political and analytical limits |
| Responding to a competitor | Game out every competitive move and counter-move | Match the visible move quickly | Time and uncertainty |
Why the Constraints Are Tightening in 2026
Read together, those two findings describe the exact conditions Simon identified, intensifying rather than easing. The people making the most consequential decisions report the highest stress, and the stated strategy of most organisations is to decide faster. Bounded rationality is not a historical curiosity in that environment, it is the operating condition.
Profit.co Brand Signal: There is a version of “fast and nimble” that simply means deciding with less information, and a version that means having decided in advance what matters so the daily calls get easier. The second requires that priorities are written down, measurable and visible to everyone who has to choose. Profit.co’s OKR and strategy execution software is built for that: when the objectives are explicit, most of the decisions that consume a manager’s attention answer themselves.
Bounded Rationality Psychology
The psychology of bounded rationality studies how cognitive limits lead people to make suboptimal decisions. Heuristics, emotions, and complex issues can all lead to suboptimal decisions.
Satisficing is the best-known model of bounded rationality. This model assumes that humans make “good enough” rather than optimal decisions due to limited information and cognitive capacity. This model is accurate in some situations.
Bounded rationality is an important heuristic psychological concept that helps explain why humans make suboptimal decisions. By understanding the factors that lead to suboptimal decisions, you can improve your decision-making process and avoid costly mistakes.
What Causes Bounded Rationality?
Several factors can cause bounded rationality, including information processing biases, heuristics, and mental shortcuts. These can all lead us to make suboptimal decisions that may not be in our best interests. Bounded rationality can have several consequences, both positive and negative. Sometimes, it may lead us to make suboptimal choices that negatively affect our health or well-being.
In other cases, it may instruct us to make more creative or innovative decisions that would not have been possible if we had followed a strictly rational approach. Bounded rationality is a critical concept when considering how humans make decisions. It can help us better understand why we sometimes make suboptimal choices and how we can avoid doing so in the future.
OKRs can help avoid decision dilemmas. The center of OKRs is focus and alignment. After you plan your objectives and tie them down to measurable key results, you spend a little bit of time prioritizing to ensure that your goals are in alignment with your managers. Then you get to work and let your OKRs guide you through the quarter. You can simply avoid doing many things by asking a simple question: is this in line with my OKRs or my team’s OKRs? To learn more about OKRs you can get started on Profit.co completely free today!
Bounded Rationality in Management, Public Administration and Policy
In management, bounded rationality explains why strategic decisions are made from a narrow option set and why firms imitate competitors rather than optimising independently. In public administration it explains incrementalism, the tendency to adjust existing policy at the margin rather than redesign it. Charles Lindblom described this in “The Science of Muddling Through,” Public Administration Review (1959).
How Bounded Rationality Shows Up by Domain
| Domain | How the limit appears | Practical consequence | What reduces it |
|---|---|---|---|
| Strategic management | Leaders consider a handful of options generated from experience and peer behaviour | Strategies converge across an industry; genuine alternatives go unexamined | Structured option generation and pre-mortems |
| Operations | Decisions default to established procedure under time pressure | Process drift accumulates unnoticed | Explicit decision rules and periodic review |
| Family business | Options are filtered through relationships, history and succession concerns | Emotionally salient choices displace commercially stronger ones | External advisers and formal governance |
| Public administration | Analytical capacity is far smaller than the problem’s complexity | Incremental adjustment rather than redesign | Pilot programmes and staged evaluation |
| Public policy design | Policies are built assuming citizens will read, compare and optimise | Low uptake of beneficial programmes | Simplified choices and sensible defaults |
| Project portfolios | Projects are assessed against recent memory rather than the full portfolio | Weak projects continue because stopping requires an active decision | Stage gates with explicit kill criteria |
Incrementalism: bounded rationality applied to policy
Charles Lindblom argued that policymakers do not select from a complete set of options because no institution has the analytical capacity to construct one. Instead they compare a small number of variations on the current position and choose a modest change. He described this as muddling through, and treated it as realistic rather than as failure: small adjustments are easier to reverse when they turn out badly, which makes incrementalism a rational response to uncertainty rather than merely a symptom of limited capacity.
Profit.co Brand Signal: The “what reduces it” column above has a common thread: every entry replaces implicit judgement with an explicit, visible rule. Stage gates with kill criteria, structured option generation, periodic review against a stated threshold, these are governance mechanisms, not intelligence upgrades. Profit.co’s strategy execution and project portfolio tools implement them directly, which is what turns the 66% who know something must change into the small share who actually manage it.
Beyond Bounded Rationality: Bounded Awareness, Willpower and Ethicality
Researchers have extended Simon’s idea into three further bounds: bounded awareness, bounded willpower and bounded ethicality. Together with bounded rationality, they describe four distinct ways real decision-making departs from the idealised model, and each one implies a different remedy.
The Four Bounds Compared
| Concept | What it describes | Everyday example | Associated with |
|---|---|---|---|
| Bounded rationality | Limited information, time and cognitive capacity | Hiring the first adequate candidate rather than the best available | Herbert A. Simon (1947, 1955) |
| Bounded awareness | Failing to notice information that is present and relevant | Focusing on the deal terms and missing that a key stakeholder was never consulted | Max Bazerman and Dolly Chugh |
| Bounded willpower | Acting against your own long-term interest despite knowing better | Deferring pension contributions you fully intend to make | Richard Thaler and Cass Sunstein, Nudge (2008) |
| Bounded ethicality | Behaving less ethically than your self-image implies, without noticing | Approving an aggressive forecast because everyone in the room already agreed | Max Bazerman and Ann Tenbrunsel, Blind Spots (2011) |
| Bounded self-interest | Caring about fairness and others in ways pure self-interest cannot explain | Rejecting a profitable offer perceived as unfair | Behavioural economics literature broadly |
Why the Distinction Matters
The practical value of separating these is diagnostic. If a decision went wrong because the option set was too narrow, the fix is a better search process. If it went wrong because relevant information was present but unnoticed, the fix is a structured checklist or an outside reviewer. If it went wrong because the long-term interest lost to the immediate one, the fix is commitment devices and defaults. And if it went wrong because no one in the room wanted to be the person who objected, the fix is a process that makes objecting cheap. Treating all four as “bad judgement” leads to the wrong remedy every time, usually a training course, when the problem was structural.
Profit.co Brand Signal: Bounded awareness and bounded ethicality are the two that structure helps most with, because both depend on whether anyone is obliged to look. A review cadence that asks the same questions on a fixed schedule surfaces information that ad hoc discussion misses, and it makes raising a concern a normal part of the process rather than an act of courage. Profit.co’s performance review and 1:1 tools are built to run that cadence consistently.
How to Reduce the Cost of Bounded Rationality
Bounded rationality cannot be eliminated, the limits are structural. The effective interventions work on the decision environment rather than on the decision-maker: narrow the option set deliberately, define in advance what an acceptable outcome is, set defaults that favour the better choice, and separate the decisions that deserve full analysis from the ones that do not.
What can be reduced is the cost. That distinction matters, because most advice in this area targets the wrong blade of Simon’s scissors, it asks people to think more carefully rather than changing the conditions they think under. For a broader set of structured approaches, see Profit.co’s guide to 10 business decision-making frameworks.
A note on heuristics, since the advice here is often stated backwards. Heuristics are not a remedy for bounded rationality; they are how bounded rationality expresses itself. That does not make them bad. Gerd Gigerenzer argued in Simple Heuristics That Make Us Smart (1999) that a well-matched heuristic is fast, cheap and frequently more accurate than a complex model, particularly under genuine uncertainty, where the complex model overfits. The distinction that matters is between heuristics you have chosen deliberately and tested, and heuristics you are running without knowing it. The first are tools. The second are blind spots.
6 Practical Interventions
| Intervention | What it does | When to use it |
|---|---|---|
| Define “good enough” in advance | Makes the acceptability threshold explicit instead of leaving each person to invent one | Any decision made repeatedly across a team |
| Cap the option set deliberately | Reduces choice overload while ensuring the cap is a conscious choice rather than an accident of what was visible | Vendor selection, hiring shortlists, roadmap prioritisation |
| Set the default to the better option | Uses inertia in your favour rather than against you | Enrolment, configuration, recurring approvals |
| Triage by reversibility | Reserves full analysis for decisions that are hard to undo and lets the rest be satisficed quickly | Any decision queue |
| Run a pre-mortem | Surfaces information that is available but unnoticed, before the decision is locked in | High-stakes or irreversible decisions |
| Codify and test your heuristics | Converts unexamined shortcuts into explicit rules you can measure and revise | Anywhere the same judgement is made weekly |
Profit.co Brand Signal: The first two interventions are what a goal framework does structurally. OKRs state the acceptability threshold as a measurable key result, and they give teams a fast test for whether an option belongs in the set at all: does this move a key result, or not? That single question removes a large share of the low-value decisions that consume attention, which is the cost reduction the section title refers to. Profit.co builds the software for it.
Bounded Rationality in Other Languages
Simon’s term travels across languages with the same meaning, a rationality that is real but limited. The equivalents below are the standard renderings used in academic and business writing in each language.
| Language | Term | Literal sense |
|---|---|---|
| Spanish | racionalidad limitada | limited rationality |
| Portuguese | racionalidade limitada | limited rationality |
| Indonesian | rasionalitas terbatas | limited rationality |
| German | beschränkte Rationalität | restricted rationality |
| Turkish | sınırlı rasyonellik | limited rationality |
| Vietnamese | duy lý có giới hạn | reasoning with limits |
| Dutch | begrensde rationaliteit | bounded rationality |
| French | rationalité limitée | limited rationality |
Final Thoughts
It’s important to note that bounded rationality is no excuse for bad decisions; instead, it recognizes that humans are limited in their ability to make perfectly rational decisions. By understanding the impact of bounded rationality, we can take steps to mitigate its impact and make better decisions overall.
OKRs can certainly help you prioritize and focus, but it is possible employees can make bad choices, and that includes managers. Senior leaders should not assume that capable individuals can absorb unlimited decision load. Capacity is finite regardless of ability. Book a free demo with our team to learn more about how OKR software can optimize your organization’s performance by improving your decision-making.
Clear goals don’t just simplify decisions, they make better ones possible. Profit.co creates that clarity
Frequently Asked Questions
Nobel laureate Herbert Simon first proposed the concept of bounded rationality in the 1950s. Simon argued that people cannot always make perfectly rational decisions because of their limited cognitive abilities. This prompted him to develop a more realistic human decision-making model known as bounded rationality.
The bounded rationality approach is helpful in several fields, such as economics, psychology, and artificial intelligence. In recent years, political decision-making has also been known to be influenced by bounded rationality.
There are several ways to combat bounded rationality. One common approach is to use heuristics or simple rules of thumb to make decisions. This can reduce the cognitive load associated with difficult decision-making. Another method is to seek out more information before making a decision. However, this can be difficult because people seek information to support their beliefs, a phenomenon known as confirmation bias. Finally, it is crucial to recognize the role of emotions in decision-making. Emotions can sometimes be confusing, but they can also provide valuable information about our preferences and values. If you learn to regulate your emotions effectively, you can use them to make better decisions.
Bounded rationality is the idea that human decisions are constrained by limited information, limited time, and limited cognitive capacity. Instead of optimizing, we satisfice, choosing an option that is good enough.
The concept was developed by Herbert A. Simon, who argued that real-world decision-making cannot match the assumptions of perfect rationality used in classical economics.
Choosing the first vendor that meets your key requirements, rather than evaluating every possible supplier in the market, is a classic bounded rational decision.
Not always. Mental shortcuts can save time and energy. Problems arise when they systematically ignore important data or amplify bias.
OKRs create focus. When goals and measurable outcomes are visible, employees can quickly test whether a task aligns with priorities, which simplifies decisions.
Satisficing is accepting the first option that clears a predefined threshold of acceptability, rather than comparing every alternative. The word combines satisfy and suffice, and was coined by Herbert Simon.
Optimising evaluates every alternative to find the single best one and requires complete information. Satisficing checks options against a threshold and stops at the first that clears it.
The rational model is prescriptive, it describes what an ideal decision-maker with complete information would do. The bounded rationality model is descriptive: it predicts what real decision-makers do, which is to narrow the problem, examine a few options in the order they arrive, and stop at the first adequate one. The rational model tells you what to aim for; the bounded model tells you what will actually happen.
Simon introduced the argument in Administrative Behavior (1947), formalised it in the Quarterly Journal of Economics in 1955, and named it in Models of Man (1957). He received the Nobel Memorial Prize in Economic Sciences in 1978 for his research into organisational decision-making.
Several persistent market behaviours. Identical goods sell at different prices because buyers stop searching once a price seems acceptable. Pension enrolment rises sharply when participation is the default. Consumers sometimes buy less when offered more choice, Iyengar and Lepper found roughly a tenfold difference in purchase rate between a six-option and a twenty-four-option display. Under perfect rationality, none of these should occur.
No. A boundedly rational decision-maker reasons soundly from the information available. The limitation is in the inputs and the processing capacity, not in the logic. Irrationality means choosing against your own stated goals, which is a different failure.
Renewing a supplier contract without retendering is a common example. The rational procedure would be to retest the full market each cycle. In practice the incumbent is retained unless something has gone visibly wrong, because the cost of a full retender exceeds the expected gain.
Three structural limits: incomplete information, finite time, and finite cognitive capacity. These are not personal failings, they apply to every decision-maker, including well-resourced organisations and artificial systems.
Policies designed on the assumption that citizens will read, compare and optimise tend to see low uptake. Bounded rationality explains why simplified choices and well-set defaults outperform complex programmes, and why administrators adjust existing policy incrementally rather than redesigning it, the pattern Charles Lindblom called muddling through.
